Veritas D1.5.1

Accessible and Assistive ICT
VERITAS
Virtual and Augmented Environments and Realistic User
Interactions To achieve Embedded Accessibility DesignS
247765
Abstract Behavioural and Psychological

User models definition
Deliverable No. D1.5.1
SubProject No. SP1 SubProject Title User modelling

Workpackage W1.5 Workpackage Behavioural and
No. Title psychological models
Activity No. A1.5.1 Activity Title Existing Models
A1.5.2 Existing rules from SoA and
Standards
Authors Ana Navarro, Juan Carlos Naranjo, Hctor Galn
(ITACA), Vitaliy Kolodyazhniy, Marcel Delahaye
(COAT), Tamara Aguilar, Roco Zaragoza (AIJU)
Elena Tamburini (I+), Eleni Chalkia (CERTH-HIT),
Giuseppe Varalda (CRF)
Status F (Final)
Dissemination level Pu (Public)

File Name: VERITAS_D1.5.1_DRAFT_v1.doc
Project start date and 01 January 2010, 48 Months

duration
VERITAS_D1.5.1 PU Grant Agreement # 247765
Version History table

Version Dates and comments
no.
st
1 1 December 2010. Draft version. ITACA, COAT, I+, AIJU. Chapter 1, 2 and 3.
2 15th December 2010. Chapter 3 final. (ITACA)
3 21st December 2010. Chapter 4 first contributions ITACA, I+
4 5th of January 2011. Chapter 5 contributions ITACA,I+. (Tables and ontology

approach)
th
5 8 January: Chapter 4 final contributions all partners involved (ITACA, AIJU, I+,
CERTH, CRF), and conclusions draft.
th
6 10 January: AIJU final contributions to Chapter 3. Closing Chapter 3
7 13rth January 2011. Final contributions to the Abstract User Model tables by
CERTH and CRF (Chapter 5). Second version Chapter 6 (conclusions).
Executive Summary included.
8 15th January 2011. Review of Conclusions (Chapter 6) by all partners involved.
9 Final formatting of the deliverable.(ITACA)
st
10 31 January 2011. Changes following Peer reviewers comments. (ITACA)
December 2010 iii ITACA

Table of Contents
Version History table ....................................................................................... 3
Table of Contents ............................................................................................. 4
List of tables ..................................................................................................... 6
List of Figures................................................................................................... 7
Abbreviations list ........................................................................................... 10
Executive Summary ......................................................................................... 1
1 Introduction ............................................................................................... 3
2 State of Art in Behavioural and psychological computational models 5
2.1 B&P modeling current computational approaches .......................... 5
2.1.1 Use of the models for elderly & disabled user groups ........................................... 8
2.1.2 Description of existing computational models for each targeted domain ............ 10
2.2 Analysis of the P&B computational models ................................... 13
3 Cognitive and behavioral integrated architectures .............................. 25
3.1 Human behavioral and psychological theories .............................. 25
3.1.1 THE BEHAVIOURAL APPROACH ...................................................................... 25
3.1.2 THE COGNITIVE APPROACH............................................................................ 29
3.2 Cognitive architectures: ACT-R..................................................... 36
3.2.1 ACT-R structure ................................................................................................... 37
3.2.2 Human knowledge in ACT-R ............................................................................... 40
3.2.3 ACT-R modeling methodology: model variation, model space and parameters for
different domains and users ............................................................................................... 41
3.2.4 Functionality of the overall architecture: Integrated Driving Modeling Example 42
3.3 Cognitive and behavioural moderators: Overlays ....................... 45
3.4 Cognitive Architecture & Cognitive and Behavioural Moderators:
suitable approach for VERITAS? .................................................................. 48
3.4.1 Advantages .......................................................................................................... 49
3.4.2 Disadvantages ..................................................................................................... 51
3.4.3 Analysis and recommendations of psychological and behavioural modeling for
VERITAS ............................................................................................................................ 52
4 Psychological facets: stress, fatigue, emotions and motivation as
cognitive and behavioural moderators......................................................... 53
4.1 Stress as cognitive and behavioural moderator ............................ 54
4.1.1 Definitions of Stress ............................................................................................. 54
4.1.2 Types of stress..................................................................................................... 55
4.1.3 Causes of stress: stressors or stress stimuli ....................................................... 57
4.1.4 Physiological, cognitive and behavioural stress responses ................................ 58
4.1.5 Relationship of Stress to Task Performance ....................................................... 63
4.1.6 Measurement of human stress responses .......................................................... 64
4.1.7 Stress as cognition moderator: analysis of psychological studies ....................... 73
4.1.8 Stress and disabled ............................................................................................. 89
4.1.9 Stress and elderly ................................................................................................ 93
4.1.10 Stress models approaches .................................................................................. 94
4.2 Fatigue as cognitive and behavioural moderator .......................... 97
4.2.1 Definitions of Fatigue ........................................................................................... 97
4.2.2 Types of fatigue ................................................................................................... 98
4.2.3 Causes of fatigue: fatigue stimuli ......................................................................... 99
4.2.4 Physiological, cognitive and behavioural fatigue responses ............................. 100
December 2010 iv ITACA

4.2.5 Assessment of fatigue and human fatigue effect............................................... 103

4.2.6 Relationship of Fatigue to Task Performance ................................................... 105
4.2.7 Fatigue as cognition moderator: Analysis of psychological studies .................. 107
4.2.8 Fatigue and disabled ......................................................................................... 112
4.2.9 Sleepiness and Fatigue in Elderly people ......................................................... 113
4.2.10 Fatigue modeling approaches ........................................................................... 115
4.3 Emotions as cognitive and behavioral moderator ....................... 122
4.3.1 Definition ............................................................................................................ 122
4.3.2 Theories of emotions ......................................................................................... 122
4.3.3 Approaches to emotions .................................................................................... 125
4.3.4 Physiological, cognitive and behavioural responses to emotions ..................... 136
4.3.5 Dynamics of emotional responses ..................................................................... 142
4.3.6 Bodily expressions of emotions ......................................................................... 145
4.3.7 Measurement methods for emotions ................................................................. 148
4.3.8 Six basic emotions as cognitive and behavioural moderators ........................... 163
4.3.9 Emotions and elderly ......................................................................................... 168
4.3.10 Emotions and disabled ...................................................................................... 169
4.3.11 Emotions modelling approaches........................................................................ 170
4.4 Motivation as cognitive and behavioral moderator ...................... 178
4.4.1 Definitions of motivation ..................................................................................... 178
4.4.2 Types of motivation ............................................................................................ 178
4.4.3 Motivation theories and theoretical models ....................................................... 180
4.4.4 Physiological, cognitive and behavioural responses to motivation.................... 193
4.4.5 Measurement techniques for motivation ............................................................ 194
4.4.6 Motivation as cognitive moderator ..................................................................... 196
4.4.7 Motivation and elderly & disabled ...................................................................... 196
5 Psychological and Behavioural VERITAS Abstract User models ..... 201
5.1 Psychological and Behavioural Abstract User Models tables ..... 202
5.2 Towards an ontologies description.............................................. 263
5.2.1 Introduction and methodology ........................................................................... 263
5.2.2 Formal and explicit description approach: ......................................................... 264
6 Conclusions and next steps................................................................. 269
7 References ............................................................................................. 273
ANNEX A: ACT-R Parameters ..................................................................... 308
December 2010 v ITACA

List of tables
Table 1: Behavioural & psychological computational modeling approaches
analysis ...................................................................................................... 17
Table 2: Behavioural & psychological computational modeling approaches
analysis (cont.)........................................................................................... 23
Table 3: Difference between Classical Conditioning and Operant Conditioning28
Table 4 Distribution of stressors ....................................................................... 58
Table 5 Measurement of stress response ........................................................ 66
Table 6 Cognitive-affective scales .................................................................... 67
Table 7 Psychological diagnosis of the stress response .................................. 71
Table 8 Common causes of fatigue ................................................................ 100
Table 9 Assessment of fatigue and human fatigue effect ............................... 105
Table 10: Emotions illustration and description. ............................................. 128
Table 11: Various theories form the history of emotions research.................. 131
Table 12: A possible mapping between the facial actions and their meaning for
emotional state from by Smith and Scott. Smith and Scott (1997) ........... 147
Table 13: Physiological variables and behavioral indicators (Backs and
Boucsein 2000) ........................................................................................ 162
Table 14: Happyness ..................................................................................... 164
Table 15: Anger .............................................................................................. 164
Table 16: Fear ................................................................................................ 165
Table 17: Sadness ......................................................................................... 166
Table 18: Surprise .......................................................................................... 167
Table 19: Disgust ........................................................................................... 168
Table 20: Intrinsic and extrinsic motivation for working purposes................... 180
Table 21 Psychological and Behavioural Abstract User Models..................... 218
Table 22: Metrics, computational models and measurement techniques for
stress: Abstratct User Model Cont. .......................................................... 245
fatigue: Abstract User Model Cont. .......................................................... 246
emotions: Abstratct User Model Cont. ..................................................... 254
motivation: Abstract User Model Cont...................................................... 262
Table 26: List of available parameters in the ACT-R framework .................... 312
December 2010 vi ITACA

List of Figures
Figure 1: Information processing model. Mahoney, (1974) .............................. 31
Figure 2:Stages of the information processing models (Norman 1981)............ 33
Figure 3: How the human mind functions (Rey Ardid) ...................................... 34
Figure 4: How the human mind functions (Rey Ardid) ...................................... 35
Figure 5: ACT-R 6.0. Each ACT-R module is associated with a function ......... 38
Figure 6: ACT-R 6.0. Each ACT-R module is associated with a specific brain
region ......................................................................................................... 40
Figure 7: Integrated driver model ..................................................................... 44
Figure 8: Decision Making under stress ........................................................... 47
Figure 9: Stimuli / Response ............................................................................ 54
Figure 10: Stress response (G. S. Everly & J.M. Lating, 2002) ........................ 59
Figure 11: Epiphenomenological model of the stress response ....................... 62
Figure 12: The human function curve. .............................................................. 63
Figure 13: Coping / Pressure relation. (Haarmann, accessed in 2010) ............ 64
Figure 14: Value ranges for urinary epinephrine and norepinephrine............... 68
Figure 15: Arousal / Reaction time response.................................................... 73
Figure 16: Memory tests, Luethi, Meier and Sandi 2009 .................................. 76
Figure 17: Stress / Reaction time relation. (M. A. Ellenbogen et al., 2002) ...... 79
Figure 18: Performance comparison stressed and non-stressed (Morelli and
Burton, 2009) ............................................................................................. 81
Figure 19: Mean score stressed and non-stressed stressed (Morelli and Burton,
2009).......................................................................................................... 81
Figure 20: Inverted U model of stress (Hanckock, 1986c)................................ 84
Figure 21: Avoidance conditioning ................................................................... 85
Figure 22: Pitch frequency distribution (Steeneken and Hansen, 2002)........... 87
Figure 24: Speech production process(Steeneken and Hansen, 2002) ........... 89
Figure 23: Stressor taxonomy (Steeneken and Hansen, 2002) ........................ 88
Figure 25: Correlation matrix constructed from 31 variables (stressors) .......... 92
Figure 26: Effect of mental fatigue on perception of effort during high-intensity
cycling exercise: # is the significant main effect of time (P=0.05). ............ 99
Figure 27: Effects of mental fatigue on physiological responses during high-
intensity cycling exercise ......................................................................... 102
Figure 28: Decline of Performance with Total Sleep Deprivation, Sleep &
Performance Model vs Angus & Heslegrave (1985) ................................ 106
Figure 29: Human performance on the psychomotor vigilance test across 88 h
of total sleep deprivation (TSD), averaged over each day for the protocol (n
= 13). ....................................................................................................... 107
Figure 30: Repeated nights of sleep loss have cumulative cognitive impairment.
NOTE: B, baseline day. ........................................................................... 110
Figure 31: Changes across Time for Mean SEM (left, Figure a) Visual
Perceptual (CVPT) Response Omissions and (right, Figure b) Complex
December 2010 vii ITACA

Motor (ARPTT) Azimuth Deviations. Adapted from Russo, Kendall, Johnson

et al 2005. ................................................................................................ 111
Figure 32: The two process model of Borbely et al. (1982): Error! Marcador no
definido.
Figure 33: James and Lange theory illustration. ............................................. 124
Figure 34: Cannon and Bard theory illustration. ............................................. 124
Figure 35: Schacter-Singer theory illustration................................................. 125
Figure 36: Emotions mapping to arousal, valence, and stance dimensions
[A;V;]. ....................................................................................................... 132
Figure 37: Plutchik circumplex model. (Plutchik, 1980) .................................. 133
Figure 38: Combinations of primary emotions in dyads. (Plutchik, 1980) ....... 134
Figure 39: Dimensions of emotions according to V.J. Wukmir (1967) ............ 135
Figure 40: Emotions position in the negative-positive scale. .......................... 136
Figure 41: Simplified representation of the nervous system and the
physiological responses under its control. (Andreassi 2000) ................... 137
Figure 42: Map of the physiological components of emotions. ....................... 139
Figure 43: Emotional brain or limbic systems criterion .................................. 141
Figure 44: Various facial expressions created by emotional states like affection,
anger, anxiety, despair, rage, disappointment, fear, happiness and love. 146
Figure 45: Emotion and facial actions. Russell (2003) ................................... 148
Figure 46: Different gestures that provoke emotional statements. ................. 148
Figure 47: Implementation of PANAS methodology, from Crawford and Henry.
................................................................................................................. 153
Figure 48: SAM model illustration. (Lang, 1980) ............................................ 155
Figure 49: ProEmo animations. (Desmet, 2002) ............................................ 155
Figure 50: Overview of warmth monitor methodology. ................................... 156
Figure 51: Muscles for facial expressions. (Ekman, FACS)............................ 157
Figure 52: Velasquez baseline framework. (Velasquez, 1998) ...................... 171
Figure 53: Mapping of the OCC Model emotions and mood octants. ............. 173
Figure 54: Representation of emotions. (Cochran, Lee and Chown, 2006).... 174
Figure 55: Class diagram of the designed software and calculated fear emotion
intensity. (Pour Mohammad Bagher 2008) .............................................. 177
Figure 56: Motivational reaction to external and internal stimuli ..................... 181
Figure 57: Classification of expectancies proposed by Bandura (1977) ......... 185
Figure 58: Scheme of explanatory parameters according to the dynamic action
theory ....................................................................................................... 186
Figure 59: General scheme of motivaton following the reactive theories
approach .................................................................................................. 187
Figure 60: Marlow's hierarchy of needs .......................................................... 189
Figure 61: Scheme of the Action Control Theory (Kuhl, 1985) ....................... 191
Figure 62: Scheme of Rubicon theory of action phases (Heckhausen, 1987) 191
Figure 63: Relation of psychological processes with the world (Kuhl, 1986) .. 192
Figure 64: Ontology methodology .................................................................. 263
Figure 65: PsychologycalState attributes ....................................................... 264
December 2010 viii ITACA

Figure 66: Cause hierarchy ............................................................................ 265

Figure 67: ICD hierarchy ................................................................................ 265
Figure 68: ICF hierarchy ................................................................................. 265
Figure 69: CognitiveFunction hierarchy .......................................................... 266
Figure 70: Types of Parameter ....................................................................... 267
Figure 71: Types of Memory........................................................................... 268
December 2010 ix ITACA

Abbreviations list
Abbreviation Explanation
ACT-R Adaptive Control of Thought - Rational
ANN Artificial Neural Networks
ANS Autonomic Nervous System
AUM Abstract User Model
B&P Behavioural & Psychological
CFS Chronic Fatigue Syndrome
CS Conditioned Stimuli
CVR Cardiovascular Response
D Deliverable
DBP Diastolic Blood Pressure
DoW Description of Work
DSSQ Dundee Stress State Questionnaire
ECG Electroencephalography
EDA Electrodermal activity
EDR Electro dermal Reaction
EMG Electromyogram
EMS Elderly Motivation Scale
EPIC Executive-Process/Interactive Control
FACS Facial Action Coding System
HR Heart Rate
ICD International Classification of Diseases
ICF International Classification of Functioning, Disability and
Health
ICT Information and Communications Technology
LTM Long Term Memory
MRI Magnetic Resonance Image
MS Multiple Sclerosis
PD Parkinson's Disease
PF Personality Factor
December 2010 x ITACA

Abbreviation Explanation
PMF Performance Moderator Functions
POMS Profile of Mood States
PUM Programmable User Model
PVT Psychomotor Vigilance Test
SAB Scientific Advisory Board
SAM Self Assessment Manikin
SBP Systolic Blood Pressure
SSS Subjective Stress Scale
STM Short Term Memory
TSD Total Sleep Deprivation
TSS Total Stress Score
TSST Trier Social Stress Test
VeSR Verbal Self Report
ViSR Visual Self Report
VR Virtual Reality
VUMS Virtual User Modelling and Simulation (Cluster)
WMS Work Motivation Scale
December 2010 xi ITACA

December 2010 xii ITACA

Executive Summary
This deliverable entitled D1.5.1 Abstract behavioural and psychological user models
definition aims at giving an overview of respective existing models within subproject
User modelling of the VERITAS Project Virtual and Augmented Environments and
Realistic User Interactions To achieve Embedded Accessibility DesignS - Grant
Agreement no 247765.
Chapter 1 presents VERITAS motivation and needs for modelling behavioural and
psychological aspects of people with disabilities and elderly. It also gives an overview
of the modelling approach necessary for VERITAS.
In chapter 2, the current state of art of computational approaches for behavioural and
psychological models has been described. It provides a deep review of the current use
of the different models, focusing on how they are applied for VERITAS users (elderly
and disabled) and domains (automotive, infotainment, smart living places, workplace
and healthcare). It finally provides a complete analysis of the encountered models,
extracting the primary strengths and weakness concerning VERITAS. This state of art
has been extracted from the survey of behavioural and psychological models carried
out in A1.1.2 Benchmarking of existing models, standards and problems. This
analysis serves as basis for a first approximation to the B&P models that can be used
in VERITAS and shows that, up to date, cognitive architectures such as ACT-R with
cognitive and behavioural moderators as overlays of the architecture are the most
promising approaches for modelling cognitive processes and psychological-affective
aspects.
Chapter 3 describes in depth what cognitive and behavioral integrated architectures

are. It offers an overview of the behavioral and psychological theories underneath, a
description and analysis of the most used cognitive architecture: ACT-R, and how
psychological states can be overlaid into cognitive architectures. It provides a final
analysis of the pros and cons of this approach and the recommendations for VERITAS.
Chapter 4 describes the psychological facets selected in VERITAS as cognitive and

behavioural moderators. Stress, fatigue, motivation, and emotions are proposed as
necessary attributes of the behavioural and psychological user models. A first analysis
of the existent models and psychological studies of stress, fatigue, emotions and
motivation is included. This analysis focuses on the theoretical models of the facets,
which provides a deep understanding of the psychological dimensions of each state. It
also provides a second analysis of the existing computational models for the
psychological states focusing on VERITAS users, domains and measurement
methodologies.
Chapter 5 includes the Abstract User Model of the psychological facets selected:
stress, motivation, fatigue and emotions. The Abstract User Model is an explicit
representation of the multiple dimensions of the psychological facets. It compiles all the
information of the previous chapter, addressing for each psychological state, the
category, description, types, ICD classification, causes, age-relation, ICF functional
limitations and the physiological, cognitive and behavioural responses. To do so we
December 2010 1 ITACA

have adopted a common semantic and structure, that is also consistent with the
physical and cognitive Abstract User Models.
This chapter also includes a first approach towards a formal and explicit representation
for describing the Abstract User Model in an ontological way.
The last chapter finalizes with the conclusions of the here-presented research and the
steps that will follow in order to develop the Psychological and Behavioural User
Models.

1 Introduction
A common concern amongst virtual user models developers is to increase the realism
of models regarding behaviour and cognition. This is not an ephemeral fashion, but a
serious objective that directly affects the bottom line of commercial concerns and
mission achievement in non-profit organizations: products and applications with a
better cognitive fit and that are more emotively natural tend to sell better.
The central problem for user acceptance of human current models is missing diversity
and the low psychological nature likeness of the models, so for designing and
developing products with high user acceptance, not only functional limitations should
be considered, but also behavioural and cognitive aspects should be taken into
account.
VERITAS aims to develop, validate and assess tools for built-in accessibility support at
all stages of ICT and non-ICT product development. The goal is to introduce
simulation-based and virtual reality testing at all stages of the design process in order
to ensure that future products and services can be universally used, including people
with disabilities and elders.
For creating VR tools with high acceptability amongst users, behavioural and
psychological models should be integrated in the design process. Nevertheless an
integrated design theory is still missing. Different theories of human psychological
modelling and interaction should be therefore integrated and validated within VERITAS,
focusing on those models that can represent the psychological and behavioural facets
of elderly and disabled people.
An initial analysis of current theoretical and computational psychological models

approaches shows that there is a lack of consensus in psychological definitions and
models. In fact, the problem is less that there arent many models, so much as the fact
that the ones that exist are none integrated. The situation worsens when looking for
specific models for elderly and disabled, since literature shows little references about
them. This reveals both the need of developing a psychological user model that
integrates and unifies the current knowledge and of exploring and developing models
for elderly and disabled.
Elderly and people with disability or functional limitations face multiple life and body
changes that lead them to experiment different psychological sates during their life:
depression, fear, physical and emotional fatigue, distress, lack of motivation, amongst
many other. These psychological states can have a great impact on the quality of life
and affect the overall performance of these groups when performing a task.
Cognitive theories (Ellis, 1962; Beck, 1964) postulate that the way people feel is
associated with the way they interpret and think about a situation and that emotions
and behavior of people are influenced by their perception of events. From an
information processing perspective, the human mind thinks and feels as follows: a real
or virtual stimulus is received from the environment (situation, physiological state or
mental state - fantasy, memory, recollection, dream, etc.). The human mind processes
the information and a response (emotional, cognitive behavioural or physiological) is

generated. However, this response is never automatic; it is not directly conditioned by

the stimulus, but also takes into account the emotional processes, the intentions,
desires and motivations of each individual. Between environment and response there is
a person that acts like a filter. Understanding and categorizing the different
psychological concepts that take part on how a human being thinks and acts is
essential to construct the psychological models.
In VERITAS, we should focus on how different psychological facets affect VERITAS

users, elderly and disabled groups. Stress, fatigue, emotions and (de)motivation are
amongst the most relevant P&B states that influence elderly and disabled, so within
VERITAS we will pay special attention to model these psychological facets, trying to
find parameters, qualitative and quantitative values and rules of behaviour.
An Abstract User Model is a descriptive representation of the different dimensions if a

human being. A psychological and behavioural abstract user model of each state
should therefore categorize the representative features of the psychological facets of
an individual: definitions, types, stimulus triggering the psychological states, the human
responses (physiological, cognitive and behavioural) that characterize the
psychological states and the functional limitations that may generate, including relevant
parameters, measurements and metrics found in literature and computational
modelling approaches.
A formal and explicit representation of any knowledge can be of great help when
sharing understanding of the domain among experts. A normalized description is useful
since it is easy to manipulate and transformed by computers. An ontology is a formal
representation of knowledge as a set of concepts within a domain, and the
relationships between those concepts. Using ontologies for describing the Abstract
User Models will enable the reuse of psychological knowledge and to make theoretical
and computational assumptions explicit. In addition, it will allow expanding the
psychological data in a future.
The dimensions of the Psychological and Behavioural Abstract user models described
in the ontology will be the basis for the parameterized Psychological and Behavioural
User Models. The integration of these models with the physical and cognitive models,
will give as a result a multidimensional Virtual User Model that includes physical,
cognitive and psychological aspects.

2 State of Art in Behavioural and psychological

computational models 1
In behavioural science, system theory and dynamic systems modelling, a behavioural
model reproduces the required behaviour of the original analyzed system, such as
there is a one-to-one correspondence between the behaviour of the original system
and the simulated system. That namely implies that the model uniquely predicts future
system states from past systems states
(http://en.wikipedia.org/wiki/Behavioral_modeling).
According to Dinerstein et al. (2004), a behavioural model is an executable model

defining how the character should react to stimuli from its environment. Alternatively, a
cognitive model is an executable model of the characters thought process. A
behavioural model, as opposed to a cognitive one, is reactive (i.e., seeks to fulfill
immediate goals), whereas a cognitive model seeks to accomplish long-term goals
through planning: a search for what actions should be performed in what order to reach
a goal state. Thus a cognitive model is generally considered more powerful than a
behavioural one, but can require significantly more processing power.
On the other hand, a psychological model is considered to be a synonym of a cognitive

model (http://en.wikipedia.org/wiki/Psychological_model). Alternatively, psychological
models can simulate, e.g., personalities, moods or relationships (Mac Namee &
Cunningham (2003a, 2003b), e.g. the Eysenck model of personality represented on
two orthogonal axes of introversion-extroversion and neuroticism-stability.
Numerous authors (Silverman, 2001; Libin, 2001; Funge, 1998; Rickel & Johnson,
1998; Tambe et al., 1995) summarize the different social and psychological theories on
how to create/simulate human behaviour in a realistic Virtual Reality (VR) scenario. In
order to achieve VR tools with a high user acceptance, behavioural and psychological
models should be integrated in the design process. Nevertheless an integrated design
theory is still missing. VERITAS is aiming to validate and integrate different theories of
human modelling and interaction, integrating physical, cognitive and behavioural and
psychological models.
2.1 B&P modeling current computational approaches

According to (Libin, 2001), the central problem for user acceptance of human models is
missing diversity and the low psychological nature likeness of the models. One
solution for this could be (according to Silverman) an enhanced individual profile.
Each profile should be created considering emotional expression and specific
interaction (Silverman, 2001). For example in cognitive learning experiments emotional
words are remembered much better than neutral world (Kensinger & Corkin, 2003 This
state of art has been extracted from the survey of behavioural and psychological
models carried out in A1.1.2 Benchmarking of existing models, standards and
problems.). So far it is unclear, how emotional expression in the VR interacts with
1
This state of art has been extracted from the survey of behavioural and psychological models carried out
in A1.1.2 Benchmarking of existing models, standards and problems.

learning capacity and user acceptance. No guidelines are defined which describe the
appropriate level of emotional content. Definitely the implementation of emotions in VR
would help to boost the presence, i.e. the sensation that they are actually inside the
computer-generated environment, interacting with virtual objects.
Cognitive architectures such as ACT-R (Anderson & Lebiere, 1998), EPIC (Kieras &
Meyer, 1997), and SOAR (Laird et al., 1987) have been employed extensively in
psychology and the cognitive sciences to model human behaviour (Salvucci et al.,
2001). A cognitive architecture is a blueprint for intelligent agents
(http://en.wikipedia.org/wiki/Cognitive_architecture). It proposes (artificial)
computational processes that act like certain cognitive systems, most often, like a
person, or acts intelligent under some definition. Cognitive architectures form a subset
of general agent architectures. The term 'architecture' implies an approach that
attempts to model not only behaviour, but also structural properties of the modelled
system. These need not be physical properties: they can be properties of virtual
machines implemented in physical machines (e.g. brains or computers).
The GOMS family models (http://en.wikipedia.org/wiki/GOMS) are used for modelling

of human-computer interaction (John & Kieras, 1996) with skilled users.
Models developed using cognitive architectures like SOAR consider the uncertainty of
human behaviour in detail but have not been widely adopted. Usability issues for
cognitive architectures are also supported by the X-PRT system (Tollinger et al., 2005)
for the CORE architecture. Salvucci & Lee (2003) have developed the ACT-Simple
model by translating basic GOMS operations into ACT-R production rules. Young et al.
(1989) propose a Programmable User Model (PUM) based primarily on the SOAR
architecture and incorporating ideas from some other architectures as ACT-R.
Blandford et al. (2004) also implemented a PUM based on SOAR.
Galata et al. (1999) presented an approach for automatically acquiring stochastic

models of the high-level structure of an activity without the assumption of any prior
knowledge. The process involves temporal segmentation into plausible atomic
behaviour components and the use of variable length Markov models for the efficient
representation of behaviours.
Cognitive models, such as GOMS, can model human interaction with almost any
devicesince they are concerned with the cognitive issues involved. In contrast, the
Markov model approach (Thimbleby et al., 2001) starts with the specifications of
artefacts, and we are limited in the sorts of interaction that can be modeled. It cannot
handle continuous interaction, but can handle all discrete systems, whether they are
multimodal or involve parallel input and output. Such devices are ubiquitous and
pervasive in modern culture. Most other approaches only study individual traces of user
behaviour. The Markov approach is statistical, and tells the designer about all
possible behaviours of the user. The results of the analyses are expressed in terms like
on average or can be plotted graphically. Whether the distributions describe individual
users over a period of time or a population of users is determined by how the
probabilities and models are set up. One of the most important advantages of the
Markov approach lies in direct modelling of user errors. Most other approaches like
GOMS ignore user error or find it difficult to handle. Error is central to a Markov

approach. Users can make choices, slips, or errors, with varying probabilities, which
Markov models use directly.
Biswas & Robinson (2008a) developed a simulator to help with the evaluation of
assistive interfaces using the CPM-GOMS (John & Kieras, 1996) model to simulate the
optimal behaviour and a new model based on Markov processes for sub-optimal
behaviour.
Dinerstein et al. (2004), Charles & McGlinchey (2004) propose an approach to creating
behavioural and cognitive models for virtual character animation based on artificial
neural networks (ANN). In this approach, an ANN can be used to learn (i.e.,
memorize) the decisions made in a particular situation to achieve a certain goal.
Thereafter, it is possible to quickly recall these decisions by executing the trained ANN.
Training is done offline and then the trained network is used online. So, for instance,
intelligent virtual characters can act in real time using very few computational
resources. It is also possible to use adaptive (online) learning or evolving approaches
for the characters to adapt themselves to changing environment or enable them to start
acting when little prior information is available and refine their behaviour and/or skills as
more and more information is incoming.
In general, the advantage of the ANN-approach is the inherent learning capabilities that
are beneficial for modelling when little or no explicit prior knowledge is available. The
disadvantages of ANN are the need for training prior to use, the need for a
representative training set for satisfactory performance of the model, and difficulty in
interpreting the trained model because the ANNs are essentially black box models.
Fuzzy logic (Chen, 1996) can be used to control the behaviour of animated characters
in computer games (El-Nasr et al., 1999; Karunaratne & Yan, 2001; van Waveren,
2001). This logic is a superset of conventional (Boolean) logic. It was extended to
handle the concept of partial truth, also using values between "completely true" and
"completely false" and emerged as a consequence of the 1965 proposal of fuzzy set
theory (Zadeh, 1965). Prof. Lotfi Zadeh of UC/Berkeley introduced fuzzy sets in the
1960's as a means to model the uncertainty of natural language. El-Nasr et al. (1999)
used fuzzy rules to relate sets of emotions to corresponding behaviours. Fuzzy logic
can be used by intelligent agents or bots in 3D action games (van Waveren, 2001) to
express how much they want to have, or do certain things.
Mac Namee & Cunningham (2003a, 2003b) proposed the -SIC system, which - uses
a number of quantitative psychological models to simulate characters personalities,
moods and relationships. The values of these models are used as inputs to an ANN
which drives characters social behaviour and determines which interactions the
character should engage in at a particular time. The ANN was trained with a data set
generated from a small set of hand crafted interactions. A psychological model used in
-SIC simulates a character's mood as it changes over time through interactions with
other characters or players. An agent's mood is measured according to valence and
arousal (Lang, 1995) where valence refers to whether the mood is positive or negative
and arousal refers to the intensity of the mood.

Barry G. Silverman proposes human behavior models for Agents in Simulators

based on PMFs (Performance Moderator Functions) [G. Barry G]. This work
explores emergent macro-behavior due to micro-decisions of bounded-rational agents
and with developing a framework that permits one to examine the impacts of biology,
stress, personality, culture, emotion, social relations, and decision making upon human
coping behavior. Barry G. Silverman postulates a High Level of Unified Architecture of
Behaviour with the following modules: Perception Module, Biology Module/Stress,
Personality Culture & Emotion module, Social, Relations and Trust module, Cognitive
module, Memory module and Expression module. It describes a model (Ortony, Clore
and Collins, 1998) for emotions: using 11 pairs of opposite balanced emotions; pride-
shame. These expressions capture the major dimensions of concern in any emotional
model values, relationships, and temporal aspects. In this research the determinants
of the stress: effective fatigue, event stress and time pressure are modelled.
2.1.1 Use of the models for elderly & disabled user groups
The early input of psychology into the field of disability focused on the emotional sequel
of disability rather than on disability per se. However, the conceptualization of disability
as behaviour enabled the application of theories of behaviour and behaviour change to
describe, explain and reduce disability (Johnston, 1994). Operant behaviour models
applied to activity limitations associated with chronic pain were perhaps the original
demonstration of both the concept of disability-as-behaviour and the importance of
psychological constructs in relation to disability reduction (Fordyce et al., 1968).
Social cognition models have been used to predict a wide variety of health related
behaviours and health outcomes in clinical and non-clinical samples (Armitage &
Conner, 2000; Armitage & Conner, 2001; Godin & Kok, 1996; Hagger et al., 2002;
Hobbis & Sutton, 2005). Control cognitions in particular are consistent predictors of
disability. Perceptions of control predict disability associated with several chronic
conditions including, osteoarthritis, stroke and myocardial infarction (Bonetti &
Johnston, 2008; Ewart, 1992; Jonston, 1999; Orbell et al., 2001; Rejeski et al., 2001).
Salvucci et al. (2004) present an approach to model the effects of age on driver
behaviour based on their ACT-R-based driven model (Salvucci et al. 2001) architecture
and a previous work in the EPIC cognitive architecture applied to modelling of cognitive
ageing (Meyer et al., 2001).
The GOMS family of HCI models (e.g. KLM, CMN-GOMS, CPM-GOMS) is mainly
suitable for modelling the optimal behaviour (skilled behaviour) of users (John & Kieras,
1996) but do not model human failures. On the other hand, models developed using
cognitive architectures consider the uncertainty of human behaviour in detail but have
not been widely adopted (Biswas & Robinson, 2008a).
Researchers have already attempted to combine these two forms of model to develop
more usable and accurate models. Salvucci & Lee (2003) have developed the ACT-
Simple model by translating basic GOMS operations into ACT-R production rules. The
model works well to predict expert performance but does not work for novices. Young
et al. (1989) propose a Programmable User Model based primarily on the SOAR
architecture (PUM) and incorporating ideas from some other architectures as ACT-R.

Blandford et al. (2004) also implemented a PUM based on SOAR and developed a
program, STILE (Soar Translation from Instruction Language made Easy), to
convert the PUM Instruction Language into SOAR productions. However, this approach
also demands good knowledge of SOAR from an interface designer. The second
problem of existing modelling approaches comes from the issues with disability. There
is not much reported work on systematic modelling of assistive interfaces (Biswas &
Robinson, 2008a).
The AVANTI project (Stephanidis et al., 1998) models an assistive interface for a web
browser based on some static and dynamic characteristics of users. However, this
model does not address the basic perceptual, cognitive and motor behaviour of
users and so it is hard to generalize to other applications.
Biswas et al. (2005) breaks down the task of user modelling into several steps that
includes clustering users based on their physical and cognitive ability, customizing
interfaces based on user characteristics and logging user interactions to update the
model itself. However the objective of this model is to design adaptable interfaces and
not to simulate users performance.
Keates et al. (2000) measured the difference between able-bodied and motor impaired
users with respect to the Model Human Processor (MHP) and motor-impaired users
were found to have a greater motor action time than their able-bodied counterparts.
A motor behaviour model simulates movement limits and capabilities of users for
different input devices and interaction techniques (McKenzie, 2003). For able-bodied
users, most of the motor-behaviour models are based on Fitts law (Fitts, 1954) and its
variations (McKenzie, 2003). For users with disabilities, there is growing evidence that
their interaction patterns are significantly different from those of their able-bodied
counterparts (Keates & Trewin, 2005; Trewin & Pain, 1999).
Fitts' law (Fitts, 1954) is a model of human movement in human-computer interaction

and ergonomics which predicts that the time required to rapidly moving to a target area
is a function of the distance to and the size of the target. Fitts's law is used to model
the act of pointing (related to response time), either by physically touching an object
with a hand or finger, or virtually, by pointing to an object on a computer display using a
pointing device (http://en.wikipedia.org/wiki/Fitts%27s_law)
The applicability of Fitts law for motor-impaired users is a debatable issue (Biswas &
Robinson, 2008a). In general real life pointing tasks demand very fine coordination
between vision and motor-action and in those situations motor-impaired persons do not
precisely follow Fitts Law. There was some work to develop an alternative of Fitts
law for motor-impaired people. Gump et al. (2002) found significant correlation
between the movement time and the root of movement amplitude. Gajos et al. (2007)
estimated the movement time by selecting a set of features from a pool of seven
functions of movement amplitude and target width, and then using the selected
features in a linear regression model. Biswas & Robinson (2008b) did a more detailed
analysis of different phases of movement for several pointing tasks undertaken by
motor impaired users and developed a statistical model to predict the movement time
for a pointing task.

Biswas & Robinson (2008a) developed a simulator to help with the evaluation of
assistive interfaces. It can predict the likely interaction patterns when undertaking a
task using a variety of input devices, and estimate the time to complete the task in the
presence of different disabilities and for different levels of skill. The authors modelled
the optimal (expert) and sub-optimal (non-expert) behaviour separately. They used the
CPM-GOMS (John & Kieras, 1996) model to simulate the optimal behaviour. For sub-
optimal behaviour, they developed a new model. This model takes a task definition as
input and produces a sequence of operations needed to accomplish the task as output.
It simulates interaction patterns of non-expert users by two interacting Markov
processes.
The ANN-based approach (e.g. as in Dinerstein et al., 2004; Chan et al.,1995) is

promising for modeling elderly and/or disabled people because developing explicit
behavioural models for this target group can be quite difficult. Therefore, it might be
useful to implement such models or approximations thereof based on self-learning
techniques such as ANN which do not require explicit models and can learn the
desired input-output relations instead provided there is representative data available for
training and/or a possibility for the network to adapt itself online in a changing
environment and/or under prior uncertainty.
2.1.2 Description of existing computational models for each targeted

domain
Automotive: Salvucci et al. (2001) propose the use of cognitive architectures for
developing models of driver behaviour that integrate cognitive and perceptual-motor
processes in a serial model of task and attention management. A cognitive architecture
is a computational framework that incorporates built-in, well-tested parameters and
constraints on cognitive and perceptual-motor processes. All driver models
implemented in a cognitive architecture necessarily inherit these parameters and
constraints, resulting in more predictive and psychologically plausible models than
those that do not characterize driving as a multi-tasking activity. The authors
demonstrate these benefits with a driver model developed in the ACT-R cognitive
architecture. The model is validated by comparing its behaviour to that of human
drivers navigating a four-lane highway with traffic in our fixed-based driving simulator. A
recent version of the ACT-R based driver behaviour model can be found in (Salvucci,
2006).
While computational cognitive modeling has made great strides in addressing complex
dynamic tasks, the modeling of individual differences in complex tasks remains a
largely unexplored area of research. Salvucci et al. (2004) present an approach to
modelling individual differences, specifically age-related cognitive differences, in
complex tasks, and illustrate the application of this approach in the domain of driving.
They borrow ideas from rigorous work in the EPIC cognitive architecture (Meyer et al.,
2001) and extend them to the ACT-R architecture (Anderson et al., in press) and the
ACT-R driver model (Salvucci et al. 2001) to model the effects of age on driver
behaviour. Two validation studies are described that demonstrate how this approach
accounts for two important age-related effects on driver performance, namely effects
on lateral stability and brake response during both normal driving and driving while
performing a secondary task.

Smart Living Spaces: Chan et al. (1995) use machine learning techniques to control
environments and automatically raise alarms. An artificial neural network (ANN) is used
to learn the habits of a group of elderly or disabled people (presence/absence, location,
mobility, and comfort temperature). The network is trained over a given period, and
then is used, e.g., to control the temperature of a room based on expected occupancy,
or raise an alarm when the expected location of a person does not match the response
of a multisensor system. Similarly, Rivera-Illingworth et al. (2007) proposed an integral
framework consisting of the development of both a simulated and two real
environments monitored by a system using an ANN-driven embedded agent able to
work with online, real-time data from a network of unobtrusive low-level sensors. The
objective of the system is to discover a normal habitual set of activities of elderly
people and identify novelties in ongoing behaviour, which do not fit the model of
previously constructed behaviours or activities. Such novelties could, for example,
relate to falls, or deteriorating medical conditions.
Mihailidis et al. (2003) developed a system for an intelligent environment to support

aging-in-place of people with dementia. The system consists of modules for tracking,
planning, and prompting. The tracking module monitors the actions of the user by
determining the spatial coordinates of the persons body and hands within the
environment. Once these coordinates have been determined, the systems planning
module determines what step the user is completing, whether the step being completed
is correct, and which plan (i.e. sequence of steps) the person is attempting. If the
system detects that the user has made an error, such as completing a step out of
sequence or missing a step altogether, a prompt will be selected and played. The
planning module is developed using partially observable Markov decision process
(POMDP) algorithms.
More references to works on smart living spaces for the elderly and disabled can be
found in (Rivera-Illingworth et al., 2007) and under
http://en.wikipedia.org/wiki/Home_automation_for_the_elderly_and_disabled
Workplace: Ghaffari, Genaidy (2005) employ ANNs for optimizing human performance
in workplaces. The work compatibility is an integrated work design criterion that
improves different aspects of human performance in a workplace. A neural network
model has been developed for the work compatibility matrix. Operating zones have
been established in which an employee can perform with a good performance and
sustainability. A survey has been conducted and an application example has been
provided to demonstrate the applicability of the human performance optimization
model.
Infotainment: Charles & McGlinchey (2004) review some of the successful uses of
ANNs in games, identify the positive elements of their use, and discuss some of the
factors that have deterred their use amongst game developers. Intelligent character
animation is one approach to improve this aspect of player immersion (Charles &
McGlinchey, 2004; Dinerstein et al. 2004). Alternatively, fuzzy logic to control
intelligent agents (Karunaratne & Yan, 2001; van Waveren, 2001). Mac Namee &
Cunningham (2003a, 2003b) proposed the -SIC system modelling of non-player
characters (NPCs) in computer games. The -SIC system uses psychological models

to simulate characters personalities, moods and relationships. The outputs of these

models are used as inputs to an ANN which determines which interactions the
character should engage in.
The approaches used in modelling intelligent characters in 3D games can be also

utilized for creating virtual user models for other VERITAS application domains.
ANNs have been applied to assess the emotional state of humans based on
physiological responses from watching videos (Wilhelm et al., 2007; Kolodyazhniy et
al., 2010).
The GOMS family models are used for modelling of human-computer interaction (John
& Kieras, 1996) with skilled users. Models developed using cognitive architectures like
SOAR consider the uncertainty of human behaviour in detail but have not been widely
adopted. Usability issues for cognitive architectures are also supported by the X-PRT
system (Tollinger et al., 2005) for the CORE architecture. A problem of many existing
modelling approaches comes from the issues with disability. The AVANTI project
(Stephanidis et al., 1998) models an assistive interface for a web browser based on
some static and dynamic characteristics of users. However, this model does not
address the basic perceptual, cognitive and motor behaviour of users and so it is hard
to generalize to other applications. More on models for human-computer interaction for
elderly or disabled users see in section
Personal Healthcare & Well-being: Psychological and neurophysiological studies

have shown that a computer game can be an effective and amusing tool for personal
improvement in general (Eskelinen, 2001; Vincelli, 1999; Riva, 1997; Riva et al., 1999;
Palmer, 1990), and for the aged group in particular (Weisman, 1983; Mcconatha et al.,
1994; Czaja, 1996; Lawhorn et al., 1996).
Research has proven that the use of electronic media, especially computer games, by
older adults and frail populations has positive effects such as ameliorate overall
cognitive functioning (Dustman et al., 1992), training of the sensory perceptual and
motor neuropsychological functions, attention and memory (Sano, 1988; Ott-Chevert et
al., 1998; Vroman et al., 1989; Peniston, 1991), increasing feelings of self-confidence
and life-satisfaction (Temple & Gavillet, 1990; Goldstein et al., 1997), and developing
affirmative attitudes toward aging and health (Butler & Lewis, 1973). Diverse virtual
applications are widely used for the diagnosis of individual problem areas (Yarnold et
al., 1996), and in health care (Greenleaf & Tovar, 1994; Moline, 1997; Bodenburg &
Technow,1992).
Beside testing/ training of brain activities, there are also PC-based desktop virtual
reality systems for rehabilitating (e.g. hand function in stroke patients, Jack et al.,
2001). Those systems use input devices, such as Cyber Gloves and force feedback
devices, that allow user interactions with a virtual environment. Those systems are
designed to exercise specific hand movements. Performance-based target levels
increase patient motivation and individualize exercise difficulty to a patient's current
state. Patients showed improvement on most of the hand parameters over the course
of the training. Subjective evaluation by the patients was also positive.

ANNs are used to model and assess the circadian rhythms in humans based on
ambulatory monitoring (Kolodyazhniy et al., 2009). The utility of the developed
circadian model lies in the possibility to accurately assess the endogenous circadian
(nearly 24h) rhythmicity and improve the well-being and/or cognitive and physical
performance by scheduling optimal sleep/wake times, hours of physical activity or
exercise, etc.
In (Virone, 2009) a pattern recognition model for assessing behavioural rhythms in the
framework of aging and technologies is presented. The method, previously tried and
tested using motion sensors installed in assisted living units, has permitted to establish
motion-based behaviours of older-people based on their habits in term of
displacements and activity levels. The method is expanded to measure more specific
patterns of everyday life activity assuming an activity can be pre-identified on the long
term using an activity recognition system. The study feasibility, carried out using semi-
artificial data, includes an attempt to model disruptive patterns of living linked with
dementia. The alert triggering method, part of the model is improved, and has been
evaluated using a real-case study to detect behavioural changes with a higher
sensibility. The changes can be related to particular chronic disease symptoms.
2.2 Analysis of the P&B computational models

The following tables contain an analysis of the P&B computational approaches taking
into account primary strengths and weakness.
Strengths of a model approach are:
it is an explicit model (descriptive)
applies for VERITAS users and for VERITAS domains
focused on non-skilled behavior
cost and efforts-effective
accurate, allows modeling multitasks based on psychological theories
allows modeling individualities
Weaknesses of the model:
not directly related with behaviour
no experiences with VERITAS domains and users
not explicit models (black boxes)
focused on skilled behaviour
allows single tasks

does not allow modelling individual differences
not based on psychological theories
This analysis is based on the tabled included in A1.1.2 Benchmarking of existing

models, standards and problems. However, just the most relevant approaches have
been included.
Modelling
Brief Descriptions Primary Strengths Primary Weakness
approach
GOMS (or Original GOMS A GOMS estimate of a Cannot address user

CMN- model. The GOMS particular interaction can unpredictability, e.g.
GOMS) concept is useful to be calculated with little due to fatigue, social
analyze knowledge effort, at little cost, and in surroundings, or
of how to do a task a short amount of time if organizational
in terms of Goals, the average Methods- factors. None of the
Operators, Time Measurement data GOMS models allow
Methods, and for each specific task has for any type of error.
Selection rules. been previously It is assumed that
measured experimentally user knows what to
GOMS to a high degree of do at any point, so
methodology accuracy. the models apply only
estimates how a to experts, not
simulated user Experience available with novices.
would perform in a mobile phone technology
simulated system to validate model Represents only
using a specified parameters for older skilled behavior.
set of procedures. adults by comparing Not useful for ill-
goodness-of-fit to defined problem
Can be modified for previously validated solving, exploration,
modelling of models of younger adults etc. Need to start
disabled and using the GOMS with a list of top-
elderly users modeling technique in level tasks or user
(Schrepp, 2006; both a simple dialing goals. List must be
Tonn-Eichstdt, task, and a more provided outside of
2006; Jastrzembski complex text messaging GOMS.
& Charness, 2007). task.[Jastrzembski(2007)]
Other models from An extension of GOMS

the family are: models has been used to
calculate the time to
KLM-GOMS, faster execute a task on a web
than CMN-GOMS page. This model [Tonn-
but less accurate. Eichstdt, H. (2006)]
Especially useful describes the interaction
when determining model of blind users
time it takes to type interaction strategies.
a phrase, correct a

Modelling
approach
realized error, or (Visual impairments)

select something
with a mouse.
CPM-GOMS Cognitive- Modelling of parallel The most difficult

Perceptual-Motor information processing by GOMS technique to
model of the the user. Can model implement
GOMS family (also multitasking behavior that
Critical Path can be exhibited by
Method) experienced users.
The most
sophisticated
GOMS model, also
for perception and
motor functions and
modelling of
multitasking. Based
on Model Human
Processor
(modification for the
elderly users see in
Jastrzembski &
Charness, 2007)
CORE Implementation of a Development of a set

(Constraint- new general of constraints is
based approach to difficult.
Optimal cognitive modeling
Reasoning based on Focused on predict
Engine) constraint- the time that a user
reasoning and will require to achieve
optimization. It has tasks with a given
associated a tool to interface.
help interface
designers predict
human
performance.
ACT-R ACT-R is a Learning and Memory. High-Level Human-

cognitive Machine-Interaction
architecture that Human-computer

Modelling
approach
can be used to interaction to produce and Multi-Tasking.

model low-level user models that can
cognitive tasks. assess different Efforts needed to
computer integrate the tasks as
its constructs interfaces,education. production rules and
reflect assumptions memory as chunks.
about human Allows Individual
cognition, which differences.
are based on
numerous facts Widely applied approach,
derived from for cognitive and
psychology behavioural modelling,
experiments. also of the elderly and
disabled (see e.g. Serna
Production rules of et al., 2007).
the form IF
condition, THEN An experience with
actions. Alzheimer has been
performed.
Last version of
ACT-R includes Stress, fatigue and
EPIC approach for emotions have been
the perceptual and overlay in ACT-R
motor systems. Gunzelmann, G., &
Gluck, K. A. (2008), Fum,
D. & Stocco, A. (2004).
Ritter, F. E., Reifers, A.
L., Klein, A. C., &
Schoelles, M. J. (2006).
ACT- A higher-level Works well to predict Does not work for

SIMPLE framework built on expert performance. novices
the ACT-R
cognitive
architecture. ACT-
Simple includes a
set of basic
perceptual, motor,
and cognitive
commands that
compile down to
ACT-R production
rules. ACT-Simple
models by
translating basic
GOMS operations

Modelling
approach
into ACT-R
production rules.
Table 1: Behavioural & psychological computational modeling approaches analysis
Modelling
approach
EPIC EPIC is designed to Multi-Tasking Low-Level Cognitive

assess low-level Processes.
multi-tasking Well-developed
performance. approach for regular
users, adaptation for
It has been behavioural modelling
integrated in last in the elderly and
version of ACT-R. disabled needs to be
checked. In (Biswas et
al., 2008a) it is
mentioned that EPIC
can be adapted for
visually impaired
users.
Distinct perception
models.
SOAR SOAR is a cognitive Reasoning, Learning, High-Level Human-

architecture that can Problem Solving. Machine-Interaction.
be used to model
low-level cognitive Well-developed Not so focused on
tasks and for approach for regular emulating the human
developing systems users. Can model user behaviour but on
that exhibit errors (Jones et al., creating intelligent
intelligent behavior. 2007). Further agents.
adaptation for
behavioural modelling
in the elderly and
disabled needs to be
checked

Modelling
approach
PUM Programmable User Demands good

Model (PUM) by knowledge of Soar
using the SOAR from an interface
architecture (and designer.
possibly some other
modelling
frameworks)
Artificial ANNs are inspired Can approximate a Need a representative

neural by the parallel behavioural model data set for training.
networks information when it is not available Generally, represent
(ANNs) processing in natural explicitly or is difficult the black-box
brains. They are, to develop. A huge approach to modelling
essentially, networks variety of neural (i.e. interpretation of
of simple processing network models exist the trained net can be
units (artificial with many different difficult)
neurons) working in training algorithms.
parallel. Knowledge
acquisition is done Pre-trained models
via training based on can be easily shared
examples using a and re-used.
representative set of
input-output data. Experiences modeling
the motivation of a
Generic modelling user [ Portegys, T.
approach is not (1999)]
limited to cognitive
and/or behavioural
modeling.
Fuzzy logic An extension of the Fast execution, low Fuzzy rules need to be
conventional memory requirements, defined (can be learnt
Boolean logic, in much less rules are automatically if
which a truth value required compared to combined with ANN-
can be anything the conventional like training methods)
between 0 and 1. An Boolean logic. either manually without
intelligent agent can any training data, or
use fuzzy relations, extracted from the data
e.g., to specify how if available
much it wants to do,
have or use
something or to
relate sets of

Modelling
approach
emotions to
corresponding
behaviours.
Generic modelling
approach, not
limited to cognitive
and/or behavioural
modelling.
Markov Statistical approach Can model sub- Limited in the sorts of

models (Also that tells the optimal (non-expert) interaction that can be
Mixture designer about all behaviour, including modeled
Models) possible behaviours user errors, also for
of the user. The modelling assistive
results of the interfaces for special
analyses are needs (Biswas et al.,
expressed in terms 2008a)
like on average or
can be plotted
graphically. Whether
the distributions
describe individual
users over a period
of time or a
population of users
is determined by
how the probabilities
and models are set
up.
User behavior is
viewed as a
probabilistic model.
-SIC Uses psychological Can model individual Needs a

models to simulate psychological representative data set
characters characteristic of an for training.
personalities, moods intelligent agent.
and relationships.
The values of these Simple psychological
models are used as models of personality,
inputs to an ANN mood, and
which drives relationship. Generic

Modelling
approach
characters social model that can be

behaviour adapted for VERITAS
provided training data
are available.
HUVAS- Human Values Defines a formal scale It does not reflect

SUMM Scale from Smart of universal human behavioural attributes
User Models, values that can be such as fear, anxiety,
[Guzman,J. Methodology. applied to any nervousness, etc
(2008)] Methodology based individual. These
on users values reflect the No experiences with
customisation personality of the elderly and disabled
considering the person. people.
Human Values
Scale acquired from Categorizes the Experience in the
the Smart User human values in commerce area, no
Model, which different levels: results in the 5
improves the clients Universal Human VERITAS domain
recommendation by Values, Basic Human areas.
using a message Values, Human
and one-to one Values per each Basic
dialogue. Human values.
The Smart User Defines

Model is defined as questionnaires and
an adaptative user metrics parameters to
model that captures extract the results.
the evolution of a
users emotion An experience
(preference, taste an measuring the
emotions). The HUVAS-SUMM in
emotional different times of the
component of the users life has been
Smart User Model is performed. This
a set of attribute- experience shows that
value pairs the Human Values
representing the depends on the users
emotional state of a age.
user.

Modelling
approach
Human Explores in A more theoretical Games domain. The

Behavior emergent macro- description of the other domains arent
Models for behavior due to model, not software addressed.
Agents in micro-decisions of oriented.
bounded-rational No experiences with
Simulators agents and with Good description of Aged and Disabled.
based on developing a the stress with rules
PMFs framework that and values.
(Performance permits one to Description of the
Moderator examine the impacts determinants of the
Functions) of biology, stress, stress: Effective
personality, culture, fatigue, event stress
emotion, social and time pressure.
relations, and
decision making Description of Human
upon human coping Values Tree.
behavior.
Functions that
Defines a High Level describe the intensity
of Unified of emotions.
Architecture of
Defines a Decision
Behaviour with the
Making Module based
following modules:
on emotions,
Perception Module,
stressors, memories,
Biology
etc. Using a stress-
Module/Stress,
constrained
Personality Culture
subjective-expected
& Emotion module,
utility formulation.
Social, Relations
and Trust module,
Cognitive module,
Memory module and
Expression module.
Defines the
PMFserv
implementation of
the Architecture.
OCC model for

emotions: using 11
pairs of opposite
balanced emotions;
pride-shame. These
expressions capture
the major

Modelling
approach
dimensions of
concern in any
emotional model
values,
relationships, and
temporal aspects.
Unified 3D Models affective Ability to express the Only defines affective

Affective states for context- secondary states as areas.
Model based aware applications nuances of the basic
on fuzzy where a generic 3D affective states and to It is not directly related
sets. affective model explicitly express the with behaviour.
starting from the intensity of each state.
activation evaluation Not experiences in
plane is proposed by MMI database with VERITAS domains and
giving a metadata and the users.
methodology to diversity of the people
define basic in terms of sex, age,
affective areas. and ethnicity.
Basic or 'prototypic' Defines intensity of

patterns of emotion emotions.
are described:
happy, sad, angry,
fear, neutral,
surprise and disgust;
With a mixture of
these basic
emotions, hundreds
of mixed emotions
can be produced.
Combination Models the optimal Used for modeling the Objective of the
of Models: (expert) and sub- use of an assistive complete model is to
CPM-GOMS optimal (non-expert) application by a design adaptable
+ Markov for behaviour disabled. interfaces and not to
Cognitive + separately. Uses the simulate users
Motor- CPM-GOMS model Simulates HCI of both performance.
Behaviour to simulate the able-bodied and
model optimal behaviour. disabled users.
(Biswas et For sub-optimal
al., 2008a) behaviour, a new Work for users with
model: This model different levels of skill.
takes a task
Easy to use and
definition as input
comprehend for an

Modelling
approach
and produces a interface designer.

sequence of
operations needed
to accomplish the
task as output. It
simulates interaction
patterns of non-
expert users by two
interacting Markov
processes. One of
them models the
users view of the
system and the
other signifies the
designers view of
the system.
The cognitive model

takes the output of
the perception
model and
determines an
action to accomplish
the current task. The
motor-behaviour
model predicts the
completion time and
possible interaction
patterns for
performing that
action. (Biswas et
al., 2008a)
Table 2: Behavioural & psychological computational modeling approaches analysis

(cont.)
The present analysis of current computational models shows a high number of models
that focus on different aspects and serve to specific purposes:
GOMS family models are more focused on modeling interactions, not

psychological assumptions.
ACT-R, EPIC and SOAR are cognitive architectures that integrate psychological
theories and assumptions of how the human mind thinks and feels.

Version 6.0 of ACT-R includes EPIC approach for the perceptual and motor
systems.
ACT-R allows modelling individual differences with overlays. Several

experiences with elderly and driving domain have been performed.
SOAR does not focus on emulating human behaviour but on creating intelligent
agents with optimal behaviour.
PMFs (Performance Moderator Functions) can be used as moderators of

known psychological facets: stress, emotions, etc
ANNs models and difficult to interpret because they are essentially black box
models and do not offer a descriptive approach.
-SIC system is based on ANNs and is focused on social interactions of the

characters.
In Markov models, user behavior is viewed as a probabilistic model. It is limited

in the sorts of interaction that can be modelled.
Fuzzy logic is useful to model simplified animated characters behaviour,

without including psychological theories of human mind.
The present analysis points out ACT-R as the most promising and extended modelling
approach, that is consistent with the Abstract User Model intended in VERITAS: an
explicit descriptive model of human cognition and behaviour that can be afterwards
parameterized. In the following section a complete analysis of ACT-R as reference
cognitive architecture will be shown.

3 Cognitive and behavioral integrated

architectures
A cognitive architecture is a specification of the structure of the brain at a level of
abstraction that explains how it achieves the function of the mind. They are used for
describing the mechanisms of cognition that are fixed across users and across tasks
(Anderson, 2007; Newell, 1990).
Cognitive science and cognitive modeling are increasingly interested in creating

models that include psychological facets such as emotions, stress as a way to define
and understand psychological states. These models provide useful definitions of
psychological states and provide a way to explore them, how they may be related, and
how they influence cognition and performance. The psychological facets are modeled
as moderators of cognition and behavior, modifying architectural parameters (ie:
overlays).
In the last years, several experiences for including psychological facets into cognitive
architectures have been performed. They have been created by modifying architectural
parameters. These cognitive parameters can be moderated at runtime, leading to
systematic variations in behavior.
In the following sections we well describe the behavioural and psychological theories
underneath, an extended example of the cognitive architectures: ACT-R, and the
cognitive moderators that modify the overall performance of the users.
3.1 Human behavioral and psychological theories

Psychology is the science of mind and behavior (Merriam-Webster, accessed 2010).
Psychology attempts to understand the role of mental functions in individual and social
behavior, while also exploring the physiological and neurobiological processes that
underlie certain functions and behaviors. However, there is not a unified theory about
how the human mind works. Amongst the most important psychological theories, we
have focused on those that describe the behavior and the cognitive processes of
human mind: behaviorism and cognitivism theories. These theories describe
theoretical models of human cognition and behaviour, and are relevant for
comprehending the complexity beneath the models and for understanding the theories
which cognitive architectures have been based on.
3.1.1 THE BEHAVIOURAL APPROACH
3.1.1.1 Classical Conditioning

Studies on classical conditioning began with experiments conducted by the Russian
physiologist Ivan Pavlov, at the beginning of the 20th century. Pavlov discovered that
animals can learn to respond to certain stimuli from their unconditioned reflexes,
by experimenting with dogs that were taught to salivate after listening to the sound of a
bell, in principle a neutral stimulus. This way of learning is known as Classical
Conditioning.

Key elements in Classical Conditioning (Paulov, 1927):
1. Unconditioned stimulus: (non-conditioned). The piece of meat. Since food is a

stimulus and the view of it produces salivation without further help.
2. Unconditioned response: (non-conditioned). The dogs salivation when it

occurs naturally, not learned.
3. Neutral stimulus: The sound of the bell itself.
4. Conditioned stimulus: The bell. When the bell sounds and produces the
salivation in the dog. It is a naturally conditioned stimulus.
5. Conditioned response: The dogs salivation, when it occurs in response to the

sound of the bell. It is a learned response.
In summary we can say that Classical Conditioning is to learn a conditioned response

involving the construction of an association between a conditioned stimulus and an
unconditioned stimulus. Using them together, the conditioned stimulus that was
naturally neutral, adopts the properties of the non conditioned stimulus.
Generalization: It is the response to similar stimuli. For example, once the dog
has learned to salivate with the sound of the bell, it will produce saliva when
listening to similar sounds.
Extinction: If the bell no longer rings when food is presented to the dog,
salivation will eventually disappears with just the sound of the bell.
Discrimination: It explains that specific responses are given to specific stimuli.

For example, the dog can learn to discriminate between similar sounds and
distinguish which sounds are associated with the presentation of food.
Spontaneous Recovery: Extinct replies can be retrieved after a short period of

stimulation, but cease again if the food is not presented.
Conditioning Of Higher Order: Once the dog has acquired conditioning with the
association of the sound of the bell and food, then other elements can be tried,
at the same time it can be tried, such as turning on a spot light. In this way the
dog will also produce saliva just by switching on a spot light (without the sound
of the bell).
As learning is a way to modify behavior, then behavior modification procedures

developed by the behaviorists have proven to be very useful for many teachers and
schools during the last generations. Although in disagreement with much of the
influence that Behaviorism has had in education, due to the viewpoint that students are
empty individuals that acquire behaviour and that unwanted behavior can be replaced
or removed, we have to admit the great influence of Behaviorism in traditional
education.

In relation to learning, these theories take into account the process involved in the
acquisition of knowledge and the interactions that occur between the different
elements of the environment.
Behaviorism and Pavlov encouraged the use of procedures destined to manipulate

behaviors in the school system. The information and data organized in a certain way
were basic stimuli to which students, merely as recipients, had to make elections and
associations within a narrow margin of possible correct answers which, if carried out,
received the corresponding reinforcement. Although the principle of reinforcement not
always implied the use of stimulus, the consequences of this in school practices were:
outside motivation to the student, repetition and memorization, predominance of the
test method - error, teaching and evaluation subject to award/punishment and almost
absolute student dependence on external stimuli.
The relationship between educator and learner was extremely poor, because it was not
conceived that students had intellectual initiative. On the contrary, it provided technical
and operational resources so that the traditional role of the educational process would
be strengthened thanks to a highly sophisticated education plan.
3.1.1.2 Operant Conditioning

The paradigm of Skinner was eminently positive, influenced by the typical scientific
event of the decade in which his work was developed, based on the work of Pavlov
(1927) and Thorndike.
Discriminative stimulus: It is that where a certain type of behaviour is highly

likely, due to having been reinforced earlier by a stimulus.
Operant behavior: Is the typical natural conduct of an organism or being, in

other words, how it behaves in the environment.
Reinforcing stimulus: It is a stimulus which increases the likelihood of a

contingent response.
Generalization: Wwhen a response is reinforced, this produces an increase in

another similar response.
Discrimination: When an organism behaves differently in the presence of two

stimuli.
Extinction: It is a procedure in which an operant behaviour which has been

reinforced ceases to be strengthened and results in response ending, in other
words, it ceases.
Types of Reinforcements:
1. Positive: all stimuli that, when present, increases the probability of the
appearance of a particular behaviour.
2. Negative: any aversive stimulus that when removed increases the probability of
that behaviour.

3. Extinction: occurs when a stimulus that previously strengthened a type of

behaviour ceases to act.
4. Punishment: similar to extinction, works to reduce the conduct.
5. Multiple: the application of two or more different programs.
6. Composite: reinforcing one response with two or more different programs.
7. Concurrent: reinforcement of two or more responses with one or more

programs.
8. Punishment: This occurs when using an aversive stimulus for a response rate
reduction.
Skinner sees learning as the product of a stimulus-response relationship, thus

appropriate behaviours are reinforced through awards and the inappropriate
through punishment.
Skinner attempted to demonstrate in the area of school learning that, through threats
and punishments much lower positive results are achieved and with worse side effects
than through the system of positive reinforcement. His principle for progress in the
courses and maximum achievement is based on the activity of the students. His best-
known application is programmed teaching based on the premise that success in
certain tasks acts as reinforcement for subsequent learning.
3.1.1.3 Difference between Classical Conditioning and Operant Conditioning:
CLASSICAL CONDITIONING OPERANT CONDITIONING
A neurological stimulation becomes an An apprentice operates in the

associated reflection. Example: the sound environment, and receives a reward for a
of the bell is associated with salivation. particular conduct. Example: the individual
operates a lever and receives food.
It is a reflection. A simple innate It is not a reflection; the subject must carry

behaviour, which produces an inevitable out an activity to get something in return.
response to a change in the
environmental situation.
It is a combination of two factors, stimulus Behaviour is external, as it has an effect

and response. Behaviour is internal and on the individuals outside world. The
innate. relation of the behaviour has an effect
which increases the likelihood that under
similar conditions the same behaviour
could appear.
Table 3: Difference between Classical Conditioning and Operant Conditioning

3.1.2 THE COGNITIVE APPROACH

In 1956 Dr. Albert Ellis started a new approach in psychotherapies which today is
known as Rational Emotive Behavior Therapy (REBT). Some years later, in 1962, Dr.
Aaron T. Beck initiated a new proposal in psychotherapies called Cognitive Therapy
(CT). Very soon this new therapy would acquire great prestige for its effectiveness in
addressing depression. Both approaches, together with other new integrative,
behavioural and social guidelines, are included in what could be called today the
"cognitive behavioural approach" in psychotherapies, which adopts the cognitive
psychology model.
The paradigm of cognitive psychology (and cognitive therapies) is based on the

principle of cognitive mediation. In summary: reaction (or emotional or behavioral
response) when facing a situation (or stimuli) is mediated by the process of
significance (or assessment) in relation to objectives (goals, desires) of each
person. Supporters and defenders of this model include Richard Lazarus (1991).
Cognitive structures, generally called schemes, which give rise to evaluations and
responses, are the result of genetic predispositions and learning generated through
each person's own experiences.
Aaron T. Beck started the development of what he called "Cognitive Therapy" at the
University of Pennsylvania in the early 1960s (1962), as a psychotherapy for short-
term, present-oriented depression.
The model agreed with the concept of cognitive mediation proposed by Albert Ellis in
1956. The headquarters of this approach, which for many years was located in the
Center for Cognitive Therapy at the University of Pennsylvania in Philadelphia,
Pennsylvania, is currently the "Beck Institute for Cognitive Therapy and Research"
located in Bala Cynwyd (suburban area of Philadelphia, the State of Pennsylvania,
USA), founded in 1994 by its current President, Aaron T. Beck, and directed by his
daughter, Judith S. Beck.
3.1.2.1 BASIC CONCEPTS

"The Cognitive Therapy is based on the cognitive model which postulates that the
emotions and behavior of people are influenced by their perception of events. A
situation is not in itself the one that determines what a person feels, but rather how the
person interprets the situation (Ellis, 1962; Beck, 1964). Therefore the way people feel
is associated with the way they interpret and think about a situation. The situation itself
does not directly determine how they feel; their emotional response is mediated by their
perception of the situation." (Beck, J.,1995)
In other words, the cognitive therapies are based on the following basic assumptions:
People are not simply environmental stimuli receptors, rather, they actively build
their "reality".

Cognition is a mediator between stimuli and responses (cognitive, emotional or

behavioral).
People can access their cognitive content.
Modification of the cognitive information processing (attribution systems, beliefs,

thinking, etc) is central to the change process.
The cognitive system:
Using "S" as the stimuli or situation, "P" as the system and cognitive processing
(thoughts) and "R" as response or reaction. The model is represented by this relation:
Reactions may be: emotional, behavioural or physiological.
3.1.2.2 MOST IMPORTANT COGNITIVE THEORIES
3.1.2.2.1 The Social Learning Theory

The social learning theory is derived from the work of Cornell Montgomery (1843-
1904), who proposed that social learning takes place through four main stages: close
contact, imitation of superiors, understanding of concepts and role model behavior.
Julian Rotter distanced himself from the theories based on psychosis and behaviorism,
and developed a theory on learning. In Social Learning and Clinical Psychology
(1954), Rotter suggests that the result of behaviour has an impact on the
motivation of individuals to realize a specific behaviour. People want to avoid
negative consequences, and obtain positive results. If one expects a positive
result, or thinks that there is a high possibility to produce a positive result, then he/she
will be more likely to engage in such behaviour. Behaviour is reinforced by positive
consequences, leading the person to repeat it. This social learning theory suggests that
behaviour is influenced by factors or stimuli from the environment, and not only by the
psychological.
Albert Bandura (1977) expanded Rotters idea and a previous work by Miller and
Dollard (1941), and relates to the theories of social learning developed by Vygotsky
and Lave. His theory incorporates cognitive and behavioral learning aspects.
Behavioral learning assumes that people behave in a certain way due to their
environment. Cognitive learning assumes that psychological factors are
important influences on people's behaviour. Social learning suggests that a
combination of environmental (social) and psychological factors influence behaviour.
The Social Learning Theory points out three conditions for people to learn and model
their behaviour: retention (remembering what one has observed), reproduction (ability

to reproduce behaviour) and motivation (having a good reason) for wanting to adopt
such a conduct.
Learning occurs not only by direct and personal experience, but also, mainly in human
beings, by observation of others (imitation) as well as information received from verbal
or visual symbols, constituting critical variations in the human learning process.
3.1.2.2.2 Information Processing Theory

This is a theory that emerged in the 1960s and is a psychological explanation about
the process of learning. It fits into the scientific-cognitive range, and is influenced by
computer and communication theories. It is not a single theory, but rather a synthesis
known generically as: information process.
It is important to state, with Gimeno and Perez (1993), that the anthropological concept
of this theory is that "man is an information processor, whose fundamental activity
is to receive information, develop it, and act on it. In other words, every human
being is an active processor of experience through the complex system in which
information is received, encoded, transformed, accumulated, recovered and used".
Therefore, it can be inferred that the subject does not necessarily interact with the real
world, but rather the interaction is with the subjective representation, therefore ensuring
an internal process (cognitive) of learning. That is, how it is assumed and processed.
During the formulation of this theory, the emphasis given to the internal resources is
noteworthy, without forgetting the inclusion of the external resources of the
behaviorist model, as set out by Mahoney (1974) (see Figure 1). So, it is seen that
there are three structural elements: sensory register: receives internal and external
information; short-term memory: short term storage of selected information; and
long-term memory: organizes information and keeps it available for a longer
length of time. However, these early theories are too simplistic and do not truly collect
all the brain functioning processes nor do they take into consideration the influence of
emotional and motivational aspects.
Figure 1: Information processing model. Mahoney, (1974)

Indeed, in recent years a series of focused problems have been raised, mainly
centered around the areas of perception, attention, memory systems ((in a broad
sense), knowledge structures, prose processing, etc., and these have gradually been
configured into what is now known as "experimental cognitive psychology" (Arnau
1986).
It is not at all surprising that the classic psychological topics have been addressed by
new "experimental paradigms" and theoretical schemes coming from information
processing. Due to the collection of studies carried out by Broadbent (1958), Peterson
and Peterson (1959), Posner (1967), Sperling (1960), Sternberg (1966), Mahoney
(1974) etc., specific experimental plans were standardized, and thus, the underlying
mechanisms responsible for information processing from a cognitive perspective
became apparent. All of this has led to the formulation of integrative models that
embrace the experimental results and, at the same time, have promoted the rapid
development of cognitive issues. In a recent study (published in 1984) by Donald A.
Norman, one of the most important driving forces behind current cognitive psychology,
the three basic models that have been proposed in the recent history of cognitive
psychology have been combined.
First, we can highlight the models known as "Stages Models" which are amongst the
first attempts to theoretically systematize the initial results obtained from experimental-
cognitive research (Atkinson and Shiffrin, 1968;) (Shiffrin and Atkinson, 1969).
These models, of linear and descriptive characteristics, collate a number of stages or

phases of processing between the entry of the physical stimulus and the corresponding
response. From the cognitive perspective, the stages models can be considered as
processing systems that analyze, develop and transform information. As a first stage,
the system analyzes and extracts the physical characteristics of the stimulus. At this
level, the model requires the presence of some memories, known as sensory registers,
capable of storing such features in accordance with specific coding. On the other
hand, in the final stages a series of decision-making, comparison and extraction of
meaning mechanisms are activated, connecting directly with the rest of the global
system processes. Due to the restrictive characteristics of the system and lack of
attention, not everything that goes into the model can be processed and elaborated.
As Norman (1984) (Figure 2) showed, this simply structured yet highly descriptive set
of models is often characterized by the presence of a few common mechanisms such
as analyzing characteristics, short term memory mechanisms, further processing and,
finally, decision-making sub-systems and the generation of responses (Massaro
1975).
A schematic representation of such a model is shown in Figure 2, which as you can

see, incorporates the main processing mechanisms, also known as modular
subsystems or stages in which the emotional and motivational aspects are taken into
account.
According to this schematic representation, when the stimuli or physical signals enter
the system, they undergo a sensory transduction, so as to be stored in a sensory
register. This information is then placed in (transferred to) a short-term memory

store. In the short-term memory store, the information is recognized and identified,
which implies a series of previous stages responsible for recognizing patterns. All of
these processes tend to act automatically, in a way that conscious control seems to
have an influence on the short term memory structures, enabling the subject to use
strategies to interpret and attend to the material (Baddeley 1976).
Intentions
Desires
Motivations
Thought
Emotions
Processes
(Executive
Functions)
Memory Structure
(Long Term Secondary Memory and Short

Term Primary Memory)
Pattern Recognition Motor Programs
Preliminary Analysis and

Sensory Memory Systems
Motor Control Systems
Sensory Transduction Effectors
(Perception and Attention) Speech, muscles, members.
Physical signs Sound and Movement
Figure 2:Stages of the information processing models (Norman 1981)

3.1.2.3 THE INFLUENCE OF AFFECTIVE STATES IN PROCESSING

INFORMATION
There are many definitions of 'personality'. For example, Professor Rey Ardid refers to
the personality as the structural and dynamic, perennial and unique to each individual
foundation in which the somatic make-up, the instincts and tendencies, affection, will,
intelligence, values scale and also, indirectly but evidently, everything around the
human are involved as main components.
This definition is interesting to understand the functioning of the human mind. In order
to better manage this concept, a proactive model will be drawn upon that will greatly
help to understand the different terms that will be used.
N
SAY
T
V
RESPONSE OBJECTIVE
I
R
F
O DO
SUBJECTIVE
Figure 3: How the human mind functions (Rey Ardid)
People think (T) and Feel (F) and the thoughts and feelings are not separate aspects
but are always connected and mutually influential. In other words, what the individual
thinks in a specific moment is associated to specific feelings and vice versa (although
this correlation is often unconscious, especially when the emotions are low intensity).
When a real or virtual stimulus is received from the environment (situation,

physiological state or mental state - fantasy, memory, recollection, dream, etc.) a
response is generated. However, this response is never automatic; it is not directly
conditioned by the stimulus. In other words, between both (E-R) there is a person
who decides on the one hand the response that will be given and on the other by
filtering what is received and the response that is given, or in the same scenario, the
stimulus invites but does not oblige. It is true that there are many responses that are
automatic because they are responses to stimuli of little relevance, but when the stimuli
are important we must consider which response we wish to give.

This process plays an important role, in addition to the thoughts and feelings in this
moment, our frames of reference, beliefs (the concept that each one of us has for
ourselves, everyone else, life and the world), and thus there are no two people alike
because frames of reference are created based on the relation between thoughts and
feelings and the entire life experience. This frame of reference is built from the day
of birth until the present day and as a result of this, everyone is different. The
vast majority of conflicts are frames of reference conflicts. For this reason, in order to
manage a conflict, agree to analyse the frames of reference and find what they have in
common, instead of what is usually done, which is to find differences, because in
searching for differences, positions are secured and the conflict is polarised. If possible
agreement points and what we have in common are sought instead, the conflict may
improve.
This frame of reference acts as a filter and filters the stimulus and the response.
The response may be objective (what is said/done) or subjective. The objective
response is the behaviour or conduct. The personality is what we think, feel and do
and/or say. Thus, thought (Cognition: Perception, attention, memory, etc.) (T) feelings
(emotion and motivation) (F) and behaviour (response) (B) are mutually influential, in
such a way that if there is a change in any one of them, the other two will automatically
change.
Figure 4: How the human mind functions (Rey Ardid)
What we feel is a source of internal information for ourselves that bears relation to the
satisfaction or dissatisfaction of needs and desires, in the situation the individual is in at
any given time. The emotions mobilise and inform, and when integrated with reason
make us wiser than when only the intellect is used. The emotions provide a rich source
of information about reactions when facing situations.
For this reason, they are very important: To feel (F) and be aware of what is felt (T).
When the individual feels something, a large amount of physical and psychological
energy is mobilised, which is sometimes experienced as stress and which enables the
organism to take opportune actions (B) with the aim to reduce physical and emotional
stress.

To recognise (T) and express (B) what is felt (F) enables us to know ourselves and
others better, notably improving communication.
To adapt to the environment, the organism has a sensory device which collects
information about the conditions of the environment, and a neural device that
processes this information, in order to readjust the biological parameters and maintain
organic homeostasis.
Whatever the outside world is like, we use the senses to explore and delimit. The
world is an infinite number of possible sensory impressions, and we are only able to
perceive a small part of them. The part we can perceive is then filtered by our own
experience, culture, language, beliefs, values, interests and assumptions. Each of us
live in our own unique reality, built by our own sensory impressions and individual
experiences of life, and we act according to what we perceive: our model of the
world.
The map is not the territory described. We take notice of those aspects of the world
that we find interesting and ignore others. The world is always richer than the ideas
that we hold about it. The filters that we place in our own perceptions determine what
kind of world we live in.
Albert Ellis, one of the founders of cognitive therapy, asserts that "feelings are the
parents of our thoughts and thoughts of our actions. Ellis developed the ABC model
to explain the above statement. He explains it as follows: A corresponds to
Activating Event and C to the Consequences. We can anticipate the consequences
of an Activating Event not because of the Activating Event itself, but because of B
which corresponds to Beliefs, in other words it is the beliefs that are the cause of
one consequence or another.
3.2 Cognitive architectures: ACT-R

Cognitive architectures are theories for simulating and understanding how human
mind works. They use cognitive psychology known theories (see previous section) and
their assumptions (about working memory, attention, decision making) and implement
them in a computer program that takes in knowledge representations (typically rules,
but it could include other types of task knowledge such as plans) to reproduce human
thinking. Unlike AI systems, in a cognitive architecture the application of knowledge
suffers from simulated limitations of human knowledge application, and often includes
ways for simulating errors in performance.
One of the most extended is ACT-R, which is a hybrid cognitive architecture that
provides a conceptual framework for creating models of how people perform tasks.
ACT-Rs constructs reflect assumptions about human cognition, based on numerous
facts derived from psychology experiments.
ACT-R follows a modular organization: cognition emerges through the interaction of a

number of independent modules, each of these associated with specific brain regions
(cortical regions of human brain). ACT-R contains elaborate theories about the internal
processes of these modules.

On the exterior, ACT-R looks like a programming language, and like a programming
language, ACT-R is a framework: for different tasks (e.g., memory for text or for list of
words, language comprehension, communication, aircraft controlling), researchers
create models (aka programs) that are written in ACT-R and that, beside
incorporating the ACT-R's view of cognition, add their own assumptions about the
particular task.
These assumptions can be tested by comparing the results of the model with the
results of people doing the same tasks. By "results" we mean the traditional measures
(quantitative measures) of cognitive psychology:
a. time to perform the task,
b. accuracy in the task,
c. neurological data such as those obtained from FMRI.
A computer simulation of the ACT-R architecture allows us to work out precisely what
ACT-R models predict about human cognition.
ACT-R grew out of detailed phenomena from memory, learning, and problem solving
(Anderson, 1983, 2007; Anderson & Bower, 1973). ACT-R is suited more for slightly
lower-level phenomena, and is more suited for predicting reaction times, particularly for
tasks under 10 s. ACT-R's outlook is more biased towards predicting reaction time
means and distributions because it arose out of a more experimental psychology
tradition.
3.2.1 ACT-R structure

The overall structure of ACT-R (ACT-R version 6.0) is illustrated in the following figure
and illustrates the eight modules that are standard as part of the ACT-R 6.0 simulation
system. Each module is associated with a specific brain region. The basic motivation
for a modular structure is to get the best performance possible given the limitations of
brain processing.

Figure 5: ACT-R 6.0. Each ACT-R module is associated with a function
As described in the previous section, ACT-R follows the common approach of the
cognitive theories:
There are two perceptual modules: a visual module and an aural module, which
receive the external stimuli from the environment.
There are two response modules: a manual module and a vocal module which are in
charge of provide the interaction with the environment.
The other four modules are responsible for different aspects of central processing:
The imaginal (problem state) module holds a current mental representation of the
problem. For instance, in the context of solving an equation such as 3x 7 = 5, it
might hold a representation of an intermediate equation such as 3x = 12.
The declarative module retrieves critical information (facts known by the user)
from declarative memory (ie. 7 + 5 = 12).
The goal module (control module) keeps track of ones current intentions in solving
the problem.
Finally, the procedural module embodies various rules for behavior, such as the
rules for solving equations. This central procedural system can recognize patterns
of information in the buffers and respond by sending requests to the modules.
These recognizeact tendencies of the central procedural module are

characterized by production rules. Since the production system can execute only a
single rule at a time, it becomes the central bottleneck (Pashler, 1998) in the overall
processing. Therefore, cognition can be slowed when there are simultaneous,
different demands for processing the information in the buffers of the modules. The
idea that such a central bottleneck exists reflects another emerging near-
consensus in cognitive psychology; ACT-R gives an architectural expression to this
consensus.
Each of these modules is capable of massively parallel computation to achieve its

objectives. For instance, the visual module can process the entire visual field, and the
declarative module can search through the large database of memories. However,
when it comes to communication among the modules, there are serial bottlenecks.
The only way these modules can communicate is through buffers (storage/memory)
associated with each module. Only a little information can be put into a buffer
associated with the modulea single object perceived, a single problem state
represented, a single control state maintained, a single fact retrieved, or a single
program for hand movement.
Communication among all the modules is achieved via a procedural module. The
procedural module can respond to information in the buffers of other modules and put
info on them.
A significant architectural constraint in ACT-R is that only a single production

rule can execute at a time. Moreover, it takes 50 ms for a production rule to fire
which is the time needed to complete the multisynaptic loop through the basal ganglia.
In the following figure we can see the association of each module with the specific
brain region:

Figure 6: ACT-R 6.0. Each ACT-R module is associated with a specific brain region
3.2.2 Human knowledge in ACT-R

Basically, ACT-R models how humans recall chunks" of information from memory and
how they solve problems by breaking them down into subgoals and applying
knowledge from working memory as needed.
Under ACT-R, human knowledge is divided into two disjoint but related sets of
knowledge: declarative and procedural.
- Declarative knowledge is the knowledge we are aware of (i.e.: Washington is the

capital of US). In ACT-R, declarative knowledge is represented by chunks, which
are the current set of facts that are known and goals that are active. Two examples
of chunks are:
The bank is closed on Sundays," The current goal is to run up a hill."
As a guideline, it is recommended that chunks have a small number of slots.

Millers number of 7 2 is recommended as an upper limit (Anderson & Lebiere,
1998), although this is not enforced.
- Procedural knowledge is the knowledge exercised in the performance of some

task (behavior). Human are not generally consciously aware of this knowledge, ie:
speaking a language, driving a car

In ACT-R, procedural knowledge is the set of production rules that the agent
currently knows. The set of production rules are if/then statements that specify how
a particular goal can be achieved when a specified pre-condition is met. A
production rule might state, for instance:
If I am hungry, then eat."
The human memory contains many declarative knowledge chunks and production
rules. At any point in time, when a person is trying to complete some task, a production
rule, indicating the next step to take in order to solve the problem, may fire" if the rule's
pre-condition, which is a conjunction of logical propositions that must hold true
according to the current state of declarative memory, is fulfilled. Whichever rule ends
up firing may result either in the goal being achieved or in the creation of new
knowledge chunks in working memory, which may then trigger more production rules,
and so on.
3.2.3 ACT-R modeling methodology: model variation, model space and

parameters for different domains and users
A scientifically useful cognitive model will be based in a particular theory of the
functional architecture that underlies the behavior itself. This fundamental architecture
must be able to support multiple phenomena, allowing for generalization to new
situations, users and domains (variation among models).
Model variation:
The overall ACT-R theory specifies the structures and the processes occurring in the
model, but the particular numerical values for a learning parameter or a reaction speed
may be left unspecified. The adjustment of these parameters affects the overall
performance, and a major part of the modeling process tends to involve adjusting these
parameters to find values for which the model behavior matches the empirical data.
Parameters can be adjusted to produce a certain range of outcomes, constrained by
the overall structure of the model.
ACT-R researchers have identified particular canonical parameter values that should
generally be used, and any deviation from these values should be seen as requiring
further explanation. For example, few ACT-R models have a production firing time
that is not exactly 50 ms. Other parameters, such as the latency and activation noise
levels tend to vary more widely between models.
The best approach is to identify ranges of parameter settings that produce models
that match to within a certain degree. That is, instead of reporting that the model
performs best if parameter X is set to 5, we instead indicate that the model performs
well if the parameter is between 2 and 6, for example.
In annex A, the list of the most relevant ACT-R parameters that can be modified can be
seen.
Exploring the model space:

In addition to parameter values, this approach can also be applied to modifications to

the underlying architecture itself (such as using slightly different structures or learning
rules or other implementation details). Our approach is to consider these types of
changes as qualitative, or nonnumeric, parameter changes.
The main issue with nonnumerical parameter changes is the absence of well-defined,
continuous parameter ranges.
Since we cannot, in general, expect to find mathematical proofs relating parameter

values to model behavior, we must run the model with each parameter setting of
interest. Even with only 5 parameters, if we are to consider only 5 different values for
each, we must perform 55, or 3125 simulation runs.
According to this way of viewing things, a theory specifies a set of models, each with
different parameter settings (numerical or non-numerical).
Non-numerical parameters:
It is also vital to observe that parameters need not be merely numerical values. There
is no reason why a theory might not treat an entire sub-module as a parameter. A
particular component within a model might be implemented in a number of qualitatively
different ways. We need to explore the architectural space of models as much as we
need to explore the parameter space.
3.2.4 Functionality of the overall architecture: Integrated Driving

Modeling Example
To better understand how a cognitive architecture works for modeling an individual
when performing a specific task, an analysis of an integrated model example of driving
has been performed.
Driving is a very common yet highly complex task that involves dynamic interleaving
and execution of multiple critical subtasks. In order to discern the power and limitations
of existing models, it is useful to view driving and driver modeling in the context of the
embodied cognition, task, and artifact framework. This framework emphasizes
three components of an integrated modeling effort: (1) the task that a person attempts
to perform, (2) the artifact by which the person performs the task, and (3) the
embodied cognition by which the person perceives, thinks, and acts in the world
through the artifact.
TASK: The task of driving is in fact an ever-changing set of basic tasks that must
be integrated and interleaved. Three classes of task processes for driving
identified:
Operational processes that involve manipulating control inputs for stable

driving.
Tactical processes that govern safe interactions with the environment and other
vehicles

Strategic processes for higher level reasoning and planning.
Driving involves these 3 types of processes working together to achieve stable

navigation: monitoring a traffic light, controlling the vehicle to stop and start, and
deciding to make a turn and controlling the vehicle through the turn. Some tasks
are not continual but intermittent, arising in specific situations for instance,
parking a vehicle at a final destination. In addition, driving may include secondary
tasks, perhaps related to the primary driving task (e.g., using a navigation device)
or perhaps mostly or entirely unrelated (e.g., tuning a radio or dialing a cellular
phone).
ARTIFACT: The artifact for driving is the vehicle itself and the interface between
the human and the vehicle. The interface should always include the steering wheel,
the accelerator (throttle) and brake pedals, and possibly the clutch pedal (on a
manual transmission); it also includes related controls such as turn signals,
headlights, and windshield wipers. For secondary tasks, the artifact may also
include any interface to the secondary device typically, knobs, buttons, and other
inputs, along with small displays and other outputs.
COGNITION: Embodied cognition is the integrated cognitive, perceptual, and motor

processes that manipulate the vehicle and execute the desired tasks. Driving
requires cognition during even routine driving, most obviously for higher level
decision making, more subtly for lower level vehicle control and situation
awareness. Between cognition and the vehicle lies the embodiment of the driver,
namely the perceptual processes (visual, aural, vestibular, etc.) and motor
processes (hands, feet) that provide the input from and output to the external world.
There can be parallelism in this integrated system for instance, moving the hand
while visually encoding the lead car but there are also capacity constraints and/or
bottlenecks.
The goal of integrated driver modeling is to rigorously address all three of these
components: handling as many driving-related tasks as possible, incorporating realistic
controls and vehicle dynamics, and performing the tasks through cognitive processes that
interact through realistic perceptual and motor processes.
In typical models of driving, drivers need to do the following tasks at once:
- Controlling the vehicle:
o Exercising lateral control (Steering)
o Exercising longitudinal control (Acceleration and braking)
- Monitoring the vehicles and objects around.
- Decision making
So the driver model will have the following core components: control, monitoring and
decision making modules. The three components are integrated to run in ACT-Rs

serial cognitive processor as a tight loop of small cognitive operations. The entire
model is implemented as an ACT-R production system including relevant procedural
and declarative knowledge. (See figure 7).
The core foundation of the integrated driver model is its skill knowledge (procedural
rules) specified as condition-action rules (If then rules). The model contains 45 rules
that fire serially: the various subtasks described in these rules require management to
interleave and execute. An example of these rules for controlling the vehicle is:
IF my current goal is to encode a distant perceptual point for steering
and there is a tangent point present (i.e., we are entering or in a curve)
THEN shift attention to this point and encode its position and distance
Figure 7: Integrated driver model
This system requires the visual system to update the locations of the artifact (the car)
and the motor system to translate the control information into motor commands. Once
given the control parameters, motor modules can make their adjustments
autonomously of the central production system. This type of information (location of the
lane, situation awareness, navigation) is factual knowledge that is stored as declarative
memory.

The use of ACT-Rs declarative memory for encoding the current environment provides
immediate predictions about potential driver errors: Because of memory decay
mechanisms built into the architecture, the chunks that encode vehicle position and
distance decay rapidly and may be forgotten when not refreshed often, thus providing
predictions (albeit yet-untested predictions) about mistaken estimation of surroundings
and potentially hazardous lane changes.
3.3 Cognitive and behavioural moderators: Overlays

As the tools, techniques, and implementations of human behavior representations have
begun to mature, increasing emphasis has been placed on ensuring that they
accurately reproduce the cognitive capacities, limitations, and processes of the human
participants they replace. One particular area of increased focus is the impact of
Cognitive and behavioural moderators (e.g., Gluck, Gunzelmann, Gratch, Hudlicka, &
Ritter, 2006). Cognitive moderators are factors that influence human cognition, but are
not necessarily part of the goal-directed thought that is the focus of information
processing.
An overlay is a technique for including a theory of how a behavioral moderator, such as

stress, emotions and fatigue, influences cognition across all models within a cognitive
architecture (Ritter and Reifers, 2006). An overlay is an adjustment or set of
adjustments to the parameters or mechanisms that influence all models implemented in
the architecture to reflect changes due to an altered mental state or due to long term
changes such as development (e.g., Jones, Ritter, & Wood, 2000). The main idea of an
overlay is to change the architecture in such a manner that the behavior of all models
developed under that architecture will be affected.
In many architectures, such as the one described here, ACT-R, there are a set of
mechanisms and a number of parameters that play a role in the models functioning; an
overlay will modify a combination of parameters and mechanisms to represent situation
specific but relatively long acting changes to information processing. For example, an
eyeglasses overlay would allow more inputs to be passed to the vision processor; a
caffeine overlay could increase processing speed by 3% and improve vigilance by 30%
(Ritter and Reifers, 2006).
This approach will keep the architecture consistent across tasks and populations, but
allowing that there may be differences in processing mechanisms and capabilities for
individuals or groups in certain contexts. The concept of an overlay provides a way to
describe these differences in a way that can be applied to all models in the
architecture.
Overlays are based on popular theories and studies about cognitive and behavioural
moderators that can be applied to cognitive architectures. The goal is to specify
mechanisms that can be implemented within the architecture or changes to current
mechanisms to simulate the effect of the psychological facets on embodied cognition.

These theories are typically not cast as additions to the knowledge necessary to
perform a task (it would mean creating a cognitive model), but are described as
changes to how people process information under a psychological state. Thus,
they make suggestions about process, about how the mechanisms of embodied
cognition change across all tasks under this facet.
Implementing them thus may implicate the necessity of modifying the architecture or
changing parameters of existing mechanisms. Unfortunately, many theories about
cognitive moderators are not specific enough to be directly implemented. In these
cases, the lessons serve as specification guidelines for new mechanisms.
One of the earliest theories of behavioral moderator modeling, which models a stressor
as a moderator, is the classic Yerkes-Dodson inverted u which shows that as a stimuli
or moderator is increased, performance is initially low, then rises, and then falls off
again after a threshold is passed. Thus some degree of presence of various stressors
is better than none (e.g., performance can be better in slightly chaotic, moderately time
-pressured settings). A research (Silverman et al, 2001), assembles a set of
performance moderators functions (PMFs) concerning stressors and their relationship
with emotions, motivations and fatigue. A finding of the Silverman research is that there
is no validated model of how the many diverse stimuli and PMFs interact and cross-
connect to effect stress and behavior, although there are a number of efforts that
attempt useful integrative models.
A theoretically inspired approach was taken by Hendy and Farrell (Hendy et al 1997)
who adopts an information processing theory and model that attempts to explain the
effects of time pressure, task difficulty, mental capacity, fatigue, motivation, anxiety,
and the like. This model claims that time pressure is the primary determinant of
operator workload, performance and errors, but as yet, they have little empirical results
to back up their claims.
At a more macro-behavioral level, another approach was taken by Janis and Mann
(Irving, L. J., Mann, L. 1977). This includes a meta-analysis of the literature plus an
account of many of their own numerous studies of stress and decision making under
conflict. Janis and Mann provide a taxonomy of decision strategies for coping under
stress, time pressure, and risk. This taxonomy is described on the steps of the U-curve
of Figure 8, defined below. The taxonomy includes an algorithm that indicates how
stress, time pressure, and risk drive the decision one coping strategy to another (X-axis
of Figure 8).

Figure 8: Decision Making under stress
In particular, they use the algorithm to derive the values of the so-called integrated
stress, the iSTRESS variable:
iSTRESS 1 unconflicted adherence, select mt = mt-1
2 unconflicted change to next Sm (next state)
3 m = 1, M whichever is Best Reply (vigilance)
4 near panic so, m = 1 of 1, M (if highly experienced users:

Recognition Primed Decision making. Also, defensive avoidance occurs at
this level for non experts).
> 4 panic, so Sm = avoid the stressor (hyper-vigilance)
where, according to Janis and Mann, the five coping patterns are defined:
1) Unconflicted adherence (UA) in which the risk information is ignored and the
decision maker (DM) complacently decides to continue whatever he has been doing.
2) Unconflicted change (UC) to a new course of action, where the DM uncritically

adopts whichever new course of action is most salient, obvious, or strongly
recommended.
3) Vigilance (VG) in which the DM searches painstakingly for relevant information,

assimilates it in an unbiased manner, and appraises alternatives carefully before
making a choice.

4) Defensive avoidance (DA) in which the DM evades the conflict by procrastinating,

shifting responsibility to someone else, or constructing wishful rationalizations and
remaining selectively inattentive to corrective information.
5) Hypervigilance(HY) wherein the DM searches frantically for a way out of the

dilemma and impulsively seizes upon a hastily contrived solution that seems to promise
immediate relief, overlooking the full range of consequences of his choice because of
emotional excitement, repetitive thinking, and cognitive constriction (manifested by
reduction in immediate memory Span and simplistic ideas). In its most extreme form,
hypervigilance is referred to as "panic."
Unfortunately, this model does not provide either precise threshold values i that
indicate when a user trigger a change in coping style, or any insight into how to
integrate the many diverse stimuli, factors, or PMFs that determine stress and time
pressure or risk. These values should be therefore extracted from psychological
studies and experiences and then modeled taking into account relevant information.
3.4 Cognitive Architecture & Cognitive and Behavioural

Moderators: suitable approach for VERITAS?
The aim of this work is to select the best approach for modelling the Psychological
and Behavioural Abstract User Models of VERITAS users: elderly and disables. As
we have seen in the following sections, psychological facets can be described as
moderators of the cognitive processes, and, in order to integrate these facets into a
cognitive architecture, modifications of the parameters and/or the intrinsic
mechanisms (architecture space) need to be performed. The values of these
parameters and the space variation should be consistent with psychological
theories and experiences.
The Abstract User Model will be the basis of the Virtual User Model, in which the
affected tasks will be also described.
VERITAS will likely support three simulation approaches (For more information see
Deliverable D2.1.1 Core Simulation platform):
Low-level constraints satisfaction: based on constraints that are related to the

functionality of the users. Eg: The simulation fails for a user with arthritis on the
right hand because the fist cannot close in order to grasp.
Semi-dynamic simulation, performed as a sequence of predefined tasks

(which at the same time are a sequence of primitive tasks) that have specific
completion parameters. At each simulation time-step all the constraints are
checked dynamically. Collisions and dynamic virtual user response are not
taken into account.
Dynamic simulation: Simulation is performed as a sequence of goals that

correspond to specific simple tasks. The steps to perform a task are based on

physical simulation, the virtual scene is navigated by the virtual user using
computer vision methods (like those used in autonomous robot humanoids)
while motion is performed using forces and torques. At each simulation time-
step all the constraints are checked dynamically and the environment is
checked for interaction with the virtual user.
This is another relevant issue that should be noticed when selecting the best
approach for VERITAS, and its related with the description of what a cognitive
architecture is: a theory for understanding and simulating how human mind works.
So, in the case of ACT-R, it is a cognitive architecture that represents the human
mind processes, but at the same time, it provides a computational simulation
framework for simulating how different users perform a specific task. (See example
in section 2.2). This means that the cognitive architecture selected should be
consistent with VERITAS simulation approach.
Following, an analysis of the pros and cons of using ACT-R (or analogous cognitive
architecture and the cognitive and behavioural moderators has been performed.
3.4.1 Advantages
1. ACT-R is adept at integrating lower-level perceptual and motor control
processes with higher-level cognitive and decision-making processes. It relies
on a central cognitive processor to maintain a mental model and make
decisions about higher-level strategies to execute.
2. Any model developed in ACT-R necessarily needs to bear cognitive and

perceptual-motor aspects of behavior. For example, in the case of the driving
model, this feature facilitates the integration of lower-level control models, such
as for lateral or longitudinal control, with higher-level decision making models
that act upon the current situation and the current needs and desires of the
driver.
3. It incorporates built-in mechanisms to model common psychophysical

phenomena. It has a mechanism for modeling how people store, recall, and
sometimes forget information; how people execute manual motor control for
typing and other tasks; and how people move their eyes to visual targets and
encode the information therein. Each of these mechanisms has typically been
rigorously analyzed and tested in empirical studies that emphasize basic
psychophysical tasks.
4. It allows the creation of behavioral models for complex domains, such as

driving.
5. ACT-R places certain limitations and constraints on models that mimic the
constraints of the human system cognition (e.g., memory and forgetting effects,
errors, attention allocation) and of visual-motor performance (e.g., manual
control actions, eye movements).
One of the most important constraints for complex models is that although for
instance perceptual and motor processes can run in parallel with cognition, the

cognitive processor itself is serial and, in essence, can think only one thing at
a time.
6. ACT-R provides a rigorous framework and software for simulating (building,

running, and testing) computational models of behavior.
7. ACT-R has the ability to perform some processes in parallel such that, for
example, the perceptual module can look at a new item while the motor module
performs a physical movement. This allows ACT-R models to run in simulation
often in real time, or at least generating data on the same time scale as that of
human participants and predict exactly those data and measures that are
collected from drivers (steering wheel, throttle, and brake positions; turn signals;
eye movements; etc.). The prediction of real-world measures greatly facilitates
model validation through direct comparison with human data.
8. ACT-R allows individual variability: allows for several types of variability ranging
from simple errors to those stemming from statistical variations in attention
management to variations in higher-level cognitive strategies. Cognitive and
Behavioural Moderators (overlays) can be used to tune the parameters that
account for such variability to represent different classes or types of users:
a. ACT-R allows tuning of attention and attention capacity: the amount of

information users can process through visual attention is limited and
requires movement of attention between salient regions of the view.
Architectures help to describe how much information can be processed,
how quickly it is processed, and how this processing may interfere with
other driving tasks. In addition, known overlays can be used to model a
modification of the attention capacity.
b. ACT-R allows tuning working memory load: architectures typically

include limits on the amount of information that can be mentally stored
and processed during a given time period. Such limits help to account
for cognitive limitations in dealing with heavy multitasking with multiple
tasks or instruments (e.g., a cell phone or navigation device). Overlays
can be used to model a decrease/ increase of the working memory load.
c. ACT-R allows tuning workload: architectures typically have some

interpretation of multitasking workload for their particular parameters
and constraints, though they differ greatly on these interpretations.
Nevertheless, architectures could allow a given driver model to predict
effects on mental and physical workload given different settings of these
parameters for different individuals and situations. ). Also, overlays or
cognitive moderators can be used to model a decrease/ increase of the
workload.

3.4.2 Disadvantages
1. Simulation in ACT-R is cognitive based. Its cognition who triggers the motor tasks
and not in a fixed sequential way. Two ACT-R simulations with the same input
parameters do not always produce the same output, being the motor response a
consequence of cognition. By contrast, simulation in VERITAS is more focused
on physical capabilities and constraints. In addition, the simulation models will
define a sequence of tasks to be performed by VERITAS user, so both approaches
differ.
2. Currently in ACT-R, the only motor response well developed is manual; other
movements are not well (or not at all) developed, so it wont include all the range
needed in VERITAS.
3. Regarding perception, currently in ACT-R only the visual perception module is well
developed. The other perception systems are limited (aural) or not at all developed.
4. ACT-R available models (with quantitative metrics) of VERITAS users (elderly and
disabled) are almost inexistent.
5. Except for the driving model, there are very few available ACT-R models of specific
VERITAS tasks and domains.
6. Overlays in ACT-R are practically new. There are very few models that have been
tested with information validated in psychological tests.
7. If experiences and models are not available, we should perform tests with real
users doing the intended tasks, and tuning afterwards the parameters that give as
output the same simulation results. Since many tasks and users are involved, this is
not at all a trivial task to do and requires great a great amount of efforts.
8. Simulating cognition in ACT-R implies developing an integrated simulation model in

ACT-R, where not only the cognitive architecture with the tuned parameters need to
be provided, put also a TASK model in ACT-R (implicit knowledge of how user
performs a task, really closed to VERITAS simulation models) and the interaction
with the object/artifact the user is interacting with.
9. Writing Tasks models in ACT-R means writing all the chunks and production rules
about the specific task (ie: driving). But not only this, it implies knowing also the
involved cognition processes (cognition loop) when a user is doing a specific task
of a specific domain (involving application scenarios knowledge, simulation
knowledge and task knowledge).
10. ACT-R, as simulation framework, should be part of VERITAS simulation core. The
simulation models, that deal with the sequence of tasks that virtual users perform
during the simulation, should be input of the ACT-R core.
11. In order to fit better with VERITAS approach, which is more physically focused,
extra perception and motor modules in ACT-R should be further developed.

3.4.3 Analysis and recommendations of psychological and behavioural

modeling for VERITAS
Despite the fact that research shows that a cognitive architecture and specifically ACT-
R with overlays is probably the approach best suited for modeling cognitive processes
and the influence that psychological facets have over them, the issues before
described show that adopting ACT-R as the complete solution for cognitive models and
simulation may have a great impact on the whole project.
However, and taking into account the simulation approach followed in VERITAS, based
on constraints, ACT-R may be useful for obtaining information relative to cognitive and
behavioural parameters that can be used in this approach.
Moreover, it has to be further explored if ACT-R is indicated for modeling at a primitive

task level in an individual way.
A further analysis of the suitability of cognitive architectures, and more specifically

ACT-R as core of the simulation framework, has to be performed in order to select the
final approach for modeling.
Regarding the psychological and behavioural states of a user, the state of art shows
that the best solution are the moderators of cognition and behavior, using an
overlay technique that maps the moderators into the architectural parameters of the
cognitive models.
An analysis of the state of art of behavioural and cognitive moderators of cognitive

architectures should be performed and included as part of the P&B Abstract User
Model. Moreover, an analysis of the psychological studies regarding different
psychological states (stress, fatigue, etc) that influence on VERITAS users will be also
performed. (See chapter4). The parameters extracted from this analysis will be
mapped as overlays in the cognitive architecture.

4 Psychological facets: stress, fatigue, emotions

and motivation as cognitive and behavioural
moderators
VERITAS focuses on challenges to enhance the behavioural realism of computer
generated individuals and attempts to reflect the state of the art in human behaviour of
VERITAS users, that is to say, elderly and disabled, by creating realistic models of this
behaviour.
Elderly and disabled people face multiple life and body changes that lead them to
experiment different psychological facets during their life. Late life can be a time of
multiple losses. The older person has experienced change and loss through retirement,
disability and death of friends and family as well as change in financial, social, and
physical health status. These changes may lead to different psychological states,
experimenting negative emotions such as depression, physical and emotional fatigue,
stress and de-motivation that can lead elderly towards depression. Facts about
psychological states (Healthplace, 2010) in elderly are:
Late-life depression affects about 6 million Americans age 65 and older.
Clinical depression can be triggered by long-term illnesses that are common in

later life, such as diabetes, stroke, heart disease, cancer, chronic lung disease,
Alzheimer's disease, Parkinson's disease, and arthritis.
Older patients with significant symptoms of depression have roughly 50%

higher healthcare costs than non-depressed seniors. (The direct and indirect
cost of depression in all ages is estimated to be nearly $44 billion a year.)
Severely stressful events in the elderly have been demonstrated to trigger a

cognitive decline (European Society for Traumatic Stress Studies, December
2010).
Disability is one of the main causes of depression, especially among accident victims,
senior citizens, and children born with disability. 2 out of 10 people with a disability
suffer from depression (E-zine Articles, December 2010). Often, disability occurs when
it is least expected, and the sudden shock of the trauma affects negatively people both
psychologically and physiologically, experiencing negative feelings and lack of
motivation towards their life.
These psychological states can have great impact on the quality of life and affect the
overall performance of these groups when performing a task.
In VERITAS, we should focus on how different psychological facets affect VERITAS

users, elderly and disabled groups. As we have seen, stress, fatigue, emotions and
(de)motivation are amongst the most relevant P&B facets that influence elderly and
disabled, so within VERITAS we will pay special attention to model these psychological
states.

In the following sections, an analysis of the existent models and studies of stress,
fatigue, emotions and motivation will be carried out. This analysis will focus on the
theoretical models of the facets, which will provide a deep understanding of the
psychological dimensions of each state. Then an analysis of the state of art in
computational models for VERITAS users, domains and measurement methodologies
will be performed.
4.1 Stress as cognitive and behavioural moderator

4.1.1 Definitions of Stress
Stress is the adaptive response of the body when facing various stressors. (Hans
Selye, 1936). Stress is a normal reaction in the lives of people of any age. It is
produced by the bodys instinct as a protection from physical or emotional pressure or,
in extreme situations, from danger. Alternatively, to clarify concepts, the term "stress
response" is used to refer to the nonspecific body response to any demand, and the
term "stressor" or "stressful situation" to the stimulus or situation that causes stress
response.
Figure 9: Stimuli / Response
Currently interactionist stress theories have gained popularity, which promote that
stress response is the result of interaction between the characteristics of the person
and the demands of the environment. A person is considered to be in a stressful
situation or under a stressor when it has to deal with situations involving demands that
are difficult to put into practice or to meet. Stress occurs when the demands of life are
perceived to be too difficult. In other words, it depends both on the demands of the
environment and his/her own resources (Lazarus and Folkman, 1984), or going a little
bit further, the discrepancies between the demands of the external or internal
environment, and the manner in which the person perceives that he/she can respond to
these demands (Lazarus and Folkman, 1984). So, if an imbalance between the
resources and strategies available to handle the situation and those necessary
to achieve the goals occurs, then is the moment to start talking about the stress
phenomenon. It is only in such conditions where different types of reactions appear
(emotional, cognitive, physiological and behavioural) that we could call them
reactions/responses of stress and to the condition that generated the process, a
situation of stress or stressor (Gmez, 2000).
Stress response is an automatic response of the body to any external or internal

environmental change by which it prepares to deal with possible demands that are

generated as a result of the new situation, (1992 Labrador). Since a great amount of
resources are activated (including an increase in the level of physiological, cognitive
and behavioral activation); it involves a significant wear to the body. If this is episodic,
the body will have the ability to recover, but if repeated too often in frequency, intensity
or duration, it can produce the appearance of chronic psycho-physiologic disorders.
With stress, the body responds to external conditions that disrupt the emotional
balance of the person. The physiological result of this process is a desire to flee from
the situation that causes stress or confront it violently. This reaction involves
nearly all the organs and functions of the body, including brain, nerves, heart, blood
flow, the hormone level, digestion and muscle function.
The problem with these types of definitions is that when conducting studies we cannot
assume so easily, especially in the case of human research, that exposure to a
particular event will end up being a stressor, or stressful for everyone involved and so
may not therefore generate the same physiological responses in each individual.
(Callister, Suwarno & Seals, 1992; Pack & Palkovits, 2001); the presence of
individual variables such as coping styles, biological vulnerability, perception of
control or lack of control (as examples) modulate the development of stress in
each individual (Gmez, 2000).
In fact, the definition of stress is a cloudy description of various physiological and

behavioural responses. A definition of the phenomenon should take into account the
following aspects:
The fact that stress occurs does not depend on the physical parameters of
environmental stimulation, but on the way in which the body perceives and
reacts to this stimulus.
There is no physiological status that is specific to stress. For example, it is

accepted that a rise in the level of glucocorticoids is an indicator of a state of
stress, but they can also rise in pleasurable activities such as eating food,
exercising, or sex.
The stressor controllability has a profound influence on the impact of an

aversive experience in the physiology and behaviour.
4.1.2 Types of stress

Beyond the possible definitions of stress, Humber (1986) proposes three points of view
from which stress can be considered:
1. Stress as a situation, postulating that all the stimuli in the environment can act on
the individual as stressful stimuli (aka known as stressors).
2. Stress as a short term reaction (episodic event) consists in muscle reactions,

physiological disorders and feelings experienced in a subjective way that appear as a
result of provocative stress stimuli (aka short term stress).
3. Stress as a long term reaction (chronic or sustained event), which consists of

alterations to the physical or psychological performance of a person, as a result of

having been under the influence of stress during an extended period of time (long term
stress). (Humber, 1986).
Depending on whether the effect on the human body is positive or negative, we can
also find the following classification of stress:
Eustress: Positive type of stress that you experience right before you have the
need to exert physical force. Eustress prepares the muscles, heart, and mind for
the strength needed for whatever is about to occur. Eustress can also apply to
creative endeavors. When a person needs to have some extra energy or creativity,
eustress kicks in to bring them the inspiration they need. An athlete will experience
the strength coming form eustress right before they play a big game or enter a big
competition. Because of the eustress, they immediately receive the strength that
they need to perform. Eustress prepares the body to fight with or flee from an
imposing danger. This type of stress will cause the blood to pump to the major
muscle groups, and will increase the heart rate and blood pressure to increase. If
the event or danger passes, the body will eventually return to its normal state.
Eustress motivates, focuses energy, is short-term, is perceived as within our coping
abilities, feels exciting and may improve the performance.
Distress: Negative type of stress that the mind and body undergoes when the
normal routine is constantly adjusted and altered. The mind is not comfortable with
this routine, and craves the familiarity of a common routine. There are actually two
types of distress: acute stress and chronic stress.
o Acute Stress: Acute stress is experienced in response to an immediate

perceived threat, either physical, emotional or psychological. The threat can
be real or imagined; its the perception of threat that triggers the response.
During an acute stress response, the autonomic nervous system is
activated and the body experiences increased levels of cortisol, adrenalin
and other hormones that produce an increased heart rate, quickened
breathing rate, and higher blood pressure. Blood is shunted from the
extremities to the big muscles, preparing the body to fight or run away. This
is also known as the fight-or-flight response. The response is recognized
as the first stage of a general adaptation syndrome that regulates stress
responses among vertebrates and other organisms. If a stimulus is
perceived as a threat, a more intense and prolonged discharge of the locus
ceruleus activates the sympathetic division of the autonomic nervous
system (Thase & Howland, 1995). The activation of the sympathetic
nervous system leads to the release of norepinephrine from nerve endings
acting on the heart, blood vessels, respiratory centers, and other sites. The
ensuing physiological changes constitute a major part of the acute stress
response. The other major player in the acute stress response is the
hypothalamic-pituitary-adrenal axis.
o Chronic Stress: Chronic stress will occur if there is a constant change of

routine for week after week. Chronic stress affects the body for a long
period of time. It involves an endocrine system response in which occurs a
release of corticosteroids. If this continues for a long time, it can cause

damage to an individual's physical and mental health. This is the type of

stress experienced by someone who constantly faces moves or job
changes.
Stressors are stimulus that attacks us in an emotional or physical way. If the

danger is real or perceived as such, the result is the same. It usually causes
tension, anxiety, and various physiological reactions. The physiological,
psychological and behavioral stress responses of a subject seek to adapt and
adjust to both internal and external pressures.
4.1.3 Causes of stress: stressors or stress stimuli
A distinction can be made between those stressors which are physical and those which
are mental.
A physical stressor is one which has a direct effect on the body. This may be an
external environmental condition or the internal physical/physiologic demands of
the human body.
A mental stressor is one in which only information reaches the brain with no direct
physical impact on the body. This information may place demands on either the
cognitive systems (thought processes) or the emotional system (feeling responses,
such as anger or fear) in the brain. Often, reactions are evoked from both the
cognitive and the emotional systems.
However, it is difficult to try to generalize stressors, since the same situation may
not cause the same responses in each individual (Callister, Suwarno & Seals,
1992; Pack & Palkovits, 2001).
A distribution of the mentioned stressors can be seen in the following table:
PHYSICAL STRESSORS MENTAL STRESSORS

Environmental Cognitive
Weather conditions: heat, cold, Information: too much, too many,

wetness not understandable
Noises Sensory overload versus
Hypoxia, fumes, chemicals, pollution deprivation
Infectious diseases Isolation, ambiguity, uncertainty
Physical work Time pressure versus waiting
Non-accessible environments Unpredictability
Poor housing Rules of engagement, difficult
Traffic jams judgments
Work environment Hard choices versus no choices
Recognition of impaired
functioning of the body.
Physiological Emotional
Sleep debt Fear- and anxiety- producing

Dehydration threats or injury, disease, pain,
Malnutrition, poor hygiene failure, loss

PHYSICAL STRESSORS MENTAL STRESSORS

Fatigue Grief-producing losses.
Malfunctioning immune system Resentment, anger- and rage-
Over-, under- use of muscles producing frustration, threat, loss
(excess, too little exercise), organ and guilt.
systems Boredom producing inactivity
Illness or injury Conflicting motives (worries,
Drug/alcohol misuse divided loyalties)
Excess caffeine Spiritual confrontation or
Chronic hyperventilation temptation.
Smoking Interpersonal feelings
Hypoglycaemia Unrealistic beliefs and
Surgery expectations
Chronic pain Pessimism
Lack of relaxation Excessive worrying
Health worries
Unhappy childhood
Unemployment
Financial problems
Perfectionism
Loneliness
Low self esteem and levels of
assertion
People pleasing
Personality
Rigid thinking style
Excessive self criticism
Exams, giving talks/presentation
Table 4 Distribution of stressors
4.1.4 Physiological, cognitive and behavioural stress responses

Stressors induce physiological, cognitive and behavioural/emotional reactions or
responses. These stress responses of a subject seek to adapt and adjust to the
physical and mental pressures/stimuli. For instance, physiological responses affect
brain, nerves, heart, blood flow, hormone levels, digestion, muscle function, etc.
Cognitive responses affect concentration, memory, attention, etc, while
psychological reactions may be tension, anxiety, etc.
A vast literature states that when stress becomes excessively chronic or intense in
amplitude, target-organ (the organ affected by the stress response) disease and/or
dysfunction will result (Everly, 1986; Selye, 1956; Harris, 1991; Stoudemire, 1995).
G. Everly and J. Lating define stress (G. S. Everly & J.M. Lating, 2002) as a
physiological response that serves as a mechanism of mediation linking any given
stressor to its target-organ effect or arousal.

Stressor Stress Response Target organ signs/ Symptoms
( stimulus ) ( physiological mechanism of mediation) (pathological effect)
Figure 10: Stress response (G. S. Everly & J.M. Lating, 2002)
The stress response, as a physiological mechanism of mediation, can be

characterized by a diverse constellation of physiological mechanisms (Cannon,
1914; Makara, Palkovits, & Szentagothal, 1980; Mason, 1972; Selye, 1976) that
may be categorized as (1) neurological response pathways, (2) neuroendocrine
response mechanisms, and (3) endocrine response pathways.
Selye (1956, 1976) has argued for the nonspecificity of the stress response. Other
authors (Everly, 1972; Humphrey & Everly, 1980; Mason, 1971; Harris, 1991;
Mason et al., 1976) have argued that the psychophysiology of stress may be highly
specific with various stressors and various individuals showing different degrees of
stimulus or response specificity, respectively.
4.1.4.1 Epiphenomenological model of the stress response

G. Everly and J. Lating describe human stress within the context of the dynamic
process it represents. This process may then be delineated from a systems
perspective, that is, one of interrelated multidimensionality. Figure X details a
systems perspective brought to bear upon the phenomenology of the human stress
response. The key elements of the system are:
1. Stressor events
2. Cognitive appraisal and affective integration: Cognitive appraisal refers to the

process of cognitive interpretation, that is, the meanings that we assign to the
world as it unfolds before us. Appraisal is a function of any existing biological
predispositions, personality patterns, learning history, and available coping
resources. Affective integration refers to the blending and coloring of felt
emotion into the cognitive interpretation. The resultant cognitiveaffective
complex represents how the stressors are ultimately perceived. It should be
noted that there exists substantial evidence supporting the cognitive primacy
hypothesis that is, cognition determines affect (felt emotion) and thus assumes
a superordinate role in the process of restructuring human behavior
patterns. Influential authors such as Arnold (1970, 1984), Cassel (1974),
Lazarus (1966, 1982, 1991), Meichenbaum (1985; Meichenbaum & Jaremko,
1983), and Selye (1976) strongly support the cognitive primacy position as it
relates to human stress.
3. Neurological triggering mechanisms (e.g., locus ceruleus, limbic nuclei,

hypothalamic nuclei): Impulses project back to the highly sensitive emotional
anatomy in the limbic system (Arnold, 1984; Cullinan et al., 1995; Gellhorn &
Loufbourrow, 1963; Gevarter, 1978; Nauta, 1979), especially the hippocampus
(Reiman et al., 1986).
4. The stress response (a physiological mechanism of mediation): In most

individuals, activation of the limbic centers for emotional arousal leads to

expression of the felt emotion in the form of visceral activation and

neuromuscular activity. Such visceral and neuromuscular activation represents
the multiaxial physiological mechanisms of mediation: the stress response. It
possesses at least three major efferent axes: neural, neuroendocrine (also
called the fight-or-flight response axis), and endocrine, and represents
numerous combinations and permutations of efferent activity directed toward
numerous and diverse target organs, amongst them the brain.
The effects of neural activation via the sympathetic system are those of
generalized arousal within the end organs. The effects of activation via the
parasympathetic system are inhibition, slowing, and restorative functions.
Skeletal muscular is also a prime target for immediate activation during stress
and emotional arousal. Such activation, if excessive, may lead to a host of
neuromuscular dysfunctions.
Due to the neural activation, when a person is exposed to a 'dangerous or
stressful' situation, his Internal Homeostasis gets unbalanced and the body
fights to restore it, to escape or confront the peril, so the body needs more
energy for that. The extra energy is got by the 'Fight or Flight' response.
Basically, the mind feels the danger and the body tackles it. These actions are
coordinated by hormones. Typical physiological responses are: diversion of the
blood from less vital to more vital organs, increase of the strength of skeletal
muscles, increase in the heart rate to supply more blood quickly, etc
The most chronic and prolonged somatic responses to stress are the result
of the endocrine axes (Mason, 1968b). Typical responses are: exacerbation of
gastric irritation, suppression of immune mechanisms, appetite suppression,
feelings of depression, hopelessness, etc.
The General Adaptation Syndrome: As a means of integrating his

psychoendocrinological research, Hans Seyle (1956) proposed an integrative
model for the stress response, known as the General Adaptation Syndrome
(GAS). The GAS is a triphasic phenomenon. The first phase Selye refers to as
the alarm phase, representing a generalized somatic shock, or call to arms
of the bodys defense mechanisms. The second phase is called the stage of
resistance, in which there is a dramatic reduction in most alarm stage
processes and the body fights to reestablish and maintain homeostasis. Stages
1 and 2 can be repeated throughout ones life. Should the stressor persist,
however, eventually the adaptive energy, that is, the adaptive mechanisms in
the second stage may become depleted. At this point, the body enters the third
and final stage, the stage of exhaustion, which, when applied to a target
organ, is indicative of the exhaustion of that organ, and the symptoms of
disease and dysfunction become manifest. When the final stage is applied to
the entire body, life itself may be in jeopardy. The three stages of the GAS are
detailed in Table 2.6.
Regarding to the classification of stress, the alarm phase would be connected

with the eustress, the resistance stage with the acute stress, and the
exhaustion stage with chronic stress.

5. Target-organ activation: the most immediate response to a stressful stimulus

occurs via the direct neural innervations of end organs. The term refers to the
phenomenon in which the neural, neuroendocrine, and endocrine constituents
of the stress response just activate, increase or inhibit normal activation, or
catabolize some organ system in the human body. Target - organ activation and
subsequent signs and symptoms of disease may affect the patients cognitive
affective behavior and, therefore, further neurological triggering and continued
stress-response activity.
6. Coping behavior: the manifestations of human stress are highly varied and
individualistic. Whereas biological predisposition certainly plays a role in this
process, a major factor in determining the impact of stress on the patient is his
or her perceived ability to cope. Coping is defined as efforts, both action-
oriented and intrapsychic, to manage (that is, master, tolerate, reduce,
minimize) environmental and internal demands, and conflicts among them,
which tax or exceed a persons resources. Coping can occur prior to a stressful
confrontation, in which case it is called anticipatory coping, as well as in
reaction to a present or past confrontation with harm. (Cohen & Lazarus, 1979,
p. 219). Coping may be thought of as environmental or cognitive tactics
designed to attenuate the stress response.

Figure 11: Epiphenomenological model of the stress response

4.1.5 Relationship of Stress to Task Performance

Stress is an internal process which presumably evolves because it helps the individual
to function better, stay alive, and cope successfully with stressors. However, there is an
optimal range of arousal (or motivation or stress) for any given task.
a. If there is too little arousal, the job is done haphazardly or not at all because the
individual is easily distracted, makes errors of omission, or falls asleep. If arousal
(stress) becomes too intense, the individual may be too distractible or too focused on
one aspect of the task. He may have difficulty with fine motor coordination and with
discriminating when and how to act. If the individual is unfamiliar with his own stress
reflexes and perceives them as dangerous (or incapacitating, or as a threat to self-
esteem), the stress itself can become a stressor and magnify itself.
b. With extreme arousal, the individual may freeze (become immobile or petrified by
fear). Alternately, he may become agitated and flee in disoriented panic. If stress
persists too long, it can cause physical and mental illnesses. Extreme stress with
hopelessness can even result in rapid death, either due to sympathetic nervous system
overstimulation (such as stroke or heart attack) or due to sympathetic nervous system
shutdown (not simply exhaustion). An individual giving up can literally stop the heart
from beating.
Figure 12: The human function curve.

This optimal range of stress differs from task to task. Tasks which require heavy but
gross muscular exertion are performed best at high levels of arousal (Figure 11). Tasks
that require fine muscle coordination and clear thinking or that require inhibiting action
will be disrupted unless the stress process is kept finely tuned. If the stress process
allows too much or too little arousal or if arousal does not lessen when it is no longer
needed, stress has become harmful.
The Yerkes-Dodson law demonstrates an empirical relationship between arousal

(pressure) and performance (ability to cope) (Figure 12). The law depicts the
importance of avoiding extremes of too little or too much workload.
Figure 13: Coping / Pressure relation. (Haarmann, accessed in 2010)
4.1.6 Measurement of human stress responses
4.1.6.1 Stressor Scales

The most widely used measurement tool for the assessment of human stress, in reality,
does not measure stress at allit measures stressors.
SCALE METHODOLOGY SCORES
Social Readjustment The SRRS contains 43 items consisting of commonly Total LCU score:
Rating Scale (SRRS), experienced life events. Each life event is weighted
Thomas Holmes, with a life change unit score (LCU). Respondents are Score of 300+: At
risk of illness.
Richard Rahe (1967) simply asked to check each of the items they have
experienced within the last 12 months. The arithmetic Score of 150-299+:
summation of LCUs represents the total LCU score, Risk of illness is
which can then be converted to a relative health risk moderate (reduced
statement, that is, the risk of becoming ill within a by 30% from the

stipulated time period. The association between high above risk).

LCU scores and risk of subsequent illness is assumed
to be a function of the fact that organisms must adapt to Score 150-: Only
have a slight risk of
novel stimuli and otherwise new life events. Ie: death of
illness.
a spouse: 100 LCU.
Life Experiences Survey Is concerned with the views on the levels of life Life event= [-3,3]=
(LES), (Sarason et al., satisfaction along with current daily habits and views of [extremely negative,
1978) current social issues. Lists a series of life events but extremely positive];
also inquires into the desirability of each of the events.
The original instrument includes 60 items divided into Event not
two sections. Section 1 contains 50 life changes that occurred=0;
are common to individuals in a wide variety of situations
(e.g., In the last year, did you get married?). Section 2 Total Score=
contains 10 items domain related. Most of the items recent_life_events.
were based on existing life stress measures (SRRS).
Every event that occurred is coded as one life change
unit. These units can then be summed for a total score
of recent life events.
Hassles Scale (Kanner Lists a series of minor daily hassles, that is, sources of Score of 136+: Very
et al., 1981) frustration that commonly recur to many individuals. High Level of
The scale also includes an uplifts assessment that Stress,
theoretically serves to mitigate the adverse impact of
Score of 116-135:
negative life events. High Stress,
Score 76-115:
Average stress,
Score 56-75: Low

stress,
Score 51-55: Very

low stress
Stressful Life This instrument consists of 20 items that inquire as to

Experiences Screening (1) the presence of a stressful life experience, and (2)
(SLES; Stamm, 1996). the degree of stressfulness of that experience.
Life Stressor Checklist Consists of 30 events that satisfy the DSM-IV

Revised (Wolfe & definition of traumatic. The scale is not only an indicator
Kimerling, 1998) of traumatic events, but it also serves as an
assessment of the events current impact upon the
individual or endorsed events, respondents are asked
to provide: age when event began, age when event
ended, belief that they were in harm (yes or no),
feelings of helplessness (yes or no). In addition,
affect on life and how upsetting event was at the time
are rated on a 5-point intensity scale (1=not at all; to
5=extremely). Respondents are asked to identify the 3
events that currently have the greatest impact on them.
The LSC-R can be used for clinical or research
purposes.
Survey of Recent Life Self-report measure of exposure to daily hassles that Scores ranges
Experiences(SRLE; has been developed as an alternative to the Daily between 41 (very
Kohn & Macdonald, Hassles Scale (Kanner, Coyne, Schaefer, & Lazarus, low level stress)

1992) 1981). The intensity of experiences is ranged between and 164 (high level
1 (not at all part of your life) and 4 (very much part of Stress)
your life).
Subjective Stress Scale Provides a sensitive measure of stress for people with For each item, the
(SSS) for Mildly intellectual disabilities. Subjects are asked to response categories
Intellectually acknowledge if any of the stressful events listed in the are: Event not
Handicapped (Paul scale had occurred in the last fortnight (Yes/No), experienced, no
Bramston and Gerard J. weighting the yes-answers with a 4-point Likert scale. stressful event,
Fogarty, 1995) some stress,
moderate stress, a
great deal of stress.
Table 5 Measurement of stress response
4.1.6.2 Cognitive-affective correlate scales

The cognitive-affective correlate scales assess the cognitiveaffective domain of
respondents as an indirect assessment of the human stress response (Everly &
Sobelman,1987).
SCALES METHODOLOGY SCORES
Derogatis Stress 77-item self-report inventory derived from interactional stress multidimensional
Scale: DSP theory. It purports to assess not only emotional responses measurement:
(Derogatis, 1980) and personality mediators, but also environmental events.
Patients rate each of the items on a 5-point Likert scale (0 = TSS=n t-scoresn
not at all true of me, 4 = extremely true of me) regarding the
extent that the statement is typically true of them. The DSP SSS scores:
consists of 11 subscales for which t-score equivalents are
SSS= 0: no stress
available. A global score, called the Total Stress Score (TSS)
and derived by summing all 11 of the subscale t-scores, also SSS 42 low
has t-score equivalents available. A second global score, subjective stress
called the Subjective Stress Score (SSS), derives from an
analog 100~mm line marked by respondents to indicate their 43 SSS 55
perceived levels of current stress. The DSP permits moderate
subjective stress
formation of three groups: accurate stress appraisers, stress
deniers, and stress augmenters, as described in the test SSS >55 high
manual. subjective stress
SSS=100:
extremely highly
stressed
World Assumption Assesses three core assumptions, or beliefs, about life: the
Scale (WAS; benevolence of the world, the inherent meaningfulness of the
Janoff-Bulman, world, and self-worth. This self-report scale consists of 32
1996) items scored according to a 6-point Likert scale.
Millon Behavioral It is a 150-item self-report inventory that has been formed on scores on 20
Health Inventory medical patients. Its a truefalse inventory that includes scales (%)
(MBHI) (Millon, scales grouped into four categories: basic characterological
Green, & Meagher, coping styles, psychogenic attitudes, psychosomatic
1982) correlates, and prognostic indices.
The MBHI provides scores on 20 scales in four broad

SCALES METHODOLOGY SCORES
categories of assessment. Eight reflect basic coping styles:

Categories assessment: Introversive, Inhibited, Cooperative,
Sociable, Confident, Forceful, Respectful, Sensitive.
Six provide descriptions of psychogenic attitudes: Chronic

Tension, Recent Stress, Premorbid Pessimism, Future
Despair, Social Alienation, Somatic Anxiety.
The remaining six are for use only with patients exhibiting
specific disease syndromes. Three of these assess
psychosomatic correlates: Allergic Inclination,
Gastrointestinal Susceptibility, and Cardiovascular Tendency
and the other three gauge prognostic indices: Pain
Treatment Responsivity, Life Threat Reactivity, and
Emotional Vulnerability.
Table 6 Cognitive-affective scales
4.1.6.3 Neurological Triggering Mechanisms

The most common methods for assessing the neurological triggering mechanisms are:
Electroencephalography (EEG): The accurate assessment of subcortical activity via

electroencephalography (EEG) is very difficult and may be considered a gross
assessment at best, however. False-negative findings are a common problem with
such assessment and EEGs in general.
Electrodermal responsiveness as assessed via galvanic skin response (GSR) would be

another way of assessing the reactivity of neurological triggering mechanisms (Everly &
Sobelman, 1987).
4.1.6.4 Measuring the physiology of stress response

Stress response can be divided into three broad categories: (1) the neural axes, (2) the
neuroendocrine axis, and (3) the endocrine axes. The most common assessment
technologies used to tap these phenomenological domains are the following:
1. Assessment of the Neural Axes
Technologies used for such assessment include (i) electrodermal techniques (EDA)
(eccrine sweat gland), (ii) electromyographic techniques (EMG), that measure muscle
tension. The muscle group frontalis and related facial muscles are prime targets of the
stress arousal and therefore the prime target for measurement. (iii) Cardiovascular
measures, such as heart rate, peripheral blood flow, and blood pressure. (ECG)
These techniques are often used for detecting stress in drivers (Healey & Picard, 2005)
or in users at home through ubiquitous intelligent sensing systems Choi et al. (2006).
Picard & Du (2002).
2. Assessment of the Neuroendocrine Axis

Assessment of the neuroendocrine axis of the stress response entails measurement of

the adrenal medullary catecholamines: epinephrine (adrenaline) and norepinephrine
(noradrenaline).
Aggregated medullary catecholamines may be sampled from blood or urine and
assayed via fluorometric methods. Reference values range for random sampling up to
urine, for a 24-hour urine sample up to and for timed samples, during daylight hours
(Bio-Science, 1982). For aggregated catecholamines sampled from plasma, values
range from 140 to 165 pg/ml via radioenzymatic procedures (Bio- Science, 1982).
3. Assessment of the Endocrine Axes
According to Hans Selye (1976), the most direct way of measuring the stress response
is via ACTH, the corticosteroids, and the catecholamines. The catecholamines have
already been discussed. The most commonly used index of ACTH and corticosteroid
activity is the measurement of the hormone cortisol. Cortisol is secreted by the adrenal
cortices, activated by ACTH, at a rate of about 25 to 30 mg/day and accounts for about
90% of glucocorticoid activity.
The value ranges for urinary epinephrine and norepinephrine are:
Figure 14: Value ranges for urinary epinephrine and norepinephrine
4.1.6.5 Assessment of target-organ effects

Once the stress response has been activated to pathogenic levels, it exists another
possible assessment methodology for measuring human stressthe assessment of
the target-organ effects of the stress response. The assessment of target-organ effects
can consist of measuring physical as well as psychological variables.
1. Physical Diagnosis
The goal of such assessments is to measure the integrity of the target organs
structural and functional status. It is important to keep in mind that there is still no
certainty as to the extent of the role of stress arousal in the formation of the emergent
illnesses/reactions.
The Seriousness of Illness Rating Scale (SIRS; Wyler, Masuda, & Holmes, 1968; is
one useful self-report tool for measuring illness and weighting its impact. The Stress
Audit Questionnaire (Miller & Smith, 1982) is another. The Family Disruption from
Illness Scale (Ide, 1996) extends the assessment of physical symptoms somewhat by
assessing the degree of disruption that 53 health-related symptoms impose upon daily
functioning.

2. Psychological Diagnosis
The psychological diagnosis of the stress response refers to the measurement of the
psychological effects of the stress response.
SCALE METHODOLOGY SCORES/SCALES

NAME
Minnesota Assessment of long-term stress on the personality structure of It consists of 10 scales:

Multiphasic the patient. The numerous clinical and content scales of the
Personality MMPI-2 yield a wealth of valuable information. These scales Hs: Hypochondriasis,
Inventory2 sample a wide range of abnormal or maladjusted personality D: Depression, Hy:
Conversion Hysteria,
(MMPI-2) traits (a personality trait is a rather chronic and consistent
Pd: Psychopathic
pattern of thinking and behavior) Deviate Mf:
MasculinityFemininity,
Pa: Paranoia, Pt:
Psychasthenia (trait
anxiety), Sc:
Schizophrenia, Ma:
Hypomania (manifest
energy), Si: Social
Introversion
(preference for being
alone)
The Sixteen It measures 16 functionally independent and psychologically ReservedOutgoing,

Personality meaningful dimensions isolated and replicated in more than Less IntelligentMore
Factor 30 years of factor-analytic research on normal and clinical Intelligent, Affected by
Questionnaire groups. The 16-PF consists of 187 items distributed across FeelingsEmotionally
(16-PF) the scales. Stable, Humble
(Cattell, Assertive, Sober
1972) Happy-Go-Lucky,
Expedient
Conscientious, Shy
Venturesome, Tough-
mindedTender-
minded, Trusting
Suspicious, Practical
Imaginative,
ForthrightAstute, Self-
Assured
Apprehensive,
Conservative
Experimenting, Group-
Dependent Self-
Sufficient,
Undisciplined Self-
Conflict Controlled,
RelaxedTense
Millon Clinical 175-item self-report, truefalse questionnaire. The MCMI-II, in The scales are:
Multiaxial the diagnosis of major psychiatric disorders, is clearly the Schizoid, Avoidant,
InventoryII instrument of choice when the clinician is primarily interested Antisocial, Narcissism,
(MCMI-II/III) in personologic variables and their relationship to excessive Passiveaggressive,
stress. It includes 22 clinical scales broken down into three Compulsive,
broad categories; 10 basic personality scales reflective of the Dependent, Histrionic,
personality theory of Theodore Millon (1981); three Schizotypal,
pathological personality syndromes; and nine major clinical Borderline, Sadistic,
psychiatric syndromes (Millon, 1983). The Millon Clinical Paranoid, Anxiety,


NAME
Multiaxial InventoryIII (MCMI-III) is more current version of Somatoform,

the MCMI, (Millon, 1997). It adds two scales Depressive Hypomania,
Personality and Posttraumatic Stress Disorder. Dysthymia, Alcohol
abuse, Drug abuse,
Psychotic thinking,
Psychotic depression,
Psychotic delusions,
Self-defeating
Depressive
Personality,
Posttraumatic Stress
Disorder
Stanford Consists of 30 self-report items that assess acute stress Ranges of Values per
Acute Stress disorder. The SASRQ includes items on dissociation different Studies. Acute
Reaction (subjective sense of numbing, detachment and emotional stress over 0.8
Questionnaire unresponsiveness, reduced awareness of the surroundings,
(SASRQ) derealization, depersonalization, and dissociative amnesia: 10 Firestorm .95 (.90.91
(Cardena & items in total), reexperiencing of trauma (6 items), avoidance for subscales)
Spiegel, (6 items), anxiety and hyperarousal (6 items), and impairment
1993; Shalev, in functioning (2 items). It can be scored according to a Likert- Threatening patient
Peri, Canetti, type scale (05) or dichotomously (02: 0, 35: 1) for the .87
& Schreiber, presence of a symptom.
Vicinity of shooting .93
1996)
(.72.88 for subscales)
Sexual abuse survivors

.93 (.78.86 for
subscales)
Emergency rescue .8
(.64.98 for subscales)
Catastrophic flooding
.95 (.82.88 for
subscales
Subjective Designed to measure situational (state) effects of stress on

Stress Scale the individual. The scale consists of 14 descriptors that the
(SSS), patient can use to identify his or her subjective reactions
(Berkun, during a stressful situation. Each of these descriptors comes
1962) with an empirically derived numerical weight, which the
clinician then uses to generate a subjective stress score.
Profile of Factor-analytically derived self-report inventory that measures The scales are:
Mood States six identifiable mood or affective states. It consists of 65 tensionAnxiety
(POMS). adjectives, each followed by a 5-point rating scale that the
(McNair, Lorr, patient uses to indicate the subjective presence of that DepressionDejection
& condition.
Droppleman, AngerHostility
1971)
VigorActivity
FatigueInertia
Confusion


NAME
Bewilderment
Dundee Assesses the multiple dimensions of state response in Factors for a specific
Stress State stressful performance environments, The DSSQ samples task:
Questionnaire eleven state constructs within the domains of affect,
(DSSQ; motivation and cognition. The DSSQ consists of 29 items and Stress State
Matthews et is a multidimensional instrument designed to measure (Engagement, distress,
al., 1999; workload, mood and motivational cognitions as well as worry/ versus State
Matthews et cognitions associated with stress, arousal and fatigue (Schell change (z).
al., 2002; & Cox-Fuenzalida, 2005). Matthews et al. (2002) identified
Helton et al, three broad higher-order state factors, Task Engagement, State change: -N/+N
2009) Distress, and Worry, through factor analysis of the eleven,
correlated first-order state dimensions of the DSSQ. Task
Engagement integrates state constructs that relate to task
interest and focus: energetic arousal (affect), intrinsic and
achievement-oriented motivation, and concentration
(cognition). Low task engagement corresponds to feelings of
fatigue. Distress appears to integrate unpleasant mood and
tension (affects) with lack of confidence and perceived control
(cognition). Worry is a cognitive factor primarily composed of
selffocused attention, self-esteem, and cognitive interference
resulting from both task-related and personal concerns. The
authors validated this three-dimensional taxonomy by showing
that the state factors were differentially related to task
stressors, personality factors and situational cognitions
(appraisals and coping strategies). These three dimensions
may represent the three key aspects of conscious experience
in person-environment transactions in performance settings:
commitment of effort, overload and self-evaluation,
respectively. In the following figure an example of this
questionnaire can be seen:
Table 7 Psychological diagnosis of the stress response
4.1.6.6 Specific-domain questionnaires

The previous tests arent specific to any domain. However, we can find examples of
tests that measure stress levels in driving and working situations.

- Trait driver stress was measured using the Driving Behaviour Inventory-General
Driver Stress scale (DBI-Gen; Gulian, et al., 1989). The DBI-Gen consists of 16
items that tap a general disposition, or trait susceptibility, to driver stress.
Responses are made on a Likert scale ranging from 0 to 100.
- The State Driver Stress Questionnaire is intended to assess the state experience of
driver stress; thus, it was designed to be administered verbally in actual driving
situations. It consisted of 11 items similar to those from the DBI-Gen and 10 items
from the Stress Arousal Checklist (Mackay et al., 1978). Half of the Stress Arousal
Checklist items indicated positive mood (relaxed, contented, peaceful, comfortable,
and calm) and the other half indicated negative mood (tense, bothered, nervous,
uneasy, and distressed).
- Job Stress Questionnaire (JSQ; Caplan et al., 1975; Harris et al., 1999) is a 13-item
questionnaire designed to reflect a four dimensional "Person-Environment Fit"
model of job stress. Each of the dimensions reflects an aspect of the job
environment which, in interaction with a particular individual, is hypothesized as
being stress producing. These four dimensions are: (a) Workload, (b) Role Conflict,
(c) Role Ambiguity, and (d) Utilization of Skills, which represent four common
sources of stress and do not represent all types of stress.
- A different version of JSQ (http://www.stress.org/topic-workplace.htm) has 11 items. If

the JSQ score is between 10-30, the person handles stress well; 40-60, moderately
well; 70-100, he or she is encountering problems that need to be addressed and
resolved.
4.1.6.7 Specific-users questionnaires

Previous questionnaires can be used to assess the stress levels of any user.
However, they do not take into account specific characteristics of elderly and
disabled users. The following questionnaires have been developed specifically for
elderly and disabled people:
- Subjective Stress Scale (SSS) is a questionnaire developed by Bramston and

Bostock (1994) that can be used to provide a sensitive measure of stressors for
people with intellectual disabilities. A recent study (Paul Bramston and Gerard J.
Fogarty, 1995) examined the underlying structure of the SSS by analysing
responses of 221 intellectually disabled people to the questionnaire. The results
suggest that although the actual stressors vary, persons with mild intellectual
disability are affected by the same major stress dimensions as the general
population.
- Elders Life Stress Inventory (ELSI): The ELSI is a self-report life event scale,
consisting of 30 items, each rated on a six-point scale (05), particularly developed
for the elderly. Two types of scores are derived from the scale, one reflecting
number of life events experienced during the last year and the other measuring the
subjective impact of those life events. The scale has satisfactory psychometric
properties (Aldwin, 1991).

4.1.7 Stress as cognition moderator: analysis of psychological studies

There have been relatively few studies on the relationship between stress and
cognitive performance. The most significant ones have been attached in the following
section:
4.1.7.1 Stress and reaction time
Piron (1920) and Luce (1986) reported that the weaker the stimulus (such as a very
faint light) is, the longer the reaction time is. However, after the stimulus gets to a
certain strength, reaction time becomes constant.
Reaction time is fastest with an intermediate level of arousal (stressors and stimuli
type), and deteriorates when the subject is either too relaxed or too tense (Welford,
1980; Broadbent, 1971; Freeman, 1933). That is, reaction time responds to arousal as
follows:
Figure 15: Arousal / Reaction time response
Many researchers have confirmed that reaction to sound is faster than reaction to light,
with mean auditory reaction times being 140-160 msec and visual reaction times being
180-200 msec (Galton, 1899; Woodworth and Schlosberg, 1954; Fieandt et al., 1956;
Welford, 1980; Brebner and Welford, 1980). Perhaps this is because an auditory
stimulus only takes 8-10 msec to reach the brain (Kemp et al., 1973), but a visual
stimulus takes 20-40 msec (Marshall et al., 1943). Reaction time to touch is
intermediate, at 155 msec (Robinson, 1934). Differences in reaction time between
these types of stimuli persist whether the subject is asked to make a simple response
or a complex response (Sanders, 1998, p. 114).
Simple reaction time shortens from infancy into the late 20s, then increases slowly until
the 50s and 60s, and then lengthens faster as the person gets into his 70s and beyond
(Welford, 1977; Jevas and Yan, 2001; Luchies et al., 2002; Rose et al., 2002; Der and
Deary, 2006). Luchies et al.(2002) also reported that this age effect was more marked
for complex reaction time tasks, and Der and Deary (2006) concurred. MacDonald et

al. (2008) found that reaction time variability in older adults was usually associated with
slower reaction times and worse recognition of stimuli, and suggested that variability
might be a useful measure of general neural integrity. Making a subject feel anxious
about his performance does shorten reaction times, at least on simple reaction time
tasks (Panayiotou, 2004). Mogg et al. (2008) found that it might be hard to disentangle
the effects of threat-induced anxiety from the simple distraction that the threat was
causing. In other words, even a non-threatening stimulus can cause distraction and
slow reaction time, but not by causing anxiety.
Monitoring single neurons in the right prefrontal cortex, University of Iowa researchers
found that these cells responded remarkably rapidly to unpleasant images, which
included pictures of mutilations and scenes of war. Happy or neutral pictures did not
cause the same rapid response from the neurons. "The changes in firing pattern of
neurons responding to the aversive visual stimuli happened within about 0.12
seconds, which is very fast and probably prior to the patient consciously 'seeing' the
image," said principal investigator Ralph Adolphs, Ph.D., assistant professor of
neurology. The findings are consistent with the idea that the brain evolved systems that
can respond extremely rapidly to potentially dangerous or threatening kinds of stimuli.
4.1.7.2 Stress and memory

There is substantial evidence that stress and enhanced glucocorticoid levels can have
complex influences on memory performance, with both negative and positive
consequences (Lupien and Lepage, 2001; Lupien et al., 2007; Sandi and Pinelo-Nava,
2007; Wolf, 2003).
In humans, explicit memory and working memory formation have been shown to be
frequently impaired after corticosteroid administration and as a result of psychological
stress (Kirschbaum et al., 1996; Lupien et al., 1999; Newcomer et al., 1994, 1999; Oei
et al., 2006; Payne et al., 2007). However, there are also examples in which a
potentation of memory was observed after corticosterone treatment (Buchanan
and Lovallo, 2001; Putman et al., 2004) or when psychological stress was applied
before (Payne et al., 2007) or after (Smeets et al., 2008) training. For implicit
(procedural) memory, the small number of studies that have addressed this issue
suggest that performance is unaffected by stress and elevated cortisol levels
(Kirschbaum et al., 1996; Lupien et al., 1997).
The prefrontal cortex, located just above the eyes, is necessary for the higher order
executive functions of the brain, including planning and working memory. Birnbaum et
al., 2004 studied rats and monkeys performing tasks that require intact working
memory. High activity of an intracellular signaling enzyme, protein kinase C (PKC),
disrupted the animals' ability to remember. A cellular correlate of working memory,
measured by recording neurons' electrical activity, showed this same sensitivity to high
PKC. Because stress causes release of norepinephrine, a known activator of PKC,
stress may exert its well-established inhibitory effect on prefrontal cognitive
function through PKC.
Luethi, Meier and Sandi 2009, assessed the effects of stress and cortisol on a
comprehensive variety of memory tasks in male human, including tests for explicit
memory (for neutral materials), working memory, and implicit memory (perceptual and

contextual priming and classical conditioning for emotional stimuli), with a particular
focus on different types of implicit memory. Stress (Acute stress) was induced in half of
the sample through the Trier social stress test (TSST; Kirschbaum et al., 1993). The
results are the following:
Measures of the verbal explicit/episodic memory test were indistinguishable

between the stress and the control group for neutral stimuli (not emotional stimuli).
Subjects in the stress group achieved higher scores in the spatial memory test (spatial
episodic memory was enhanced).
Explicit verbal memory was previously documented to be negatively affected by stress

and high cortisol levels (Lupien et al., 2005; Sauro et al., 2003; Wolf, 2006), with strong
evidence indicating that retrieval processes are particularly vulnerable (de Quervain et
al., 2000, 2003; Het et al., 2005; Roozendaal, 2002). Stress and glucocorticoids may
have opposing effects on explicit memory consolidation and retrieval, with enhancing
effects on consolidation and impairing effects on retrieval (e.g. Beckner et al., 2006;
Roozendaal, 2002; Sandi and Pinelo-Nava, 2007; Smeets et al., 2008). In this context,
a potential role of reconsolidation on the facilitating effects of stress on consolidation
has been discussed (Lupien and Schramek, 2006). Under these experimental
conditions, verbal explicit memory for emotionally neutral materials was not affected by
stress. Potentially enhancing effects of stress on consolidation might have balanced
out potentially impairing effects on retrieval.
The results showed no effect of stress on implicit learning for neutral stimuli.
However, the results showed an enhancement of implicit memory for negative, but
not positive, emotional stimuli. This latter result is particularly relevant since it
suggests a mood congruency effect of stress in conditioning.
Subjects exposed to the TSST had a lower reading span as well as significantly lower
total correct scores relative to controls. The results suggest a stress - induced
working memory impairment.
Such a mechanism could be potentially maladaptive in conditions of chronic stress and

could, therefore, be involved in the development and maintenance of psychiatric
conditions such as depression, post traumatic stress disorder (PTSD) and phobias
(Elzinga and Bremner, 2002; Wolf, 2008).

Figure 16: Memory tests, Luethi, Meier and Sandi 2009
Lower mean fragmentation levels at which objects were correctly identified, indicate
higher levels of object fragmentation and consequently, better object recognition. The
amount of priming did not differ between stressed and unstressed subjects.
To measure learning effects, reaction times of correct responses were analysed (mean
response accuracy was 98%). The results indicated no effect of stress on contextual
priming.
It has to be noticed, that other studies (Eunice Y. Yuen et al. 2009) suggest that acute
mild stress and glucocorticoids may have positive effects on the function of the PFC, at
both physiological and behavioral levels. It is necessary to realize, then, that the
severity of the stressor is of central importance, since previous studies suggest that
extreme acute stressors can impair working memory. Such seemingly dichotomous
results may be partially explained by considering the effects of stress and
glucocorticoids in the context of an inverted U-shaped curve, where too little or
too much glucocorticoid activity can have negative effects on learning, memory and
their neural underpinnings.
According to McGaugh and researchers at the University of Zurich, long-term memory

retrieval is impaired by stress: in a 2000 human study, these researchers
demonstrated how high levels of cortisol, a stress hormone, impaired memory, but only
when people tried to recall old, not recent, memories. That's why people get befuddled
and confused in a severe crisis.
Extreme or sustained stress can damage the brain's hippocampus, the part of the
limbic brain which is central to learning and memory, making it difficult to learn new

things. Animal research at the University of South Florida conducted by Dr. Diamond,
found that stressed rats continuously explored their surroundings, as if they had no
ability to retain memory. R.M. Sapolsky also holds this theory. According to his
research, the culprits are "glucocorticoids," a class of steroid hormones secreted from
the adrenal glands during stress. During a perceived threat, the adrenal glands
immediately release adrenalin. If the threat is severe or still persists after a couple of
minutes, the adrenals then release cortisol. Once in the brain cortisol remains much
longer than adrenalin, where it continues to affect brain cells. Chronic over-secretion of
stress hormones adversely affects brain function, especially memory. Too much
cortisol can prevent the brain from laying down a new memory, or from
accessing already existing memories.
According to an animal study held by James McGaugh, memory performance has a

time-dependent effect that correlates with the levels of circulating cortisol, which are
highest at 30 minutes. This effect only lasts for a couple of hours, so that the impairing
effect in this case is a temporary impairment of retrieval. The memory is not lost. It is
just inaccessible or less accessible for a period of time. In contrast, when cortisol
production is chemically suppressed, there are no stress-induced effects on memory
retrieval.
Studies with animals under chronic stress (Kazushige Mizoguchi et al, 2000) showed
that chronic stress induced impairment of spatial working memory via a D1
receptor-mediated hypodopaminergic mechanism in the PFC. In delayed-alternation
tasks, all experimental groups showed the same levels of performance accuracy under
the no-delay condition. The performance accuracy in the chronically stressed animals
markedly decreased along with the prolongation of the delay time (indicating that
chronic stress impairs the maintenance of a novel short-term memory, i.e.,
working memory, which is the term applied to the aspect of memory responsible for the
recall of information immediately after it has been presented.
Rearding specific tasks,there is evidence that stress affects working memory with
respect to skill in arithmetic problem-solving, reading and reading comprehension, and
the 11 retrieval of information from long term memory (Ashcraft and Kirk, 2001).
According to OHare (1999), less is known about the effects of stress on long term
memory. It is known (Bosch, 2000) that in a stressful situation, the body releases
cortisol, which has a negative effect on long term memory. In general, LTM functions
seem to be less affected by stress than WM (OHare). There is some evidence that
suggests that stress impairs the transfer of newly acquired information from WM to
LTM, thereby affecting learning.
A study performed by Payne (2005) investigated the impact of stress on human

episodic memory. All participants of the study watched narrated slide shows
containing emotional and neutral information. The experiment demonstrated that pre-
learning exposure to a psychological stressor (the Trier Social Stress Test or "TSST";
Kirschbaum, Pirke & Hellhammer, 1993) preserved or enhanced memory for emotional
aspects of the slide show, but impaired memory for neutral aspects of the slide show.
Moreover, stress exposure disrupted memory for information that was visually and
thematically central to the slide show. Memory for peripheral information, on the other
hand, was unaffected by stress.

4.1.7.3 Stress and attention

In a recent study published in The Journal of Neuroscience, Rockefeller University
scientists show that chronic daily stress affected neurons in two different areas of the
rat brain, showing for the first time a link between anxiety symptoms and the dynamic
anatomy of the brain.
One of the characteristic manifestations of prolonged stress is decreased

performance in tasks that require attention, including the ability to shift focus as
well as to learn and unlearn information. Bruce McEwen, Rockefellers Alfred E.
Mirsky Professor and head of the Harold and Margaret Milliken Hatch Laboratory of
Neuroendocrinology, was interested in finding out how this translates to changes in the
brain itself. So he and Conor Liston, a graduate student in McEwens lab, compared
neuronal change in stressed and unstressed rats.
The stressed rats performed as well as the unstressed ones in all but the last task: It
took them significantly more time to catch on to a new pattern. With the
collaboration of John Morrison, a neuroscientist at the Mount Sinai School of Medicine,
the researchers looked at neurons in two parts of the rats brains the medial
prefrontal cortex (mPFC) and the orbital frontal cortex (OFC) and saw a correlation.
The prefrontal cortex is typically involved in working memory: paying attention to one
thing at the expense of another, and shifting that attention from one focus to another.
Neurons in the stressed rats mPFC were shorter and had less branching than those in
the control rats. Prior experiments had shown that lesions in this area of the brain could
cause this effect. In this research, stress produced almost as large a deficit in
attention-shifting tasks as lesions.
Another research carried out by the Pittsburgh Research Laboratory- NIOSH- also
demonstrates how stress narrows the focus of attention. Time pressure studies, where
the subject is given a task and a specific, usually unreasonable, time to complete it,
generally support this conclusion. Other studies, however, report contradictory results
for this conclusion. Negative information gains become important under time pressure
because they need to be evaluated and discarded. If a situation involves risk (as in
response to an emergency), time pressure studies show that the subject
becomes more cautious and adopts risk-avoiding behaviour with importance
placed on avoiding losses [16]. These studies have shown that, under time pressure,
the subject adopts a simpler mode of information processing in which alternatives are
not explored fully and certain important cues are used to determine the decision.
From these studies, the experts conclude that stress narrows the focus of attention.
In other words, the focus of attention shrinks, and the individual focuses just on
critical issues and elements. This focused attention was assumed to be bad, but it
actually may be good because it can eliminate nonessential information and highlight
the most important sources.
M. A. Ellenbogen et al. (M. A. Ellenbogen et al., 2002) demonstrated how stress

affects selective attention: The effects of a stressful challenge on the processing of
emotional words were examined in college students. Stress induction was achieved
using a competitive computer task, where the individual either repeatedly lost or won
against a confederate. Mood, attention, and cortisol were recorded during the study.
The interesting findings are the following:

(1) Overall, subjects exhibited a lowering of mood in response to the negative

stressor, and either no change or a heightening of mood in response to the
positive stressor. Mood response of participants in the neutral condition fell between
the positive and negative stressor groups, and tended to be mildly negative. Increased
anxiety was reported in all three conditions, but it was greatest following negative
stress and least following the positive stressor
(2) Participants in a negative stressor condition were faster to shift attention away from
negative words than positive or neutral words; Rapid attentional disengagement from
negative words occurred in response to stress-induced negative affect, perhaps as an
adaptive means to regulate emotional arousal.
(3 )Negative and positive words are encoded in memory more readily than neutral
words.
(4) Attentional shifts away from negative words were associated with stress-induced
mood lowering;
(5) Low dysphoric subjects were more efficient in shifting attention away from all
stimuli following the negative stressor condition than either the positive or
neutral conditions. High dysphoric subjects, however, were slower to disengage
from all stimuli following the negative stressor than the positive stressor or the
neutral condition.
Figure 17: Stress / Reaction time relation. (M. A. Ellenbogen et al., 2002)

This response may represent a stress induced decrease in cognitive flexibility

and/or processing efficiency.
(6) elevated depression scores were associated with lower cortisol change from
baseline during the experimental phase, and with higher cortisol levels during the
recovery phase.
Another recent study held by the U.S. Army Research Laboratory (Morelli and Burton,
2009) examines the relationship between stress and cognition by measuring
performance on a multiple object tracking (MOT) task after exposure to stress-
inducing photographs. Postexposure performance revealed significant decrements
in MOT accuracy relative to a nonstressed control group. The ability to filter
distracting information and selectively attend to relevant information is critical to
effective performance on tasks.
A significant, pre-IAPS photograph exposure improvement in performance was

revealed, however, between initial blocks of trials when collapsing across experimental
conditions (stressed vs. nonstressed), with single degree of freedom Helmert contrasts
indicating a significant increase in accuracy from the first trial block (M = 58.98, SD =
20.37) to the second trial block (M = 62.58, SD = 20.50), F (4, 283) = 7.04, p < .001. No
significant differences in performance were shown between preexposure trial blocks
thereafter, indicative of an initial learning effect after completing the first block of trials
that then reached a plateau in performance.
As shown in the following figure, performance for both stressed and nonstressed
participants was best when tracking four targets and became progressively worse as
the number of targets increased to six targets, a pattern of performance that held
across groups and experimental conditions. Planned comparisons for nonstressed
participants, comparing pre- and postexposure performance, revealed that while
accuracy in tracking four targets did not significantly improve, significant gains in
accuracy were revealed for both five and six-target tracking conditions. No overall main
effect of photograph exposure on performance was shown for stressed participants.

Figure 18: Performance comparison stressed and non-stressed (Morelli and Burton,
2009)
After viewing the negative-valence IAPS photos (Post1,2,3), the experimental stressed
group revealed elevated levels of anxiety, depression, hostility, and dysphoria.
Figure 19: Mean score stressed and non-stressed stressed (Morelli and Burton,
December 2010 81
2009) ITACA
Liston et al colleagues (2009), also demonstrated how chronic stress disrupts the
brains ability to shift attention. Using a behavioral test, they measure attention-shifting
performance in 20 students before taking an exam. The students stress levels were
verified using a standard self-assessment questionnaire. Compared with a matched
control group of 20 relatively nonstressed med-student subjects, the exam-stressed
students scored significantly lower on a task in which they had to flexibly shift
their attention from one visual stimulus to another. Four relatively low-stress weeks
after their exams, the medical students performed the same tests again, but their
performance no longer differed significantly from that of controlssuggesting that the
neural change caused by stress had reversed itself after the stress abated.
Petrac et al. (Petrac et al., 2009) from the Department of Psychology of University of
Central Florida, performed am study with university undergraduates, examining recent
self-reported environmental stress and divided attention performance, while controlling
for state anxiety. Fifty-four university undergraduates who self-reported a wide range of
perceived recent stress (10-item perceived stress scale) completed both single and
dual (simultaneous auditory and visual stimuli) continuous performance tests. Partial
correlation analysis showed a statistically significant positive correlation between
perceived stress and the auditory omission errors from the dual condition, after
controlling for state anxiety and auditory omission errors from the single
condition (r = 0.41). This suggests that increased environmental stress relates to
decreased divided attention performance in auditory vigilance. In contrast, an
increase in state anxiety (controlling for perceived stress) was related to a
decrease in auditory omission errors from the dual condition (r = - 0.37), which
suggests that state anxiety may improve divided attention performance.
Bohnen et al. (1990) observed that subjects who exhibited high cortisol levels during a
session of continuous mental tasks had poorer divided attention following task
performance, compared with their divided attention following a control session.
A number of studies dealing with sustained attention (vigilance) and stress have
demonstrate that under most conditions, vigilance decrement becomes significant
within the first 15 minutes of attention, (Teichner,(1974)), but a decline in
detection performance can occur more quickly if the task demand conditions are
high ( Helton et al. (2007)). This occurs in both experienced and novice task
performers (Mackie, R.R. (1984)). Recent studies indicate that vigilance is hard work,
requiring the allocation of significant cognitive resources, and inducing significant levels
of stress. (Parasuraman, R. & Davies, D.R. (1977).
Large individual differences in monitoring task performance have been reported in a

number of vigilance studies. For a given task, however, the vigilance decrement
between subjects is generally consistent over time, such that individuals exhibiting
relatively higher levels of performance for a given task maintain that level of
performance over time (Davies, D.R. & Tune, G.S. (1969).
Subjects performing vigilance tasks exhibit elevated levels of epinephrine and

norepinephrine, consistent with high stress levels and indicative of a significant mental
workload. (Lundberg, U. & Frankenhaeuser, M. (1979)) Vigilance tasks may
therefore be assumed to be a stressful, hard mental work, and can be considered

a stressor (Gluckman et al., 1988). The mere need to maintain a vigil is sufficient to
produce a stress response, and subjects in vigilance experiments have reported
increased levels of stress following completion of the vigil.
In a study with subjects (Trackray, Bailey and Touchstone, 1977), these where asked
to rate themselves on five mood dimensions before and after a vigilance task. They
reported they were more strained and less attentive after the task. Subjects also
reported higher levels of fatigue and drowsiness after the task than before.
Two forms of environmental stress generate great interest with respect to vigilance
tasks: temperature and noise (Loeb and Jeantheau, 1958). Sustained attention is
degraded as thermal homeostasis of the observer is disturbed. Significant
breakdown capabilities occur when deep body temperature exceeds the bounds of
dynamics compensability (Hancock (1986c).
Noise effects (Kryter,1970) also cause variations in the efficiency. Results from
experiments employing continuous noise produce such effects (Lysaght, Warm,
Dember and Loeb,1984). In general, performance is degraded by a high level of white
noise (above 90 db SPL) when processing demands are high (stress situations).
Performance remains unchanged when processing demands are low whether the level
of white noise is above or below 90 dB level. Performance on low demand tasks is
facilitated in the presence of low-level varied noise.
A dynamic model of stress and sustained attention can be represented by an

inverted-U: an input stress can vary between extreme values of underload and
overload (hypostress and hyperstress). A zone of comfort is in the middle. The
attentional capacity (sustained attention) is intrinsically related with the psychological
and the physiological adaptability. In the following figure the relationship between
stress, adaptability and response capacity can be seen. The performance efficiency
(output) changes depending on the stress level. As it can be seen, the physiological
zone of maximal adaptability is wider than the psychological zone of maximal
adaptability, reflecting that increasing discomfort is followed by a rapid decreased in
psychological adaptability. Eventually, this is followed by a similar decrease in
physiological adaptability. (Hanckock, 1986c)

Figure 20: Inverted U model of stress (Hanckock, 1986c)
4.1.7.4 Stress and perception

Relatively lesser studies have been performed concerning stress and perception.
Auditory and visual information are conveyed to the cerebral cortex via the medial
geniculate and the lateral geniculate bodies of the thalamus.
Wickens et al., (1998), Wickens & Hollands (2000) note perceptual narrowing (or
tunneling) as a major effect of stress. Perceptual narrowing is where the effective visual
perceptual field becomes smaller with stress, such that items in the periphery become
less attended. Thus, the item focused on is typically the cause of stress or related to
relieving a stressor and other items are less available to cognitive processing.
Danigno-Subiabre et al. (2005), in an animal study, suggest that chronic stress

impairs both visual and acoustic perception. However, the study shows more
deleterious effects in the subcortical auditory system than in the visual system
and may affect the aversive system and fear-like behaviors. The stress-related
learning impairment observed in this report might also occur in humans and have a role
in the development of depressive disorders.

In figure 20, bars represent the percentages of

conditioned avoidance (% CAR) responses for
50 trials. For statistical comparisons were used
one-way ANOVA followed by post hoc Newman-
Keuls test. In A) Stress decreases the % CAR
response when using auditory or visual stimuli.
In B) stress produces a more dramatic decrease
in acoustic conditioning than in visual
conditioning. Percentages relative to control of
the acoustic and visual conditioning were
obtained from its respective % conditioned
avoidance response control values show in A.
Auditory stimuli can trigger aversive behaviors,

perhaps more likely than visual stimuli (Azrin,
1958; Reed et al., 1996; Macedo et al., 2005).
The auditory system and amygdala are related
to the aversive system, which is responsible for
the organization of fear and anxiety-like
behaviors (Brando et al.,1994). It is known that
the auditory receiving medial geniculate body of
the thalamus projects intensely to the BLA
(LeDoux et al., 1990), and that fear conditioning
plasticity in the auditory thalamus requires
amygdalar indemnity (Maren et al., 2001).
Figure 21: Avoidance conditioning Therefore, the auditory system seems to be
particularly linked to the amygdalar system during fear conditioning, and stress-related
alterations in the amygdala may produce downstream effects in the thalamus (Maren et
al., 2001) and, according to our results, even in the auditory system. In turn, auditory
system atrophy induced by stress may affect the auditory perception of aversive
signals and may impair the regulation of emotional and cognitive behaviors, affecting
environmental adaptation.
4.1.7.5 Stress and decision making

Stress plays a vital role in the way we make decisions (Massa et al, 2002, pg 1).
Problem solving and decision making in demanding real-world situations can be
susceptible to acute stress effects which manifest in a variety of ways depending on
the type of decision. The negative effects of an overload of acute stress include
attentional tunneling, working memory loss, and restrictions in long term memory
retrieval, with simple strategies being favoured over more complicated ones. The
underlying assumption is that stress can lead to errors, poor performance and bad
decisions. However, acute stress does not necessarily always have a detrimental effect
on decision making, rather stress may affect the way information is processed. Some
of those changes in strategy in response to stress are in fact adaptive. They reduce
and select the information being attended to and processed, in response to high time
pressure and reduced cognitive capacity (Flin, 2004, pg 42).

Keinan et al. tested the proposition that deficient decision making under stress is due
largely to an individuals failure to fulfil adequately a most elementary requirement of
the decision making process, i.e., the systematic consideration of all relevant decision
alternatives. In their study, which required participants to solve problems whilst under
stress, one group was put under stress and compared to a non-stress control group.
Stress was found to induce a tendency to offer solutions before all decision
alternatives had been considered and to scan such alternatives in a non-
systematic fashion.
However, contrary to popular opinion, judgment is not always compromised under

stress. Although stress may narrow the focus of attention (the data are inconclusive),
this is not necessarily a negative consequence in decision making. Some studies
(Kowalski-Trakofler* and Charles Vaught) show that the individual adopts a simpler
mode of information processing that may help in focusing on critical issues.
Decisions can only be made based on the information available, and studies have
shown that, on many occasions, decisions are made with incomplete information.
Positive effects of stressors are increased motivation and energy, faster reactions,
clearer thinking and improved memory retrieval in response to the stimulation of a
sudden challenge from the environment. For instance, rescuers who feel high in
response to an emergency are experiencing the same effect, and there may be short
term positive effects on thinking skills and energy levels (Flin, 2004, pg 50).
It is once the perceived level of challenge begins to exceed the individuals judged
ability to cope with the stressors that the symptoms of distress become prominent (Flin,
1997, pg 123). The critical appraisal of demands and coping resources is based on a
host of factors, such as their previous experience, training and personality, thus
resulting in distinct individual differences in the onset and extent of stress reactions.
Also one incident commander faced with an emergency may feel calm, confident and
totally in control, while another in the same circumstances could be uneasy, irritable
and losing grasp of the situation (Flin, 2004, pg 51). So, familiar and unfamiliar
circumstances are relevant regarding decision making.
However, negative acute stress is likely to have a particularly detrimental effect on

decision in the use of analytical and creative strategies. These require extensive
cognitive effort, especially Working Memory resources which are significantly depleted
under stress as attentional capacity diverts to monitoring the threat.
According to Massa et al (2002, pg 7), there are many decision making models that are
established as an ideal method for making good, rational decisions. One must learn to
use these methods at all costs to make a rational decision. One model that can be
used is the rational model called the Rational Problem Solving Model. This model has
four steps, including:
1. Understand the problem

2. Devise a plan.
3. Carry out the plan
4. Look back and evaluate.
Several researchers have suggested that exposing people to stress in training may
inoculate them from the effects of stress in task performance. Over the years, stress

inoculation has grown in popularity. In fact, several cognitive-behavioural stress-coping

training programs have been shown to be effective (Cannon-Bowers and Salas, 2000).
The study of Kowalski-Trakofler* and Charles Vaught associates improved

performance and performance degradation with increased stress. For some individuals,
heightened stress elevates their performance. Others are vulnerable to the negative
impacts of stress, which results in diminished performance. A physiological example of
this positive/negative dynamic of stress is athletic performance. An athlete desiring to
be at an optimal performance level whilst competing demands an optimal stress level.
The stress level should be enough to stimulate top performance, but not enough to
over-stress the body, because performance declines as the body moves toward
exhaustion. This is again an example of the inverted-U arousal-performance model of
stress, also applicable to the decision-making process.
4.1.7.6 Stress and speech and language

The NATO research study group on Speech and Language Technology (Steeneken
and Hansen, 2002) recently completed a three year project on the effect of stress on
speech production and system performance. For this purpose various speech
databases were collected. A definition of various states of stress and the corresponding
type of stressor is proposed. Results are reported from analysis and assessment
studies performed with the databases collected for this project. In general the average
intensity is observed which increases in noise (Lombard reflex, involuntary tendency of
speakers to increase the intensity of their voice when speaking in loud noise to
enhance its audibility), with anger or some types of high workload. It was also found
that mainly vowels and semivowels show a significant increase in intensity while
consonants did not.
Figure 22: Pitch frequency distribution (Steeneken and Hansen, 2002)

Pitch is the most widely considered parameter of stress evaluation. Pitch contours,
variance and distributions are observed. In Fig. 21 a distribution of pitch samples is
given for neutral, Lombard and angry speech. Mean word duration is a significant
indicator of speech in slow, clear, angry, Lombard (noise) and loud conditions.
Individual phoneme class duration under many conditions is significantly different for all
styles.
A possible relation between stages in the production and various stressors, presenting
a stressor taxonomy concerning speech, is the following:
Figure 23: Stressor taxonomy (Steeneken and Hansen, 2002)
These stressors affect differently on the different speech production subsystems,

depending on the nature of the stressor. The effects on the speech production complete
process are shown in the following figure:

Figure 24: Speech production process(Steeneken and Hansen, 2002)
4.1.8 Stress and disabled

Very few studies analysing the cognitive and physiological effects of stress on disabled
people have been performed.
A recent study (Paul Bramston, Corinne Mioche, 2001) aimed to reveal what aspects of
life bother people with different forms of disability, has shown that contrary to
expectation, total stress levels did not seem to be unusually high. Ninety-nine adults
with a visual, intellectual or physical disability completed a self-report stress scale. All
participants acknowledged their disability, but only half regarded it as a stressor in itself.
Participants with a visual, physical or intellectual disability reported similar levels and
patterns of stress with no highly significant differences between the groups. In accord with
general research findings, significantly higher stress scores were found for females,
those feeling unhealthy, those experiencing a recent major life event and those who found
no time to relax. Although not definitive, this study provides preliminary evidence that
people with various types of disability do not report particularly high levels of stress
despite having to come to terms with particular hardships associated with their disability.
Rosemary B. Hughes et al, 2005, have indentified correlates of stress and women with

physical disabilities. Based on multiple regression analyses, the findings indicate that
demographic (age, income) and disability (mobility, level of assistance needed) variables
explained a small but significant proportion of the variance in perceived stress.
Variables judged to be potentially amenable to change through psychosocial interventions
(i.e., social support, pain interference, and abuse) contributed significantly to stress over
and above the demographic and disability variables. Women with physical disabilities
reported high levels of perceived stress. Particularly at high risk are women who are
limited by pain, lack social support, and/or have experience with recent abuse
Other research has shown that women with disabilities "face more serious barriers to
achieving their life goals than do men with disabilities or women in general" (Nosek &
Hughes, p. 230) and that "five major psychosocial problems are disproportionately severe
for women with disabilities" (p. 226): depression, stress, low self-esteem, social isolation,
and abuse. Offering a perspective relevant to stress management, Bedard et al. (2003)
noted that, "men and women have different needs and current interventions target men's
needs more successfully" (p. 728).
Other studies within the disability literature have taken a more direct approach to
examining stress-coping strategies. In one such study, Yorkston, Johnson, Klasner,
Amtmann, Kuehn, and Dudgeon's (2003) conducted interviews with individuals with
multiple sclerosis to learn about the coping strategies and resources people used in daily
life. The key themes from their research included: (a) defining the work that needs to be
done (e.g., setting priorities, learning about resources and requirements, fixing it yourself),
and (b) changing how things get done (e.g., self-monitoring and vigilance for potential
problems, constructing strategies, weighing the costs and benefits of strategies). Yorkston
et al. (2003) also emphasized that, "many participants took the point of view that they
needed to be in charge of developing their own plans; this sense of authority was a source
of strength".
In summary, much of the existing literature concentrates on disability as both a negative

and positive stressor, which varies in its affects depending on the person's life
circumstances, broader social influences, and the nature of one's approaches to coping.
For example, proactively taking action and using positive coping strategies (e.g., positive
thinking) both have been identified as effective methods for promoting health and reducing
the stress some individuals associate with their disability (Elliot, et al., 2002; Noonan, et
al., 2004). The literature also includes work that presents disability as a highly positive, life
defining opportunity for personal growth, spiritual awareness, and self-discovery of inner
strength and resilience (Dunn, 1994; 2000; Tedeschi, Park, & Calhoun, 1998; Wright,
1983). This highly positive perspective may be a reaction to traditional, pathology-based
views about disability and a pointed reminder of Trieschmann's (1988) contention that
"most persons with disabilities have significant strengths and coping ability". Incorporating
social, psychological, spiritual, cultural, and behavioural resources in stress-coping can be
effective for capitalizing on, and extending people's existing strength and resilience.
Other studies show how stress accents chronic diseases symptoms, such as the
tremor of Parkinson disease.
Several researches also suggest the link between stress and creation of Alzheimer and
Parkinson diseases. Researchers at the University of California - Irvine, say that stress

hormones can rapidly accelerate the formation of the brain lesions that cause Alzheimer's
disease. Based on these findings, the researchers suggest that managing stress and
reducing certain medications used by the elderly could significantly slow down the
progression of the disease. In studies and by the experience of people with Multiple
Sclerosis (MS), stress has been implicated in the onset or exacerbations of the
disease. Patients who experienced qualitatively extreme events were 3.7 times as likely to
exacerbate as those not exposed to such events. (Franklin et al., 1988).
Using magnetic resonance imaging, Mayo Clinic researchers (Petersen et al.) found that
specific changes in the hippocampus were linked to changes in behavior associated with
aging and Alzheimer's disease. When certain parts of the hippocampus shrink or
deteriorate, specific, related memory abilities were affected. Furthermore, individuals with
a shrunken hippocampus tend to progress more rapidly towards Alzheimer's.
A recent study (Paul Bramston and Gerard J. Fogarty, 1995) examined the underlying
structure of the Subjective Stress Scale by analysing responses of 221 intellectually
disabled people to the questionnaire. The results suggest that although the actual
stressors vary, persons with mild intellectual disability are affected by the same major
stress dimensions as the general population.
The following table represents a correlation matrix constructed from 31 variables

(stressors). The table reflects whether these items formed recognizable groups
representing underlying dimensions of stress for this population. The extraction methods
were both maximum likelihood and principal axis factoring techniques. The study
examined a two stress dimensions.
Using a loading threshold of 0.30 as a rough guideline, factor 1 captures most of the
stressors that are associated with meeting the expectations of others, a general worry
factor. Factor 2, on the other hand, captures those stressors that are associated with
negative experiences: arguments with others; getting into trouble; being teased; and so
forth. These two factors appear to represent major dimensions of stress for intellectually
disabled people. The correlation between the two factors is 0.47, suggesting that these
two dimensions are part of a general stress syndrome.

Figure 25: Correlation matrix constructed from 31 variables (stressors)

4.1.9 Stress and elderly

The differential effects of everyday stress on the episodic memory test performances of
young, mid-life, and older adults has been analysed by VonDras et al. 2005.
Participants included 98 community-dwelling adults ranging in age from 19-89 years.
Everyday stress was assessed via the Perceived Stress Scale and the Elder Life
Stress Inventory. A brief battery of episodic memory tasks was administered which
included tests of Logical Memory, Verbal Paired Associates, Digit Symbol Substitution,
and Digit Symbol Incidental Learning. Results suggest that everyday hassles and
irritations as well as the accumulation of challenging life events (inducing chronic
stress) may exacerbate age-related decline on episodic memory tests that require
greater executive resources and more integrated and elaborative processing. The
study also suggests that young and mid-life adults who report high levels of everyday
stress (PSS and ELSI) approximate the Logical Memory Scale performances of
individuals 40 or more years older who report low levels of stress.
A study titled "Cortisol levels during human aging predict hippocampal atrophy and
memory deficits" demonstrate that aged humans with significant prolonged cortisol
elevations showed reduced hippocampal volume and deficits in hippocampus-
dependent memory tasks compared to normal-cortisol controls. Moreover, the degree
of hippocampal atrophy correlated strongly with both the degree of cortisol elevation
over time and current basal cortisol levels. Therefore, basal cortisol elevation may
cause hippocampal damage and impair hippocampus-dependent learning and memory
in humans. A third of the 60 volunteers, who were between ages 60 and 85, had
chronically high cortisol levels, a problem that seems to be fairly common in older
people.
The size of the hippocampus averaged 14% smaller in one group and showed high and
rising cortisol levels, compared to a group with moderate and decreasing levels. The
small hippocampus group also did worse at remembering a path through a human
maze and pictures they'd seen 24 hours earlier and two tasks that use the
hippocampus.
Insight into the mechanisms underlying the increased risk of cognitive impairment
observed in patients with type-2 diabetes, and a potential treatment strategy for
cognitive impairment in these patients, were gained in a study of 919 Scottish type-2
diabetes patients aged 60-75 years. Study subjects underwent cognitive testing,
physical examination and fasting blood sampling on the same morning. In age-adjusted
analyses, cortisol was not associated with current general cognitive ability, derived
from tests of memory, nonverbal reasoning, information processing speed, executive
function and mental flexibility, or with individual tests of cognitive function. After
adjustment for scores on the combined Junior and Senior Mill Hill Vocabulary Score
(MHVS), an estimate of peak prior cognitive ability, higher cortisol levels were
significantly associated with poorer working memory (letter-number sequencing)
and processing speed (digit symbol test). These associations persisted after
adjustment for potential confounders, including markers of metabolic and
cardiovascular status. Regression analysis also revealed a significant decline in
general cognitive ability with higher plasma cortisol quartiles, particularly among
subjects with subclinical atherosclerosis. If the relationship between cortisol levels

and cognitive ability is causal, lowering cortisol may be a means of improving cognition
in patients with type 2 diabetes (Reynolds, R.M. et al. Diabetes Care 2010, 33(4): 714).
Lupien and colleagues conducted a study with healthy elderly individuals on whom
they collected data on cortisol levels over a four year period (Lupien et al., 1994).
Individuals who exhibited increasing cortisol levels over the four years, and also had
high levels at the time of assessment, had significantly poorer performance on
tasks of selective attention and explicit memory than participants whose levels
of cortisol declined over the four years; and patients whose levels increased over four
years, but were moderate at the time of assessment. They also suggest that altered
cortisol responsivity to acute and/or chronic stress, with its detrimental effects on
memory, could be an important factor explaining the genesis of memory deficits in aged
populations.
Seeman et al. (1997) reported similar results from the subsample of women in their
cohort of elderly individuals (aged 7079 years). The women in this cohort exhibited an
inverse relationship between urinary cortisol and memory performance at baseline.
Furthermore, those women with increased cortisol levels over a 2.5 year period
were more likely to exhibit declines in memory performance. No relationships
between cortisol and memory were evident in the men.
Elizabeth Gould, Researcher at Princeton University, also notes that "levels of stress
hormones rise with aging, and are very likely to be responsible for the decline in
neurogenesis. Nevertheless, the aging brain doesn't appear to lose the ability to
generate new neurons, when stress is relieved.
Vigilance (sustained attention) research conducted with subjects across a range of

ages conflict regarding the ability to maintain alertness and sustained attention with
age. In 1991, Parasuraman and Giambra reported a trend towards lower detection
rates and higher false alarm rates with age when comparing groups between 19 and
27, 40 and 55, and 70 and 80 years old Parasuraman, R. & Giambra, L. (1991). Deaton
and Parasuraman reported in 1993 that beyond the age of 40 years, a trend
towards lower detection rates and higher false alarm rates occurs in both
cognitive tasks and sensory tasks, with higher and lower mental workloads
respectively (Deaton, J. & Parasuraman, R. (1993)). Age dependent differences in
cognitive tasks may differ with task type and workload, and some differences in
detection and false alarms may be due to the reduction in the sensitivity of sensory
organs.
4.1.10 Stress models approaches

Frank E. Ritter et al. (2006) have examined several theories from stress: Wickens et al.
(1998), Wickens and Hollands (2000), Hancock and Warm (1986, 1989) and from
biophysiology literature and have incorporated these theories into the ACT-R
architecture as overlays. These theories are typically not cast as additions to the
knowledge necessary to perform a task but are described as changes to how people
process information under stress.
The PMFs model provided by Silverman (Silverman, 2010), based in the previous
model Janis and Mann (1977), integrates stress as a result of three prime

determinants: event stress (ES), time pressure (TP), and effective fatigue (EF) -- each
of which is quantitatively derived and then emotionally filtered, f(.), providing a model
that combined stress, fatigue and emotions.
iSTRESS(t) = f{ES(t), TP(t), EF(t)}
Event stress (ES) is derived as the simple sum of the number of events in recent
history (t periods back) that are either positive in terms of sub-game goals, or negative
Thus ES is potentially reflective of how confident the agent is that its next actions will
be successful or not.
ES(t) = t=t- [positive-events(y) - adverse-events(y)] /
0 <= ES(t) <= 1.0, ES can vary over the interval 0,1 with 0 being over-confident and
unstressed, 0.5 is neutral, and 1.0 is totally stressed.
A relatively simple linear equation relates the magnitude of time pressure to stress.
Thus if one has a task that requires 30 seconds to do it accurately (ideal time or TI),
and there is only 20 seconds to do it in (available time or TA), there will be a stress
effect (and a performance impact that we address shortly).The magnitude of the time
stress, TP, can be computed as:
TP = TI /( TI + TA)
0 <= TP <= 1.0
Effective fatigue (EF) is modeled as depleting a cognitive reservoir that starts out full
when the agent is rested, and that begins to refill every time the agent rests and is
removed from relevant PMFs, such as sleep deprivation; exertion intervals; bruises and
non-incapacitating wounds; temperature, noise, and lack of nourishment, among
others. For each of these there is fatigue tolerance (FT) and reservoir replenishment
due to removal of the moderator from the environment. When the PMFs are present,
the effect of each of the PMFs is modeled as a linear function in a similar manner to
that of time pressure, such that PMFi / (PMFi + FTi)] is the behavior of each of the
moderators which vary from 0, 1.
EF(t) = EF(t -1) + { i=1,I wi[PMFi(t)/(PMFi(t) + FTi)]/I} - {EF(t-1)/10 if 0=PMFi(t)}
0 <= EF(t) <= 1.0
Logic rules are used to combine these three factors. For example, Hypervigilance and
unconflicted adherence can be modeled as:
Hypervigilance (ISTRESS>= 5) = VERY HIGH + HIGH or VERY HIGH + MEDIUM

HIGH + MEDIUM HIGH.
Unconflicted Adherence (ISTRESS<= 1) = VERY LOW + VERY LOW + VERY LOW

or ~0

This model also proposes how different emotions arise when confronted by a new
state, s, of the world, or in reaction to thinking about being in that state. In general, the
model proposes that any of a number of x diverse emotions could arise with intensity, I,
and that this intensity would be somehow correlated to importance of ones values or
value set (C) and whether those values succeed or fail for the state in question
(Silverman, 2010). The model captures the major dimensions in any emotional
construal: values, relationships and temporal aspects and assumes linear additivity of
multiple arousals of the same emotion from different sets of values that the state may
precipitate.
Regarding the state of art of stress models for specific domains, it focuses mainly in
the physiological and clinical aspects of stress:
Physiological parameters for detecting stress in drivers are proposed by Healey &
Picard (Healey & Picard, 2005), measuring ECG (heart rate, heart rate variability),
EMG (electromyogram), EDA (skin conductivity) and respiration.
Based on these physiological parameters listed, Healey and Picard (2005) proposed a
continuous stress metric. A driver stress prediction approach based on ECG and EDA
and Bayesian reasoning is proposed in (Rigas et al., 2008). The approaches described
above can be adapted for disabled and elderly drivers with data from the target user
groups and identifying the mathematical model as described in (Healey & Picard,
2005).
State driver stress was measured by Hennessy et al. (2000) in both low and high traffic
congestion using cellular telephones. The contributions of time urgency, trait driver
stress and hassles were also examined. Drivers showed substantially more state driver
stress under high than low congestion. Time urgency made a significant positive
contribution to state driver stress at both congestion levels. Trait driver stress also
contributed positively under low congestion. There was a significant hassle X trait
stress interaction under high congestion. Hassles exposure moderately increased state
driver stress for high trait stress drivers, but reduced state driver stress for medium and
low trait stress drivers. These findings indicate that state driver stress is influenced by a
combination of situational and personal factors, including factors external to the driving
context.
Hennessy et al. Measured the Trait driver stress using both the Driving Behaviour
Inventory-General Driver Stress scale and State Driver Stress Questionnaire (See
section 4.1.6).
Choi et al. (2006) present a ubiquitous intelligent sensing system for a smart home.
The ubiquitous intelligent sensing system acquires seven sensing contexts from four
sensor devices. The authors utilize association rules of data mining and linear support
vector machine to analyze context patterns of seven contexts. Also, Choi et al. (2006)
analyze stress rates of the human through the heart rate variability (HRV) pattern of the
ECG. If the human is suffering from stress, the ubiquitous intelligent sensing system
provides home service to reduce ones stress.

Palmer et al. (2004) present a model of work stress. The individual symptoms are
raised blood pressure, sleep and gastrotestinal disturbances, increased alcohol and/or
caffeine and/or nicotine intake, increased irritability & negative emotions, back pains,
tension, palpitations and headaches.
Picard & Du (2002) present a mobile phone-based system for prevention and care in
heart disease. The system aims to measure stress non-invasively using the heart rate
variability (HRV) derived from ECG.
4.2 Fatigue as cognitive and behavioural moderator

4.2.1 Definitions of Fatigue
Fatigue (also called exhaustion, lethargy, languidness, languor, lassitude, and
listlessness) is a state of awareness describing a range of afflictions, usually
associated with physical and/or mental weakness, though varying from a general state
of lethargy to a specific work-induced burning sensation within one's muscle. It is
believed that fatigue is a subjective experience and thus not easily assessed by
objective measures (Lewis & Wessely, 1992). It is also believed that because fatigue is
common with all who sustain a brain injury, it is therefore not related to damage within
a specific area of the brain (Lezak, 1978).
Individuals experiencing fatigue report it as a feeling of tiredness, weakness or

exhaustion (Rao et al., 2006). Fatigue has been defined as the unconscious
decreased ability for physical and or mental activity due to an imbalance in availability,
utilization or the retrieval of the physiological or psychological resources required to
perform the activity (Aaronson et al., 1999).
Fatigue can be a normal and important response to physical exertion, emotional stress,
boredom, or lack of sleep. Fatigue may be acute or chronic. Acute fatigue is normal
tiredness with occasional symptoms that begin quickly and last for a short time. Rest
may alleviate fatigue and allow a return to a normal level of functioning in a healthy
individual.
There are some persons in whom fatigue is a chronic state that does not necessarily
result from activity or exertion. In some instances this abnormal fatigue may be
associated with systemic disorders such as anemia, a deficiency of protein or oxygen
in the blood, addiction to drugs, increased or decreased function of the endocrine
glands, or kidney disease in which there is a large accumulation of waste products. If
excessive fatigue occurs over a prolonged period, exhaustion (marked loss of vital and
nervous power) may result. In most persons with chronic fatigue, however, the
condition seems to be associated with manic-depressive disorder.
Fatigue is a common state. According to data from The Columbia Encyclopedia, about
20% of Americans claim to have fatigue intense enough to interfere with living a normal
life. A physical cause has been estimated to be responsible 20% to 60% of the time,
while emotional or mental causes comprise the other 40% to 80% of cases of fatigue.

Chronic Fatigue Syndrome (CFS) describes prolonged debilitating fatigue that may
persist or relapse. Basically, two sets of criteria need to be met to establish a diagnosis
of chronic fatigue syndrome:
1. Have severe chronic fatigue for at least six months or longer with other known
medical conditions (whose manifestation includes fatigue) excluded by clinical
diagnosis; and
2. Concurrently have four or more of the following symptoms:
post-exertional malaise
impaired memory or concentration
unrefreshing sleep
muscle pain
multi-joint pain without redness or swelling
tender cervical or axillary lymph nodes
sore throat
headache
Consequently, people and their health care practitioners need to spend some time
together to clearly determine whether or not the problem or symptom is truly fatigue,
and if it is, any associated symptoms that may accompany the fatigue should be
explored.
Fatigue is recognized as a serious and disabling symptom in many chronic illnesses,

including cancer and neurological diseases (Multiple Sclerosis, Parkinson).
4.2.2 Types of fatigue

Singleton distinguishes two types of fatigue, Physical or Mental:
Physical fatigue or muscle weakness (or "lack of strength") is a direct term for the
inability to exert force with one's muscles to the degree that would be expected given
the individual's general physical fitness
Physical fatigue is associated to the contribution of muscle energy and the elimination
of toxins. In general, the producing factors are an inadequate diet (increased activity,
increased need for energy), insufficient oxygen (places with little ventilation), etc., and
more specifically, the contraction of muscles and tendons (by repetitive movements),
and insufficient blood flow (excessive time in the same position).
Mental Fatigue is a psychobiological state caused by prolonged periods f demanding

cognitive activity and characterized by subjective feelings of tiredness and lack of
energy (sleepiness). Mental fatigue symptoms are laxity sensations, mood changes,
and general decrease of attention. This type of fatigue comes primarily by an

individuals lack of motivation and brings boredom and introversion with it.
The impact of mental fatigue on cognitive and skilled performance is well known (es.
drivers and air pilots. On the contrary, the impact of mental fatigue on subsequent
physical performance remains largely unknown.
The study reported in J Appl Physiol 2009 (see figure below) aims to assess the effect
of mental fatigue on physical performance.
Figure 26: Effect of mental fatigue on perception of effort during high-intensity cycling
exercise: # is the significant main effect of time (P=0.05).
* is the significant main effect of condition (P=0.05).
Data are presented as means SD. Minute 0 represents end of warm-up.
4.2.3 Causes of fatigue: fatigue stimuli

The potential causes of fatigue are numerous. The majority of diseases known to man
often list fatigue or malaise as possible associated symptoms. This is complicated by
the fact that fatigue can occur in normal healthy individuals as a normal response to
physical and mental exertion. However, normal fatigue may begin to become
abnormal if it becomes chronic, extreme or prolonged fatigue; usually this occurs when
a person experiences chronic or prolonged physical or mental exertion. For example,
unusually hard physical or mental exertion for one day can result in normal fatigue that
may last about a day or sometimes more, depending on the exertion level, while daily
unusually hard physical or mental exertion may result in prolonged fatigue (usually
greater than 24 to 48 hours). This latter situation may develop into abnormal fatigue.
The causes of fatigue can be classified under several broad disease entities or lifestyle
problems that have fatigue as an associated symptom. The following table categorizes
numerous causes of fatigue. This table is not meant to be all inclusive.
The following table summarizes some common causes of fatigue but is not meant to be
comprehensive:

Common Causes of Fatigue
Metabolic/ Anemia; hypothyroidism; diabetes; electrolyte abnormalities;

Endocrine kidney disease; liver disease; Cushing's disease
Infectious mononucleosis; hepatitis; tuberculosis;

Infectious Cytomegalovirus; HIV infection; influenza (flu); malaria and many
other infectious diseases
Congestive heart failure; coronary artery disease; valvular heart

Cardiac (heart) and
disease; chronic obstructive pulmonary disease (COPD); asthma;
Pulmonary (lungs)
arrhythmias; pneumonia
Antidepressants; anti-anxiety medications; sedative medications;

Medications medication and drug withdrawal; antihistamines; steroids; some
blood pressure medications; some antidepressants
Depression; anxiety; drug abuse; alcohol abuse; eating disorders

Psychiatric (Mental Health)
(for example; bulimia; anorexia); grief and bereavement
Sleep apnea; reflux esophagitis; insomnia; narcolepsy; work shift

Sleep Problems
work or work shift changes; pregnancy; extra night hours at "work"
Cancer; rheumatology illnesses such as rheumatoid arthritis and

Other systemic lupus; fibromyalgia; chronic fatigue syndrome; normal
muscle exertion; obesity; chemotherapy and radiation therapy
Table 8 Common causes of fatigue
4.2.4 Physiological, cognitive and behavioural fatigue responses

In physiology fatigue is the inability to perform reasonable and necessary physical or
mental activity. The physiological mechanisms involved in this response to physical
activity are not yet known accurately.
Volkov (1990, op. quoting Garca Manso et al.), when defining fatigue, sums it up in the
following way:
"In the state of fatigue the concentration of ATP in nerve cells decreases and the
synthesis of acetylcholine in synaptic formations alters, the speed of processing signals
from the chemo-receptors slows down and in the motor centers develops the protective
inhibition linked to the formation of the gamma amino butyric acid. During fatigue there
is an inhibition of the activity of the glands of internal secretion, thereby reducing the
production of certain hormones and the activity of certain enzymes. This is projected on
the ATPase myofibrillar that controls the conversion of chemical energy into
mechanical work. When the speed of disintegration of ATP lowers, the power of the
work realized by the myofibril automatically decreases. During the state of fatigue, the
activity in the aerobic oxidation enzymes reduces and this alters the conjugation of the
oxidation reactions with the ATP re-synthesis. To maintain the necessary level of ATP,

a secondary intensification of the glycolysis occurs. Intensified catabolism of protein

compounds is accompanied by an increase of urea content in blood. As muscles get
tired, the reserves of energy substrates are exhausted, there is an accumulation of
decomposition products (lactic acid, ketone bodies, etc) and sudden changes of the
intracellular environment are observed.
When the metabolic reserves of the body are exhausted and the waste products
increased, as for example after prolonged exertion, the body finds it difficult to continue
its function and activity. The accumulation of lactic acid in muscle tissue and the
depletion of glycogen (stored glucose) results in muscle fatigue. The contractile
properties of muscle are reduced, and continued exertion is impossible unless the
muscle is allowed to rest. In the normal body a period of rest permits redistribution of
nutritive elements to the muscles and tissues and elimination of accumulated waste
products; the body is then ready to resume activity.
4.2.4.1 Fatigue and Brain Response

The neural mechanisms underlying feelings of fatigue are poorly understood. An
interesting study of Cook et al. (2007) uses functional magnetic resonance imaging
(fMRI) to determine the association between feelings of mental fatigue and blood
oxygen level dependent (BOLD) brain responses during a mentally fatiguing cognitive
task. Healthy, non-fatigued controls and chronic fatigue syndrome (CFS) patients were
included to determine the influence of chronic levels of fatigue on brain responses.
Results showed that mental fatigue was significantly related to brain activity during the
fatiguing cognitive task but not the finger tapping or simple auditory monitoring tasks.
CFS participants exhibited significantly greater activity in several cortical and
subcortical regions during the fatiguing cognitive task. This result suggests a strong
association between subjective feelings of mental fatigue and brain responses during
fatiguing cognition.

Figure 27: Effects of mental fatigue on physiological responses during high-

intensity cycling exercise
A: heart rate. B: stroke volume. C: cardiac output. D: mean arterial pressure. E: oxygen
consumption. F: minute ventilation. G: blood lactate. # Significant main effect of time (P
0.05). Significant difference between mental and control condition at exhaustion (P
0.05). Data are presented as means SD. Minute 0 represents end of warm-up.

4.2.5 Assessment of fatigue and human fatigue effect

9 statements concerning respondents fatigue The scoring is done by
Fatigue Severity Scale in multiple sclerosis and other conditions calculating the average
(FSS) including Chronic Fatigue Immune Dysfunction response to the questions
Syndrome (CFIDS) and Systemic Lupus (adding up all the answers
Erythmatosis (SLE). e.g., how fatigue affects and dividing by nine).
motivation, exercise, physical functioning, People with depression
carrying out duties, interfering with work, alone score about 4.5. But
family, or social life) (http://www.mult- people with fatigue related
sclerosis.org/fatigueseverityscale.html) to MS, SLE or CFIDS
average about 6.5.
Fatigue Impact Scale The FIS (Fisk et al, 1994b) is one of the most
(FIS) widely used tools to measure the fatigue state.
The FIS has been translated and validated in
30 languages. The fatigue impact scale (FIS)
was developed as a symptom-specific profile
measure of health-related quality of the
(HRQoL) for use in medical conditions in which
fatigue is a prominent chronic symptom
Modified Fatigue The MFIS is a modified form of the Fatigue The total score for the MFIS
Impact Scale (M-FIS) Impact Scale based on items derived from is the sum of the scores for
interviews with MS patients concerning how the 21 items. The full-length
fatigue impacts their lives. This instrument is a version of the MFIS has a
structured, self-report questionnaire providing Cronbach' s alpha of .81
an assessment of the effects of fatigue in while the short form has an
terms of physical, cognitive, and psychosocial alpha of .80.
functioning (specific subscales and scores, but
highly correlated). The full-length MFIS
consists of 21 items while the abbreviated
version has 5 items.
Daily Fatigue Impact D-FIS is an adaptation of the FIS for daily use.
Scale (D-FIS) This study describes the development and
initial validation of. Items for the daily fatigue
impact scale (D-FIS) were selected from the
pool of original FIS in order to represent an
uni-dimensional measure of self-reported
fatigue impact.
Brief Fatigue Inventory This is a self-report questionnaire composed A global fatigue score can
(BFI) by 10 items providing an assessment of the be obtained by averaging all
overall perception of fatigue state (weariness, the items on the BFI.
tiredness) and fatigue impact on daily activity Cronbach's alpha reliability
performance and social relationship. ranges from 0.82 to 0.97
Population: Patients with fatigue due to cancer
and cancer treatment. (Mendoza et al. 1999)
Fatigue Symptom The FSI (Hann ET AL, 1998), is a 14-item Each item on the FSI can
Inventory (FSI) self-report measure designed to assess the be scored as an individual
severity, frequency, and daily pattern of fatigue scale, providing information
as well as its perceived interference with about that variable.


quality of life. Severity is measured on Furthermore, a total
separate 11-point scales (0=not at all fatigued; Disruption Index can be
10=as fatigued as I could be) that assess computed by summing
most, least, and average fatigue in the past items 5 - 11. Item
week as well as current fatigue. Frequency is
measured as the number of days in the past It provides qualitative
week (0-7) that respondents felt fatigued as information only and is not
well as the extent of each day on average they intended to be used as a
felt fatigued (0=none of the day; 10=the entire quantitative scale.
day). Perceived interference is measured on
separate 11-point scales (0=no interference;
10=extreme interference) that assess the
degree to which fatigue in the past week was
judged to interfere with general level of activity,
ability to bathe and dress, normal work activity,
ability to concentrate, relations with others,
enjoyment of life, and mood. These
interference ratings can be summed to obtain
a total perceived interference score. The final
item provides qualitative information about
possible diurnal variation in the daily
experience of fatigue.
Multidimensional The MFSI (Stein, K. D., Martin, S. C., Hann, D. The MFSI can be scored for
Fatigue Symptom M., & Jacobsen, P. B. (1998)) A both the rationally derived
Inventory (MFSI) multidimensional measure of fatigue for use and empirically derived
with cancer patients. It is an 83-item self-report scales. Higher scores
measure designed to assess the principal indicate more fatigue.
manifestations of fatigue. Items are rated on a
5-point scale indicating how true each
statement was for the respondent during the
last week (0=not at all; 4=extremely). The
MFSI consists of both rationally and
empirically derived subscales. The rationally
derived subscales are designed to assess
global, somatic, affective, cognitive, and
behavioral manifestations of fatigue. The
empirically derived subscales, which were
developed using factor analysis, are
considered to assess general, physical,
emotional, and mental manifestations of
fatigue as well as vigor, an estimate of the
patients energy level.
Multidimensional MAF is a multidimensional scale provided by

Assessment of Fatigue the Dep. of Biobehaviorial Nursing and Health
(MAF) Systems, Univ Seattle. MAF includes 16
questions about fatigue and the effect of
fatigue on daily activities (to what degree has
fatigue interfered with the ability to do different
daily activities?). This questions provide also
a longitudinal study of fatigue (ex. Over the
past week, how often have you been fatigued
to what degree has your fatigue changed
during the past week?) . The MAF scale is a
good choice when selecting an instrument to


measure fatigue in chronic illness.
Psychomotor Vigilance The psychomotor vigilance task (PVT;

Test (PVT) Wilkinson & Houghton, 1982) is an objective
mental fatigue test. Typically set at a standard
duration of 10 min, the PVT is a sustained-
attention, reaction-timed task that is portable
and simple to use. Enclosed in a plastic case,
the device measures the speed with which
subjects respond to a visual stimulus (by
pressing a response button). The PVT has
become a standard laboratory tool for the
assessment of sustained performance in a
variety of experimental conditions (partial
sleep loss (Dinges & Powell, 1988, 1989),
chronic sleep restriction (Belenky et al., 2003;
Dinges et al., 1997; Van Dongen, Maislin,
Mullington, & Dinges, 2003), and napping)
Walter Reed Serial Effect of fatigue on average response time and

Addition/Subtraction accuracy in tasks affecting declarative
Task (SAST) knowledge (Thorne et al. 1985)
Table 9 Assessment of fatigue and human fatigue effect
4.2.6 Relationship of Fatigue to Task Performance
A lot of investigations have been reported about the impact of fatigue on tasks
performance, in case of fatigue as consequence of the interaction of time awake and
circadian rhythms (Dinges, 2004; Klerman & St. Hilaire, 2007), or the result of extended
time on task (Davies & Parasuraman, 1982; Hancock, Williams, Manning, & Miyake,
1995; Hockey, 1997; van der Linden, Frese, & Meijman, 2003). In these cases we can
address degradations in performance associated with extended time awake and
circadian rhythms. An Integrative Approach to Understanding and Predicting the
Consequences of Fatigue on Cognitive Performance, Glenn Gunzelmann and Kevin A.
Gluck]
The adverse effects of sleep loss on human performance are widespread throughout
industries that require shift work. Surveys, observational data, and anecdotal incident
reports reveal that shift workers often experience sleep episodes, particularly during
night shifts and sustained operations (Kogi & Ohta, 1975; Samel, Weg- mann, &
Vejvoda, 1995; Torsvall, kerstedt, Gillander, & Knutsson, 1989). Even when actual
sleep episodes do not occur, the underlying physiological state of sleepiness can lead
to deterioration in performance capability. Research indicates, for example, that as
sleepiness increases, alertness and vigilance deteriorate, problem- solving and
reasoning abilities become slower, psychomotor skills decline, and the rate of false
responding increases (Belenky et al., 2003; Buck, 1976; Graeber, 1982; Mullaney,
Kripke, Fleck, & Johnson, 1983; Wimmer, Hoff- mann, Bonato, & Moffitt, 1992). In turn,
this can lead to increased error and greater risk of accident, and, ultimately, can
threaten work safety.

Figure 28: Decline of Performance with Total Sleep Deprivation, Sleep &
Performance Model vs Angus & Heslegrave (1985)
4.2.6.1 Human performance on the psychomotor vigilance test
The PVT (Dinges & Powell, 1985) is widely used among scientists studying the effects
of sleep restriction and circadian desynchrony because human performance on the
task has proven to be highly sensitive to such factors (Dorrian et al., 2005). The task
requires simple sustained attention to a high signal rate, typically for a period of 10
minutes. Thus, participants must maintain stable goal directed alertness, which can be
quite difficult when sleep deprived. During the task, participants are seated and visually
fixed on a computer screen while holding a response box. Each time a red light
stimulus appears in the window, participants respond by pressing the button on the
box. The critical measure is how long it takes them to press the button after the
stimulus appears. Stimuli appear randomly at 2-10 s ISI. Once the response button is
pressed, the reaction time is displayed, to motivate the participant to respond as fast as
possible without making errors of commission.
To characterize PVT performance, reaction times are typically categorized into one of
four groups: false starts, alert responses, lapses, and sleep attacks. In the data we
present, false starts are defined as responses before the stimulus appears, or within
150 ms of stimulus onset.2. Based on the original study, responses between 150 ms
and 500 ms are identified as alert responses, reflecting response times associated with
appropriately noticing the onset of the stimulus and eliciting a response. Reactions to
the stimulus that took longer than 500 ms (up to 30 s) were considered lapses,
suggesting that some interruption or delay in the response process interfered with

responding optimally. Finally, there are some occasions where participants failed to
respond even after 30 s. These are classified as sleep attacks. In Doran et al. (2001),
trials were halted at this point and participants were alerted by a beep from the
computer to alert them for the beginning of the next trial. Throughout the experiment,
participants were monitored behaviorally and repeatedly admonished to perform to the
best of their ability (Doran et al., 2001), to ensure they maintained motivated effort
during all PVT sessions. One reason for the monitoring was to make certain that
participants eyes were open and looking at the monitor throughout each session.
Thus, lapses and, to a greater extent, sleep attacks appear to represent major
breakdowns in cognitive processing.
Under increasing levels of fatigue, performance can be characterized as becoming
progressively more variable from stimulus to stimulus (Doran et al., 2001). In general,
the distribution of responses tends to shift to the right, producing three characteristic
effects (see figure below). There is a decrease in the proportion of alert responses, with
a small corresponding increase in median reaction time for responses in this category.
Along with this shift, there is an increase in lapses, suggesting that participants vigilant
attention is unstable and frequently interrupted. As sleep loss becomes severe, there is
often a small but notable increase in the proportion of trials classified as sleep attacks.
Sleep attacks are important because their increased prevalence, though small on
average, illustrates a dramatic breakdown in information processing activity.
Figure 29: Human performance on the psychomotor vigilance test across 88 h of

total sleep deprivation (TSD), averaged over each day for the protocol (n = 13).
4.2.7 Fatigue as cognition moderator: Analysis of psychological studies

Neuropsychological testing confirms cognitive deficiencies; still, data is not consistent
due to the nature of the deficiencies and defects in the tests, as the results of these
tests may not be as obvious as the descriptions are that patients give about the
common cognitive problems.

In spite of the methodological limitations, neuropsychological research is quite

consistent in showing that the deterioration is centralized in the area of complex
information processing. There have been relatively few studies on the relationship
between fatigue and cognitive performance, usually affecting the following issues
(Hursh et al., 2003):
Impaired logical reasoning and decision- making
Impaired vigilance and attention
Slowed mental operations (e.g.arithmetic)
Loss of situational awareness
Slowed reaction time
Short cuts and lapses in optional or self-paced behaviors
Sleep loss had been largely dismissed as the cause of poor cognitive performance by
early, yet poorly designed, research. The prevailing view until the 1990s was that
people adapted to chronic sleep loss without adverse cognitive effects (Dinges et al.,
2005). More recent research has revealed sleep loss-induced neurobehavioral effects,
which often go unrecognized by the affected individuals. The neurobehavioral impact
extends from simple measures of cognition (i.e., attention and reaction time) to far
more complex errors in judgment and decision making, such as medical errors.
Involuntary microsleeps occur.
Attention to intensive performance is unstable, with increased errors of

omission and commission.
Cognitive slowing occurs in subject-paced tasks, while time pressure increases

cognitive errors.
Response time slows.
Performance declines in short-term recall of working memory.
Performance requiring divergent thinking deteriorates.
Learning (acquisition) of cognitive tasks is reduced.
An increase in response suppression errors in tasks requiring normal primarily

prefrontal cortex function.
The likelihood of response preservation on ineffective solutions is increased.
Compensatory efforts to remain behaviourally effective are increased.
Although tasks may be done well, performance deteriorates as tasks duration

increases (Durmer and Dinges, 2005).

Other studies have also shown that cognitive performance in a variety of functional
domains gets worse as sleep deprivation increases. In particular, with increasing
degrees of sleep loss, experimental participants exhibit impaired memory (e.g.,
Drummond & Brown, 2001; Drummond et al., 2000; Drummond, Gillin, & Brown, 2001;
Habeck et al., 2004; Thomas et al., 2000; Van Dongen, Baynard, Maislin, & Dinges,
2004; Van Dongen et al., 2003), and decrements in performance on naturalistic
tasks where errors can and do have serious consequences outside the laboratory
(Caldwell, 2003; Caldwell, Caldwell, Brown, & Smith, 2004; Dinges, 1995; Landrigan et
al., 2004).
4.2.7.1 Fatigue and Attention and Reaction Time

Attention and reaction time are altered by experimental sleep loss, which leads to
cumulative, dose-dependent deterioration of attention and reaction time. Attention and
reaction time are altered by experimental sleep loss, which leads to cumulative, dose-
dependent deterioration of attention and reaction time (Figure 29-A). Deterioration is
measured in part using the psychomotor vigilance task (PVT), a test that requires
continuous attention to detect randomly occurring stimuli and that is impervious to
aptitude and learning effects. In one study 48 healthy subjects were randomized to 4,
6, or 8 hours of time in bed for 14 days (Van Dongen et al., 2003). Investigators found
a dose-dependent effect, which increased over time (Figure 29-A). Performance
deficits in individuals who slept 6 hours or less per night were similar to those observed
in individuals after two nights of total sleep deprivation. Most striking was that study
subjects remained largely unaware of their performance deficits, as measured by
subjective sleepiness ratings. A second study (Belenky et al., 2003) showed a similar
dose-dependent, cumulative effect over 7 days of sleep loss in 66 healthy volunteers
(Figure 29-B). Subjects were followed for 3 days after the period of sleep restriction,
during which time they recovered, but not enough to return to their baseline levels.
Imaging studies have demonstrated a physiological basis for cognitive impairments
with sleep loss that has been linked with metabolic declines in the frontal lobe of the
brain (Thomas et al., 2000). Although there is not a large body of evidence,
associations are also likely between sleep loss and increased risk taking (Roehrs et al.,
2004).

Figure 30: Repeated nights of sleep loss have cumulative cognitive impairment.
NOTE: B, baseline day.

Fatigue and Perception
In the study of Knez et al (2006) maximal effort on a 30 km Time Trial (TT30) was
examined to assess whether it would elicit changes in objective and subjective tests of
the participants perception of the environment and their ability to anticipate future
occurrences (situation awareness; SA) and to determine the effect of post-exercise
recovery on SA. Nine experienced (5.22 2.77 years) road cyclists had their objective
and subjective levels of SA assessed prior to and at the completion of two TT30. The
participants results were compared to measurements of maximal oxygen uptake
(VO2max), peak power output (PPO), age and years of competitive cycle racing
experience. Fatigue resulting from maximal effort on a TT30 produced significant
changes in both the objective and subjective test of SA.
For cognitive monitoring, assessment of visual information processing would have to

occur unobtrusively, that is, without a dedicated stimulus-response requirement.
Measurement of visual perception as an indicator of visual awareness and cognitive
performance may be accomplished through two mechanisms: 1) visual evoked
potentials or occipital-parietal electrocortical signals, or 2) pupillary hippus. With the
advent of dry- application high-impedance electrodes, evoked and
electroencephalographic indices may be captured unobtrusively in pilots and
soldiers. The electroencephalographic information may be able to provide evidence of
failing cognition. Pupil size changes appear to reflect recognition and cognitive
processing of visual information. Measurements of these pupil oscillations can be made
unobtrusively in that eye tracking devices may be mounted into instrument panels,
helmet displays, or eyeglass frames. More likely is that future neurophysiological
monitoring systems will integrate electrocortical information with pupil and oculomotor
information into algorithms that interpret visual perception as an input for models that
predict cognitive performance.
Figure 31: Changes across Time for Mean SEM (left, Figure a) Visual Perceptual
(CVPT) Response Omissions and (right, Figure b) Complex Motor (ARPTT)
Azimuth Deviations. Adapted from Russo, Kendall, Johnson et al 2005.
4.2.7.2 Fatigue and Speech and Language

There is increasing interest in the development of objective non-invasive systems that
can be used to assist the identification and management of fatigue in both health and
workplace settings.

Measuring fatigue by analyzing a person's speech and quantifying any changes from
their normal, rested speech may enable doctors to make objective decisions about a
person's ability to function in a work environment. It may also be a useful tool for
monitoring fatigue in clinical trials where alertness is a key measured outcome (Fatigue
Shows In Speech Too, Feme Fashions Bureau, 2010).Components of speech usually
considered are length of pauses and total time to complete a spoken task.
The studys results show that as fatigue progresses, speech slows and variations in
pitch increase and tone diminishes. The conclusion is that we have less control over
the muscles that produce speech as we become more and more tired.
In (Ruiz et al., 2009) the impact on speech is studied in terms of duration of the words
uttered, duration between the words while reading the sentences, duration of the
pauses between sentences, fundamental frequency, and jitter.
4.2.8 Fatigue and disabled

Fatigue is a common symptom in patients with chronic illness. Most common medical
conditions associated with daytime sleepiness were cardiovascular disorders, diabetes, and
musculoskeletal/painful disorders.
Patients with Multiple Sclerosis:

Fatigue is probably the most common symptom of multiple sclerosis and the majority of
people with MS (PwMS) experience it at some point in their disease course, often
chronically and for the entire course of the disease.
Fatigue in multiple sclerosis can take two forms - mental and physical. Mental fatigue
can vary between mild and severely disabling and is usually exacerbated by exercise,
increased bodily or ambient temperature (Uhthoff's symptom).
Mental fatigue usually follows a daily pattern and, in this respect, differs from the
constant fatigue associated with depression. Many PwMS report that they feel fine
during the first few hours of the day but, by afternoon or early evening, feel completely
exhausted. Often a nap or a short period of rest will help them recover.
Fatigue in MS appears to be unrelated to disability status and many PwMS complain of
fatigue even when all their other symptoms are mild or in complete remission.
Why PwMS get so tired is not very well understood. Sleep disturbance is cited as one
possible explanation and PwMS are three times as likely to have disturbed sleep as the
rest of the population. A number of factors contribute to this, including a reduction in
sleep efficiency (demonstrated by polysomnographical studies), periodic involuntary leg
movements and urinary urgency.
Attempts to correlate mental fatigue with the location of MRI lesions have revealed
some rather weak correlations - for example, lesions in the frontal white matter and the
deep white matter of the right insula. Lesions in the cerebellum and brainstem that
cause involuntary nocturnal spasms are also implicated. Despite this, MRI lesion load
and fatigue do not correlate very well.
Physical fatigue can be easily induced in MS by a short walk or other physical activity.
For some reason, also not fully understood, the efficiency of demyelinated nerves
deteriorates very rapidly with use. Almost everyone whose physical functioning has
been disturbed through MS finds that their ability to do things reduces as they do them.

A typical phenomenon is for a PwMS to set out on a short stroll walking reasonably well
but as they continue their gait deteriorates or becomes impossible as physical fatigue
sets in. Just lifting a foot to take the next step can be an enormous effort. If they push
themselves beyond their limit they become completely exhausted and feel like they
have run a marathon.
Mental fatigue is often associated with physical fatigue and all you want to do after
physical exercise is to collapse on a bed and sleep.
Resting usually helps and many tasks can be completed if broken down into little bits
with frequent rests in between. Recovery times vary, usually depending on how severe
the disability is. Some people may be able to resume the activity within minutes,
whereas others may need hours.
As with mental fatigue and most other MS symptoms, heat makes physical fatigue
worse.
The received wisdom is to discover your limits and not to push yourself beyond them or
else you risk wiping yourself out for hours or days. It is important to keep up with what
exercise one is capable of to prevent the muscles from atrophying and to keep the
nerve pathways as operational as possible.
A number of methods and scales of measuring fatigue in multiple sclerosis have been
developed including Fatigue Impact Scales, Fatigue Description Scale.
Sleep Disfunction and Parkinson:

Parkinson's disease (PD) is a progressive neurodegenerative disorder affecting about
300 per 100000 people. Motor hallmarks of the disease are tremor, rigidity,
bradykinesia, and impaired balance. Non-motor symptoms of PD are a prominent
cause of disability in the PD population, and may even have a greater impact on the
quality of life than the motor symptoms themselves. Sleep dysfunction, initially
recognized by James Parkinson in his famous monograph An essay on the shaking
palsy, is one of the most striking non-motor symptoms of PD. It is only recently that
sleep disturbances in PD have received the attention of the medical and research
community.
This was steered by several important observations. Frucht and colleagues reported
sleep attacks in PD patients treated with dopamine agonists, drawing attention to
daytime sleepiness in PD. Observations that REM behavior disorder (RBD) may be an
indicator of pre-symptomatic PD opened a window of opportunity to explore early
markers of the disease, which are so vital to the development of neuroprotective
therapies. Recent studies of the hypothalamic hypocretin system dysfunction in PD
proposed a novel and exciting hypothesis of the pathophysiology of sleep dysfunction
in PD that may further translate into a better understanding of the neurodegenerative
process of PD. Sleep disorders are linked to increased level of fatigue.
4.2.9 Sleepiness and Fatigue in Elderly people

Daytime sleepiness and fatigue in the elderly are frequent complaints. It appears also
that there is a common belief that increased sleepiness during the day, and also
fatigue, tiredness, and lack of energy, are normal signs of an advanced age.
It is true to say, however, that older age comes with increased incidence of a
variety of medical disorders that may cause daytime sleepiness and fatigue, and
also with physiological changes that affect the quality and duration of overnight

sleep, as well as the rhythm city of sleep distribution. All of this may result in an
increased need to sleep or rest during the day.
Trajanovic and Shapiro report close to a third of elderly males and a quarter of elderly
females reported feeling sleepy during the day, and 30% and 15%, respectively, took
daytime naps (in contrast to 9% of the general population who report daytime
sleepiness). Conversely, excessive daytime sleepiness (EDS) is a predictor associated
to a number of medical conditions, including stroke, congestive heart failure,
myocardial infarct, depression, dementia, and global functioning.
There are many age-related changes that affect circadian rhythms. It is generally
believed that age associated circadian disruption occurs at various levels of biological
organization. These age-related disruptions have been studied in laboratory animals
and have been extensively reviewed in the literature. Such age-attenuated changes
have been associated with several neural, endocrine, metabolic, and behavioral
rhythms in animals, and are closely linked to the period of the circadian pacemaker
which controls these rhythms.
Sleep loss has a strong impact in elderly people, also because compensatory
processes are less effective. (Kuhns et al, 2004)
A study done by researchers at Columbia University urges physicians takes complaints

of exhaustion among the elderly seriously. There were 2,130 people in the study, which
took place from 1989 to 1995. The average age was 74, and 20% of the participants
were older than 80. The majority were women. They were evaluated at a baseline in
1989 and checked on every 18 months after that to 1995. The study members rated
features of their physical and mental health status. A lack of energy (fatigue state) is
more common in women than men. Twenty-two percent of the women reported
having a lack of energy, compared with 12% of the men. Those with a lack of energy
were more often not married (21% compared to 13%) than married, and older. Lack of
energy was linked to poorer reported health and physical function, such as being able
to walk fewer blocks before needing to rest and more limited ability to perform
activities of daily living. Having lack of energy also was more likely to be linked to
having joint problems, needing to take pain drugs, urinary incontinence, hearing
problems, depression and social isolation.
People with a lack of energy went to hospitals more often than those who were not
exhausted, which included trips to the emergency room, office visits, and all home care
health services.
Falls are a common and costly problem in older people (65 years and older), whether
in the community or in long-term care facilities. Each year, more than 30 percent of
older people fall (Hausdorff et al., 2001). Falls are the leading cause of death for this
particular age group (Murphy, 2000). Although most falls are not directly fatal, they are
a leading cause of injuries and trauma-related hospital admissions (Alexander et al.,
1992).
Insomnia increases the risk of falling (Brassington et al., 2000). One of the major
questions raised by this finding is what is responsible for the increased risk of fallsthe

underlying insomnia or the use of medication to treat it? Until recently, most of the
studies addressing this question were not large enough to yield an answer. In 2005, a
large, prospective study of 34,000 nursing home residents across the state of Michigan
ruled out use of hypnotic medications as a risk factor for falls (Avidan et al., 2005). In
fact, the study found that treated insomnia, and untreated insomnia, but not hypnotic
medications, were predictors of falls. Although the results of this study did not find that
insomnia increased the risk of hip fractures, other studies have found an association
(Fitzpatrick et al., 2001). Preliminary data from the Study of Osteoporosis in Women
also indicate an increased risk of falls associated with decreased sleep efficiency
and sleep time (as measured objectively using actigraphy) in a large group of older
women, with effects persisting after adjustment of health status and mood and other
confounders (Stone et al., 2004).
Nutrition is generally a foremost issue with elderly people. Kuhns (2004) reports the
effect of not equilibrated nutrition in elderly people (80 years old): one of the most
important is the increase of fatigue, linked to experienced brain fog and experienced
significant memory loss.
4.2.10 Fatigue modeling approaches

A long history of research has revealed many neurophysiological changes and
behavioral impacts of fatigue, especially caused by sleep restriction or extended time
on task.
Relevant effort has been recently invested to develop computational models of the
dynamic of fatigue and the impact on cognitive functioning. There are different
approaches but they are linked by a common overarching goal: the desire to make
principled, quantitative predictions about how increased levels of fatigue will impact
performance in applied settings.
The Sleepiness-Induced Lapsing and Cognitive Slowing (SILCS) Model (Neville et

al., 2000) and the Fatigue Degradation (FADE) Algorithm (French and Morris (2003))
are the most important models in the mathematical area (Gunzelman 2008). The aim of
these approaches is to model the dynamic of fatigue.
SILCS is an empirically-derived model for describing the behavioral dynamics of

human performance under conditions of total sleep deprivation. Interesting strengths
include a solid theoretical basis for the components of the model and fits to multiple
dependent measures. SILCS itself is comprised of three functions representing the
three empirical findings of interest to its developers: lapse frequency, lapse duration,
and general response slowing.
Each of these three functions includes parameters representing hemi-circadian,

circadian, and sleep homeostatic components, as well as a baseline performance
parameter.
The SILCS modeling methodology involves finding the best-fitting mathematical

parameters for the empirical result relevant to that function. For instance, Neville
et al. (2000) computed the number of lapses occurring per minute in their empirical
results, and then fit the Lapse frequency per min function to that result. They did the

same, separately, for mean lapse durations and also for mean response times, with
different best-fitting parameter values for the different dependent measures. Neville et
al. did not provide quantitative fit metrics, such as correlations or deviations, but they
did plot their model predictions and the empirical human data from two studies, which
allows for a qualitative visual assessment of the goodness of their account. For many
measures, the fit appears to capture qualitative trends, although there are several
cases where the model predictions stray considerably from the observed data.
The most part of studies have been conducted in the area of Biomathematical models.
Biomathematical models represent the main approach to modeling the effects of
fatigue that have come out of the sleep research community (Gunzelman 2008). In
particular these analyze the effects of sleep loss and circadian rhythmicity in terms of
impact on individuals overall level of cognitive functioning, which we refer to as
alertness.
Biomathematical models have been originated from the need to ascertain the formal
underlying relationships among sleep and circadian dynamics in the control of
alertness and neurobehavioral performance capability. Participants in these
studies are monitored continuously (often in a hospital environment) and perform a
wide array of experimental tasks over the course of sometimes extensive laboratory
stays. In addition to performance data from experimental tasks, physiological measures
are also obtained, including core body temperature, hormone levels, and often
neurophysiological data (e.g., EEG, fMRI, etc.). From these data, models of the
dynamics of the sleep homeostatic pressure and circadian rhythm have been
developed that accurately capture the relative changes in human neurobehavioral
functioning observed under conditions of sleep deprivation that occur across a variety
of dependent measures (performance data).
Dinges (2004) reviews the scientific research on biomathematical models of fatigue

and performance. The author suggests that priority should be given to research that
further establishes their basic validity, including the accuracy of the core mathematical
formulae and parameters that instantiate the interactions of sleep/wake and circadian
processes.
Two- Process approach
Sleep/Wake Predictor
Sleep Activity Fatigue & Task Effectiveness (SAFTE)
System for Aircrew Fatigue Evaluation (SAFE)
Interactive Neurobehavioral
Fatigue Audit InterDyne (FAID)
Circadian Alertness Simulator (CAS)
The two-process model of sleep regulation, first proposed by Alexander Borbly

(1982), is the leading paradigm used to predict the effects of sleep and sleep
deprivation on human cognition and performance (the other models analyzed by

Dinges have emerged from modifications to the two-process model). This model
consists of a circadian oscillator with a period of slightly over 24 hours, as well as a
homeostatic process (reflecting sleep pressure) that builds exponentially during periods
of wakefulness and dissipates exponentially during periods of sleep (Figure 1). These
two processes interact to promote alertness in the day and sleep during the night.
Figure 32: The two process model of Borbely et al. (1982)
The two process model of Borbely et al. (1982) show that sleep drive is derived from an
interaction of homeostatic sleep pressure (process S) and circadian oscillation (process
C), leading to alternating periods of wake (white areas) and sleep (shaded areas).
When subjects are sleep deprived (2nd 24-hour period), additional sleep pressure
builds, leading to greater slow wave activity (SWA) and total sleep (TST) time on the
subsequent night of recovery sleep.
Numerous refinements have since been proposed to improve the predictive validity of
the two-process model. Adjustments have been made for environmental variables
(such as light levels, and location), sleep variables (such as length and quality), and the
level of cognitive workload during waking periods.
The most important biomathematical models are the Circadian Neurobehavioral

Performance and Alertness (CNPA) (Jewett & Kronauer, 1999), and the Sleep,
Activity, Fatigue, and Task Effectiveness (SAFTE) (Hursh et al., 2004).
The alertness measures produced by CNPA and SAFTE can be seen as an inverse
measure of fatigue. The models combine sleep-loss-induced decreases in performance
with a cyclical model of circadian rhythm.
SAFTE is a simulation of how time of day (circadian rhythms) and sleep/wake patterns
influence cognitive capacity and risk of performance error. This represents a fatigue
management system to anticipate worker fatigue, optimize schedules to reduce risk of
error, and improve operator safety, effectiveness, and quality of life. The most part of
applications of SAFTE is in US Air Force, Navy and Marine. (Fatigue and Alertness
Management using the SAFTEmodel andFAST, S. Hursh and Johns Hopkins,
University School of Medicine, 2003).

Van Dongen (2004) reviewed a number of such models and showed that they capture
some of the important dynamics associated with fatigue under conditions of total sleep
deprivation.
However, the models cannot make predictions about performance on specific

tasks. The alertness measures produced can be scaled or transformed to fit particular
datasets, but this is necessarily a post hoc process.
The increasing number of biomathematical models of human performance and their

applications to both experimental and operational contexts has fueled growing interest
in the potential of these models as technologies for fatigue (or risk) management. Many
fatigue assessment tool based on the SAFTE model have been developed for the
US Air Force and the US Army (ex Fatigue Avoidance Scheduling Tool FAST- ).
Biomathematical models suffer from some important limitations. First, whereas

they make fairly accurate predictions regarding the relative changes in overall
functioning resulting from sleep deprivation, they are much less accurate in predicting
the dynamics associated with long-term restricted sleep. The other main limitation of
biomathematical models of alertness is that they produce only predictions of overall
level of functioning. That is, these models do not generate predictions of performance
in particular tasks.
Moreover the scaling process is done post-hoc; it relies on existing empirical data to
drive the process, and it must be performed separately for each dependent measure
and task of interest. Of course, in many naturalistic contexts the behavioral data are not
available and may not be possible to collect. This exposes a major drawback to these
models, since those real-world contexts are precisely the situations and tasks where
quantitative predictions are the most valuable. More importantly, biomathematical
models provide no insight regarding the cognitive mechanisms responsible for
observed declines in performance, effectively eliminating the ability to generalize the
quantitative predictions to other tasks, contexts, or dependent measures.
Regardless of how biomathematical models of fatigue and performance may ultimately

prove useful, there is widespread agreement on the merits of developing such
quantitative technologies to predict the impact on performance and safety of acute
sleep loss, cumulative sleep loss, circadian desynchrony, recovery during sleep
periods, the effects of fatigue countermeasures, and related aspects of work/rest
schedules.
An overview of other approaches and mathematical models of fatigue can also be

found in (Gunzelmann & Gluck, 2008).
Little research has been conducted in the area of computational cognitive modeling to
understand the information processing mechanisms through which neurobehavioral
factors operate to produce degradations in human performance.
In contrast to biomathematical models of human alertness, computational cognitive

modeling represents an approach to understanding human cognition that focuses
explicitly on producing performance predictions in particular tasks and contexts (e.g.,
Ritter et al., 2003, 2006). Cognitive models have been developed in a variety of areas,

providing insight into the basic information processing mechanisms that support human
cognition and performance (e.g., Anderson, 2007; Gray, 2007; Newell, 1990; Ritter et
al., 2003).
The most important cognitive approaches to modeling of fatigue effect are based on
the ACTR architecture (Jongman et al, 1999) and SOAR architecture (Neville 1998).
These models are focused on extending the potential of existing biomathematical
models of alertness by integrating them with a theory of the human information
processing system, implemented as a set of computational mechanisms that
instantiates the theory - a cognitive architecture (Anderson, 2007; Newell, 1990). This
integration is explicitly targeted at grounding estimates of alertness in cognitive
processes, which allows for the generation of quantitative performance predictions on
particular tasks.
To link the biomathematical models of alertness to a computational theory of cognition

involves determining which information processing mechanisms in the architecture are
impacted by decreased alertness, and how.
A critical assumption of the current cognitive approaches (Gunzelman 2009) is that

fatigue causes existing mechanisms in the architecture to operate less effectively or
efficiently, which can be represented by changing their associated parameters. Using
the biomathematical models to drive the dynamics of those parameters over time
serves to embed a theory of alertness in the architecture, allowing a models
performance to vary as a function of time awake and circadian rhythms. Besides
allowing for the generation of in situ performance predictions, using a cognitive
architecture provides a window into the information processing mechanisms in the
cognitive system and how they may be affected by decreased alertness.
An example of this approach is in Gunzelman 2009: a fusion of modeling approaches,

capitalizing on the strengths of both biomathematical fatigue models and computational
cognitive models. They used the alertness (overall cognitive function) predictions from
the biomathematical models to drive changes to architectural parameters in ACT-R.
Mathematical models that capture the dynamics of the human arousal system are
integrated with a cognitive architecture that instantiates a unified theory of the
mechanisms of human cognition. The integration of these approaches leads to an
enhanced ability to quantify the impact of fatigue on performance in particular tasks
(VEDI GUNZLM 2009)
The approach is to use biomathematical model of the fatigue to drive parameters

changes in ACTR, to produce decrements in the models performance like those
observed in human participants (Gunzelm, Understanding Decrements in Knowledge).
Gunzelam reports some examples of hybrid models to model the fatigue as cognitive
performance moderator (Gunzelm 2008): the unified framework PMFServ (Silverman
2007) , the task network modeling IMPRINT PRO, the solution based on SOAR
architecture described by Jones, Laird, & Neville (1998) added mechanisms for fatigue
to enhance the realism of behavior representations within the TacAir- Soar model of
human pilot performance (Laird et al., 1998), by adding mechanisms to model the
effects of cognitive moderators on performance. To accomplish this, Jones et al. (1998)
made reference to a mathematical model developed by Neville (1997) and colleagues

(Neville et al., 2000) to represent the dynamics of fatigue (the SILCS model described
before).
The most relevant approach is the ACT-R/F provided by Gunzelmann in Gunz 2008: a
linear function is used to map the biomathematical model of alertness to parameters in
ACT R.
So we have,
Px=s. At + I
Where Px is a specific parameter in ACT-R, s is the slope of the linear function, Ix is the
intercept, At is the predicted level of alertness according to the biomathematical model
Using this approach we can model changes on human performance as a consequence

of fatigue in different tasks emphasizing different ACT-R components (eg. sustained
attention, reaction time, declarative memory).
EU FP6 Project EUCLOCK (http://www.euclock.eu) in its subproject Entrainment in

Humans also aims at development of subject-independent models of circadian
rhythms that compensate for inter-individual differences. It was established that the
circadian component can be detected using ambulatory physiological monitoring and
multivariate regression modelling (Kolodyazhniy et al., 2009). Important non-invasive
monitored variables are: skin temperatures; motion acceleration; ambient light; sleep
log.
Subject-independent circadian ambulatory models using multi-channel recordings of

the variables listed above are being developed and validated in real life conditions and
under constant routine in a chronobiology laboratory in Basel within the project
EUCLOCK (Kolodyazhniy et al., 2009).
Gundel et al. (2005) reports that the German Aerospace Center DLR developed a
computer tool for the prediction of fatigue based on a biomedical model of fatigue
combining four components. The circadian component describes the course of fatigue
during the day with the maximum of alertness during the day and high fatigue during
the night. The sleep-related component describes the effect of sleep timing, duration
and quality. In general, fatigue increases with the time awake. The third component,
sleep inertia, describes fatigue occurring after awakening even from a good sleep.
These three components are task invariant. The fourth component, the time-on-task
component, describes the fatiguing effect of a specific task. The availability of the
software needs to be checked, as well as the applicability in the target user groups of
VERITAS.
Williamson (2007) provides an analysis of the following parameters of fatigue for

drivers: slower reaction times (consequently slower information processing), reduced
vigilance and change in mood state. Some measurable parameters for fatigue
detection are:
Steering and speed performance (Peters, 2001). Increased variation in lateral and
longitudinal control would then indicate possible signs of fatigue due to physical
workload

Eyelid movement (Liying & Haoxiang, 2008) (percentage of eyelid closure over the
pupil, PERCLOS, and average eye closure speed, AECS)
Tracking of gaze (Liu et al., 2002; Eriksson & Papanikotopoulos, 1997)
Speech (Ruiz et al., 2009) (duration of the words uttered, duration between the
words while reading the sentences, duration of the pauses between sentences,
fundamental frequency, and jitter).
ECG and EEG (Yang et al., 2010).
Yang et al. (2010) propose a driver fatigue recognition model based on the dynamic
Bayesian network, information fusion and multiple contextual and physiological
features (eye movement, ECG, and EEG), and apply the first-order Hidden Markov
Model to compute the dynamics of the Bayesian network at different time slices. The
experimental validation shows the effectiveness of the proposed system; also it
indicates that the contact physiological features (especially ECG and EEG) are
significant factors for inferring the fatigue state of a driver.
Driver fatigue was modelled with the cognitive architecture ACT-R (Gunzelmann et al.,
2009). This approach needs to be adapted for the elderly and disabled users by
introducing respective correction factors and comparing the model performance with
data from the VERITAS target user groups.
Perry (2010) considers ergonomic design of workplaces for the elderly and lists
changes in physiological changes that typically contribute to fatigue.

4.3 Emotions as cognitive and behavioral moderator

4.3.1 Definition
Emotions are psycho-physiological phenomena that represent modes of adaptation to
certain environmental stimuli. In humans, emotions fundamentally involve
"physiological arousal, expressive behaviours, and conscious experience" (Myers,
2004). The creation of the emotional experience has been charged from James-Lange
theory and the experience of bodily changes (Cannon, 1927) to other more functional
approaches to emotions, like the one form Nico Frijda and Freitas-Magalhaes, where
emotions have evolved for a particular function, such as to keep the subject safe.
Psychologically, emotions can increase or lower the amounts of stress, create

anxiety, depression or even panic and can elevate a range of certain behaviours in the
individual's response hierarchy.
Cognitively, emotions can alter attention, activate relevant associated networks in the
memory, reduce perception, affect the decision making procedures and the orientation
and even create temporary speech and language disorders.
Physiologically, emotions quickly organise the responses of the different biological

systems, including facial expressions, muscles, voice, SNA (sympathetic nervous
activity) and the endocrine system, with the aim to establish an optimal internal
environment for the most effective behaviour.
Behaviourally, emotions establish our position with respect to our environment,

impelling us towards certain people, objects, actions and ideas and averting us from
others. The emotions also act as a repository of innate and learned influences,
possessing certain invariable characteristics, and others that display a certain variation
between individuals, groups and cultures (Levenson, 1994).
4.3.2 Theories of emotions

Each individual experiences an emotion in a particular way, depending on previous
experiences, learning experience, character and the specific situation characteristics.
Some physiological and behavioural reactions that trigger emotions are innate, while
others may be acquired (Levenson, 1994). This is obviously making the objective
categorisation of the emotions a very difficult task.
Until now, no definitive classification of emotions exists. Numerous taxonomies and

theories have been proposed during the history of emotions. Some examples of the
various theories are, namely:
Somatic theories, claim that bodily responses rather than judgments are essential to
emotions. (Damasio, 1991)
James-Lange theory. The first modern version of such theories comes from William
James in the 1880s. (Cannon, 1927)
Neurobiological theories, based on discoveries made through neural mapping of the

limbic system, the neurobiological explanation of human emotion is that emotion is a

pleasant or unpleasant mental state organized in the limbic system of the mammalian
brain.
Prefrontal cortex. Two neurobiological models of emotion in the prefrontal cortex

made opposing predictions. The Valence Model predicted that anger, a negative
emotion, would activate the right prefrontal cortex. The Direction Model predicted that
anger, an approach emotion, would activate the left prefrontal cortex.
Homeostatic emotion. "Homeostatic emotions" are feelings evoked by internal body

states, which modulate our behaviour. Each internal body state is a signal from a body
system saying that something is going wrong. When one of these systems drifts out of
balance the human being starts to feel a homeostatic emotion, and the feeling prompts
him/her to do what is necessary to restore that system to balance.
Cognitive theories, argue that cognitive activity in the form of judgements,

evaluations, or thoughts is necessary in order for an emotion to occur.
Perceptual theory. This is the most recent theory of the somatic and cognitive theories
of emotion that is arguing that bodily responses are central to emotions, yet it
emphasizes the meaningfulness of emotions or the idea that emotions are about
something, as is recognised by cognitive theories.
Affective events theory. This a communication-based theory developed by Howard

M. Weiss and Russell Cropanzano (1996), (Freedheim 2003) that looks at the causes,
structures, and consequences of emotional experience (especially in work contexts).
This theory suggests that emotions are influenced and caused by events which in turn
influence attitudes and behaviours.
Cannon-Bard theory. This theory is also known as the thalamic theory and it is a
theory of emotion developed by physiologists Walter Cannon and Philip Bard,
suggesting that individuals experience emotions and physiologically react
simultaneously.
Two-factor theory. This is based on experiments purportedly showing that subjects

can have different emotional reactions despite being placed into the same physiological
state with an injection of adrenaline.
Component process model. A recent version of the cognitive theory regards

emotions more broadly as the synchronization of many different bodily and cognitive
components. Emotions are identified with the overall process whereby low-level
cognitive appraisals, in particular the processing of relevance, trigger bodily reactions,
behaviours, feelings, and actions.
Of all the aforementioned theories three are the prevailing ones about emotions and
they are based on physiology, behaviours and the interaction of physical and mental
factors, namely:
1. James (1884) and Lange (1885) theory.

Both authors suggest that the experience of emotion is based on the awareness of the
physiological responses or physical sensations that cause the emotion, such as
increased heart rate and muscular spasms. This theory proposes that emotions
happen as a result of the physiological events, rather than being the cause of them.
Figure 33: James and Lange theory illustration.
2. Cannon (1927) and Bard (1938) theory.
These authors emphasise that when a stimulating event happens, we feel emotions
and physiological changes (such as muscular tension, sweating, etc.) at the same time.
All physical reactions are the same for different emotions. One emotion cannot be
distinguished from another based (only) on physiological signals experience of the
emotion with a stimulus.
Figure 34: Cannon and Bard theory illustration.
3. Schacter-Singer (1962) theory.
This theory maintains that emotions are stimulated due to the cognitive evaluation of an
event and corporal responses. The person notices the physiological changes, points
out what happens around him/her and names the emotions in accordance with both
types of observations.
The first theory explains the emotions from only physiology, the second from cognition
and the third theory is based on the interaction of both (physiological and cognitive
factors).

Figure 35: Schacter-Singer theory illustration.
4.3.3 Approaches to emotions

According to the research in psychology, three major approaches to emotion modelling
can be distinguished (Grandjean, Sander, & Scherer, 2008):
(1) categorical approach,
(2) dimensional approach, and
(3) appraisal-based approach.
1. Categorical approach
The categorical approach is based on research on basic emotions, pioneered by

Darwin (1998), interpreted by Tomkins (1962, 1963) and supported by findings of
Ekman et al. (1992, 1999). According to this approach a small number of emotions
exist that are basic, hard-wired in our brain, and recognized universally (e.g., Ekman &
Friesen, 2003).
It is claimed that basic emotions are to be biologically fixed, innate and as a result
universal to all humans and many animals as well. Complex emotions are then either
refined versions of basic emotions, culturally specific or idiosyncratic.
Before beginning our presentation of the emotions classification, a detailed list of all the
existing emotions and their description is presented at the table below.
Emotion type Description
Affection Affection is the expression of care. It symbolizes security, protection,

comfort and approval and is often associated with a feeling or type of
love.
Anger A normal emotion that involves a strong uncomfortable and

emotional response to a perceived provocation.
Annoyance An unpleasant mental state that is characterized by such effects as

irritation and distraction from one's conscious thinking.

Angst Is used in English to describe an intense feeling of strife.
Apathy A state of indifference, or the suppression of emotions such as

concern, excitement, motivation and passion.
Anxiety Psychological and physiological state characterized by cognitive,

somatic, emotional, and behavioural components combined to create
an unpleasant feeling that is typically associated with uneasiness,
apprehension, fear, or worry.
Awe An emotion comparable to wonder but less joyous, and more fearful
or respectful.
Contempt An intense feeling or attitude of regarding someone or something as

inferior, base, or worthless.
Curiosity An emotion related to natural inquisitive behaviour such as

exploration, investigation, and learning, evident by observation in
human and many animal species.
Desire A sense of longing for a person or object or hoping for an outcome.
Despair Is the loss of hope.
Disappointment Is the feeling of dissatisfaction that follows the failure of expectations

to manifest.
Disgust Is an emotion that is typically associated with things that are

regarded as unclean, inedible, infectious, or otherwise offensive.
Ecstasy A subjective experience of total involvement of the subject, with an

object of his or her awareness.
Empathy Is the capacity to share the sadness or happiness of another sentient

being through consciousness rather than physically.
Envy an emotion that "occurs when a person lacks another's (perceived)

superior quality, achievement, or possession and either desires it or
wishes that the other lacked it."
Embarrassment An emotional state experienced upon having a socially or

professionally unacceptable act or condition witnessed by or
revealed to others.
Euphoria A mental/emotional state defined as a profound sense of well-being.
Fear Is an emotional response to a perceived threat.

Frustration Is a common emotional response to opposition that is related to

anger and disappointment, it arises from the perceived resistance to
the fulfilment of individual will.
Gratitude A positive emotion or attitude in acknowledgment of a benefit that

one has received or will receive.
Grief Is a multi-faceted response to loss, particularly to the loss of

someone or something to which a bond was formed.
Guilt A cognitive or an emotional experience that occurs when a person

realizes or believesaccurately or notthat he or she has violated a
moral standard, and bears significant responsibility for that violation.
Happiness is a state of mind or feeling characterized by contentment, love,

satisfaction, pleasure, or joy
Hate is an intense feeling of dislike
Hope Is a belief in a positive outcome related to events and circumstances

in one's life.
Horror The feeling of dread and anticipation that precedes the horrifying
experience.
Hostility Is a form of angry internal rejection or denial in psychology. . In

everyday speech it is more commonly used as a synonym for anger
and aggression.
Hysteria A state of mind, one of unmanageable emotional excesses.
Interest A feeling or emotion that causes attention to focus on an object or an

event or a process.
Jealousy Is an emotion and typically refers to the negative thoughts and

feelings of insecurity, fear, and anxiety over an anticipated loss of
something that the person values, such as a relationship, friendship,
or love.
Joy/Happiness Is a state of mind or feeling characterized by contentment, love,

satisfaction or pleasure.
Love Is the emotion of strong affection and personal attachment.
Lust Is a craving for sexual intercourse, which can sometimes assume a

violent or self-indulgent character.

Misery Is a feeling of great unhappiness, suffering and/or pain.
Pity A tender or sometimes slightly contemptuous sorrow or empathy for

people, a person, or an animal in misery, pain, or distress.
Pride A high sense of one's personal status or ego (i.e., leading to

judgments of personality and character) or the specific mostly
positive emotion that is a product of praise or independent self-
reflection.
Rage A mental state that is one extreme of the intensity spectrum of anger.
Regret Is a negative conscious and emotional reaction to personal past acts

and behaviours.
Remorse Is an emotional expression of personal regret felt by a person after

he or she has committed an act which they deem to be shameful,
hurtful or violent.
Sadness Is an emotion characterized by feelings of disadvantage, loss,

helplessness, sorrow and rage.
Shame A cognitive or an emotional experience that occurs when a person

realizes or believes that this/her cultural or social values have been
violated.
Shyness The feeling of apprehension, lack of comfort, or awkwardness

experienced when a person is in proximity to, approaching, or being
approached by other people, especially in new situations or with
unfamiliar people.
Sorrow Deep distress, sadness, or regret especially for the loss of someone
or something loved
Suffering Is an individual's basic affective experience of unpleasantness and

aversion associated with harm or threat of harm.
Surprise Is a brief emotional state experienced as the result of an unexpected

significant event
Wonder Is an emotion comparable to surprise that people feel when

perceiving something rare or unexpected.
Worry Is thoughts and images of a negative nature in which mental

attempts are made to avoid anticipated potential threats.
Table 10: Emotions illustration and description.

Ekman (1982) devised a list of basic emotions from cross-cultural research on the Fore
tribesmen of Papua New Guinea. There are six basic categories of emotions. He
observed that members of an isolated culture could reliably identify the expressions of
emotion in photographs of people from cultures with which the Fore were not yet
familiar. They could also ascribe facial expressions to descriptions of situations. On this
evidence, he concluded that the expressions associated with some emotions were
basic or biologically universal to all humans. The following is Ekman's (1972) list of
basic emotions:
ANGER: Rage, anger, resentment, fury, irritability.
DISGUST: Aversion, revulsion, moving away from the object that causes
aversion.
FEAR: Anticipation of a threat or danger that causes anxiety, uncertainty or

insecurity.
HAPPINESS: Fun, euphoria, gratification, contentment; provides a sensation of

well-being and security.
SADNESS: Emotional pain, loneliness, pessimism.
SURPRISE: Shock, astonishment, confusion. Surprise is very transitory. A

cognitive approximation may be provided to know what happened.
In the 1990s Ekman expanded his list of basic emotions, including a range of positive
and negative emotions not all of which are encoded in facial muscles. The newly
included emotions are:
Amusement
Contempt
Contentment
Embarrassment
Excitement
Guilt
Pride in achievement
Relief
Satisfaction
Sensory pleasure
Shame

Keeping in mind the adaptive purpose of the emotions and based upon the basic
emotions initially introduced by Ekman, it may be said that these emotions have
different functions:
Fear: Leads towards protection.
Surprise: Assists with orientation when faced with a new situation.
Aversion: Leads us to reject that which lies ahead.
Anger: Leads us towards destruction.
Joy: Leads us towards reproduction (a desire to reproduce that which made

us feel good).
Sadness: Motivates towards new personal reintegration.
Theorist Basic Emotions

Plutchik Acceptance, anger, anticipation, disgust, joy, fear, sadness, surprise
Arnold Anger, aversion, courage, dejection, desire, despair, fear, hate, hope,
love, sadness
Ekman, Anger, disgust, fear, joy, sadness, surprise

Friesen, and
Ellsworth
Frijda Desire, happiness, interest, surprise, wonder, sorrow
Gray Rage and terror, anxiety, joy
Izard Anger, contempt, disgust, distress, fear, guilt, interest, joy, shame,
surprise
James Fear, grief, love, rage
McDougall Anger, disgust, elation, fear, subjection, tender-emotion, wonder
Mowrer Pain, pleasure
Oatley and Anger, disgust, anxiety, happiness, sadness

Johnson-
Laird
Panksepp Expectancy, fear, rage, panic
Tomkins Anger, interest, contempt, disgust, distress, fear, joy, shame, surprise

Watson Fear, love, rage
Weiner and Happiness, sadness

Graham
Table 11: Various theories form the history of emotions research
To date, Ekmans theory on universality and interpretation of affective nonverbal

expressions in terms of basic emotion categories has been the most commonly
adopted approach in research on automatic affect recognition.
The table below presents a wide range of theoretical approaches identifying the basic
emotions made in 1990 by Ortony and Turne.
On the other hand, however, a number of researchers in psychology argued that it is

necessary to go beyond discrete emotions. Among various classification schemes,
Baron- Cohen and his colleagues, for instance, have investigated cognitive mental
states (e.g., agreement, concentrating, disagreement, thinking, reluctance, and
interest) and their use in daily life. They did so via analysis of multiple asynchronous
information sources such as facial actions, purposeful head gestures, and eye-gaze
direction. They showed that cognitive mental states occur more often in everyday
interactions than the basic emotions (Baron-Cohen & Tead, 2003). These states were
also found relevant in representing problem-solving and decision-making processes in
human-computer Interaction (HCI) context and have been used by a number of
researchers, though based on deliberately displayed behaviour rather than in natural
scenarios (e.g., El Kaliouby & Robinson, 2005).
2. Dimensional approach
According to the dimensional approach, affective emotional states are not independent
from one another; rather, they are related to one another in a systematic manner. In
this approach, the majority of affect variability is covered by three dimensions: valence,
arousal, and potency (dominance) (Davitz, 1964; Mehrabian & Russell, 1974; Osgood,
Suci, & Tannenbaum, 1957). The valence dimension refers to how positive or negative
the emotion is, and ranges from unpleasant feelings to pleasant feelings of happiness.
The arousal dimension refers to how excited or apathetic the emotion is, and it ranges
from sleepiness or boredom to frantic excitement. The power dimension refers to the
degree of power or sense of control over the emotion.
Russell (1980) introduced a circular configuration called Circumflex of Affect (Figure

35) and proposed that each basic emotion represents a bipolar entity being a part of
the same emotional continuum. The proposed polars are arousal (relaxed vs. aroused)
and valence (pleasant vs. unpleasant). As illustrated in Figure 35, the proposed
emotional space consists of four quadrants: low arousal positive, high arousal positive,
low arousal negative, and high arousal negative. In this way, as argued by Russell, it is
possible to characterize all emotions by their valence and arousal, and different
emotional labels could be plotted at various positions on this two-dimensional plane.

Figure 36: Emotions mapping to arousal, valence, and stance dimensions [A;V;].
Robert Plutchik supports that the emotions are spited in three dimensions within a
"circumplex model", presented in Figure 36, which describes the relations among
emotions. The vertical dimension represents intensity, and the circle represents
degrees of similarity among the emotions. Plutchik posited eight primary emotion
dimensions arranged as four pairs of opposites:
joy VS sadness;
anger VS fear;
trust VS disgust; and
surprise VS anticipation.

Figure 37: Plutchik circumplex model. (Plutchik, 1980)
Plutchik has also moved to the combination of two primary emotions to the so called
created dyad. These dyads are also classified by the frequency they may occur. The
results are illustrated at the figure below.

Figure 38: Combinations of primary emotions in dyads. (Plutchik, 1980)
The complexity with which emotions can be expressed leads to the belief that emotion
is a multi-factorial or multi-dimensional process. One always has the impression of a
lack of words to precisely describe emotions.
According to V.J. Wukmir (1967), there is a common factor to all emotions below this
complexity: each emotion expresses an amount or a magnitude on a positive/negative
scale. Thus, positive and negative emotions are experienced in variable levels and
intensities. Short or gradual changes in emotional intensity can be experienced,
whether towards the positive or the negative. In other words, all emotion represents a
magnitude or a measure along a continuum, which may have positive or negative
values.
In everyday language, emotions are expressed within a positive/negative scale in

variable magnitudes, for example, I feel good, I feel great, I feel fantastic
(intensities or levels on the positive pole) or I feel bad, I feel very bad, I feel terrible
(intensities or levels on the negative pole).
Different words are used depending on the situation that causes the emotion, such as
sadness, surprise, anger, fear, joy, etc., which also indicates their sign (positive
or negative). Depending on the intensity of the emotion, words such as none, a little,
quite, very, etc. are used to describe the emotion. For example, a person may say I
feel very well (positive) or I feel a little sad (negative).

As a consequence, two well-differentiated components can be recognised in all

emotion. On the one hand, a qualitative component expressed with a word is used to
describe the emotion (love, friendship, fear, insecurity, etc.) and which determines its
positive or negative sign. On the other hand, all emotion has a quantitative component
expressed with words of magnitude (a little, quite, a lot, great, somewhat, etc.), for both
positive and negative emotions. The following table reflects these two components of
all emotion.
Figure 39: Dimensions of emotions according to V.J. Wukmir (1967)
Nevertheless positive and negative scale for emotions constitutes an easy and
understandable way to utilise the variety of emotions with an effective an efficient way.
Thus, in VERITAS we will use this classification of emotions being inspired and based
upon the next figure from which we will use only the emotions in circle that are
considered basic from Ekman (1982), which are fear, surprise, aversion, anger, joy and
sadness.

Figure 40: Emotions position in the negative-positive scale.
3. Appraisal based approach
Scherer and his colleagues introduced another set of psychological models, referred to
as componential models of emotion, which are based on appraisal theory (Scherer et
al., 2001). The appraisal-based approach, which can also be seen as extension to the
dimensional approach, claims that emotions are generated through continuous,
recursive subjective evaluation of both our own internal state and the state of the
outside world. This approach views emotions through changes in all relevant
components including cognition, motivation, physiological reactions, motor
expressions, and feelings. The advantage of componential models is that they do not
limit emotional states to a fixed number of discrete categories or to a few basic
dimensions. Instead, they focus on the variability of different emotional states, as
produced by different types of appraisal patterns. Emotion is described through a set of
stimulus evaluation checks, including the novelty, intrinsic pleasantness, goal-based
significance, coping potential, and compatibility with standards. Therefore,
differentiating between various emotions and modelling individual differences become
possible. Nevertheless using the appraisal-based approach for automatic emotion
recognition is an open research question due to the fact that this approach requires
complex, multicomponential and sophisticated measurements of change.
4.3.4 Physiological, cognitive and behavioural responses to emotions

Several studies reveal that a large part of the physiological activity involved in emotions
is regulated by the sympathetic (excitement) and the parasympathetic (calmness)
division in the autonomic nervous system. Emotions with similar levels of
excitement/activation and the same valence are difficult to distinguish, like for example:
fear/anger.

However, science has discovered subtle differences (between these emotions) in the
cerebral cortical activity; in the use of neuronal connections and in the secretion of
hormones. For example, fear and anger both have differences in temperature in the
fingers and hormonal secretions and have different cerebral activation; fear and joy
stimulate different facial muscles; negative emotions (revulsion, depression) cause
greater activity in the right prefrontal cortex; positive emotions (joy, optimism) cause
greater activity in the left prefrontal cortex.
Many of the physiological responses under the control of the Autonomic Nervous
System (ANS) are interesting for the study of emotional and cognitive behaviour. The
function of the ANS is to regulate the internal balance of the organism in the presence
of internal or external changes by dynamically adjusting physiological parameters such
as: heart contraction force and rate, sweat gland activity, skin temperature reaction,
skeletal muscle strength. The figure that follows presents a simplified representation of
the nervous system and the physiological responses under its control (Andreassi
2000).
Figure 41: Simplified representation of the nervous system and the physiological
responses under its control. (Andreassi 2000)
The physiological component of emotions includes also the changes developed in the
central nervous system (CNS) and which is related to the presence of specific
emotional states. The following CNS centres are considered particularly active during
the emotional processes:

The cerebral cortex forms part of the CNS.
Davidoff and Sloman (Sloman 1981, Davidoff 1980) agree that the active cerebral
cortex regulates and integrates the reactions related to emotions. Based on Sloman's
Computational Architecture of the Mind (Sloman 1978), a central administrative
process is required that is dedicated to the decisions related to intentions, selecting
plans and conflict resolution: According to Sloman (1981), the decisions of an
intelligent system cannot be made independently, hence the processes developed by
the mind that involve conflicting achievements need to be resolved to a greater level of
abstraction. Sloman considers the emotional processes as the structural foundation of
the mind of intelligent systems.
The hypothalamus forms part of the limbic system
This is dedicated to the sympathetic nervous system. This centre is related to emotions
such as fear and anger and participates as an activator for sexual activity and thirst.
Amygdala
The amygdala is related to feelings of anger, pleasure, pain and fear. Extirpation of the
amygdala causes complex changes to behaviour.
The spinal cord
According to Harold Wolff (Wolff 1943, Davidoff 1980), all emotions are related to
specific physiological responses to emotions. Wolff's (Wolff 1943) experiments showed
that the walls of the stomach reacted to emotional states by changing the blood fluid,
peristaltic contractions and hydrochloric acid secretions. The relation between the
emotions and cardiac frequency, electrical conductivity in the skin (related to
perspiration), muscular tension, temperature in the face and hands and respiratory
frequency has been determined form Albert Ax (1967). It was discovered that
emotional reactions to danger caused similar reactions to adrenaline action, and that
insulting acts caused similar muscular, cardiac and respiratory reactions to the
adrenaline and noradrenaline actions. Gary Schowartz found that some emotional
reactions caused changes in facial muscular tension imperceptible to simple view but
which could be instrumentally measured. According to the above, it is possible that
emotional reactions are related to the physiological changes required for the individual
to face the situation or for the transmission of messages or response signals to other
individuals. The hypothesis that many of these changes are perhaps only
unconsciously perceptible by other individuals may also be considered.

Figure 42: Map of the physiological components of emotions.
Emotions have the following impact on cognition and behaviour:
1. Emotions organise us for action

The emotions regulate mental functioning, organising both thought and action. They
establish the goals that are priorities and organise us to undertake certain specific
actions.
On the other hand, the emotional centres of the brain receive and process the incoming
information before it reaches the centres responsible for decision-making and planning,
which, at the time this information is processed, has already been oriented in a specific
way to draw from information provided by the emotional centres.
Essentially, emotion establishes the problems and cognition resolves them. Emotion
has essentially to do with motivation and action, while cognition is related to
knowledge and analyses the situation and the action to be taken.
Emotion is necessary to tell us what it is that is affecting us and to establish a goal in

such a way that it can be achieved. Cognition is necessary to help make sense of
experiences, as is reason to help to imagine the best way to achieve a goal, or to
satisfy the interest in a specific cultural context.
2. Emotions are essentially adaptive

Emotions are not rational or irrational; they are adaptive. They are internal signals that
direct us to stay alive. Compared with cognition, the emotion makes up a biologically
older system, of quick and adaptive action; a system that improves survival.

One of the most important functions of emotions is that which connects the individual's
biological nature with the world in which it is immersed. The emotions respond quickly
to things that are really about survival.
The emotions regulate attention, control the environment, search for events that
are relevant to adaptation and alert the conscience when these things occur (fear
alerts to danger, nausea drives us away from something rotten, compassion enables
an individual to respond to someone's pain, etc.) the different emotions alert us to
different things differently to serve different functions.
The adaptive function of the emotions is better understood when observing the
differences between positive and negative emotions.
Emotions experienced positively are crucial at the time behaviour is motivated.

Interest and curiosity are essential elements that promote exploration and produce new
behaviours, and are an essential aspect of the adaptive organism as being oriented
towards growth.
Joy and happiness are the results of both contact with another person towards whom
we feel connected, as well as the sensation of effectiveness and also serves to keep
the organism proactive, continually seeking connections and rules about the situation.
In contrast with the small number of basic positive emotions, there is an extensive
repertoire of negative emotions. Feelings such as fear, desperation, anger, shame and
repugnance affect us but no one would confuse these with each other. It seems that
humans have developed more and more negative emotions and a greater
differentiation between them as a way to support the search for survival. Each one
possesses its own action tendency (anger leads the individual to face antagonists and,
in extreme situations, to want to strike or attack the object of fury; the feeling of fear
drives the individual away from that which is threatening and causes the desire to
escape from it; sadness makes the individual close up and want to hide and be isolated
from the world).
3. Emotions influence cognition and memory

The emotions are a powerful influence on reason and may improve or worsen
reasoning or decision-making.
The emotional system provides gut instincts about the things that guide us. These
feelings also help to focus attention on possibilities that may turn out to be negative,
and provide immediate responses that lead us away from them. Rational analysis
follows, although only after effective evaluation has quickly reduced the options. This
increases precision and efficiency in decision-making.
People who try to intellectually resolve decision-making conflicts without heeding the
preferences they feel go around in circles weighing up the pros and cons, unable to
make a decision. Paying attention to gut instincts gives substance to possible options,
and these must be attended to in order to make the decision or for a new meaning to
emerge.

The emotional brain or limbic systems criterion that associates different parts of the
brain system (amygdala, hippocampus, hypothalamus and thalamus) also works
together to give a new dimension to intelligence from the emotional brain's point of
view. It is quite different from the so-called rational brain or intelligent brain.
Figure 43: Emotional brain or limbic systems criterion
4. Emotions motivate
Emotional responses, through their physiological components and their tendency
towards action, prepare and motivate people to be able to relate to the events that elicit
these emotions. The emotions also often try to obtain their own rewards by attempting

to motivate behaviours that increase the likelihood of others occurring that cause
certain emotional states.
Fear, as previously mentioned, is aversive and motivates towards escape or evasion at

the same time as providing the motor and physiological responses that enable us to
make the escape. Joy is pleasurable, and at the same time provides physiological and
motor responses that enable people to open up and become closer.
However, living in the moment, paying attention to the sensations felt about what is
good for us is not the same as living for the moment, which simply implies doing what
makes us feel good, without considering the consequences.
5. Emotions inform
The emotions provide information about reactions when facing situations. The
emotions are there to help decide how to react in situations.
For example, it is important to be aware of the feeling of fear so that we can be ready
to flee. Thus, a person who feels anger and expresses I am angry because this is not
fair begins to clarify that they feel unjustly treated and that they would like a fair
exchange.
6. Emotions communicate
The emotions provide information to others about our intentions and our availability to
act. In interpersonal relations, the constant sending and receiving of emotional states,
especially through facial expressions, largely informs and regulates the interaction.
If people change emotionally on the inside, this will manifest on the outside and others
will treat them differently. In effect, if they feel safer, they will be able to express
themselves more assertively, and if they are less frightened, they will be more capable
of working and will be treated with greater respect.
4.3.5 Dynamics of emotional responses
The affective reaction (emotional response) is in fact not static. It is displayed over a
characteristic temporary course of time, in accordance with whether or not the stimulus
that has elicited it is present or absent, and depending on the length of time this
stimulus has been or was acting. The dynamics of this emotional response can be
explained using two complementary models: the opponent-process theory of motivation
(Solomon & Corbit, 1974; Solomon, 1977) and the excitation transfer theory (Zillmann,
1978, 1983).
4.3.5.1 The opponent-process theory of motivation

This theory considers that the organism responds using two affective processes when
faced with an instigator: Process A and Process B.
Process A is the primary and natural reaction response to this instigator (UR in the
case of a US).
It has the following characteristics:

a. Short reaction time.

b. It quickly establishes its maximum extent.
c. Once the instigator has disappeared, it declines quickly.
Every Process A is followed by an opponent Process B. Process B has opposite
affective qualities to that of Process A, as follows:
d. A long reaction time or latent period.

e. Slow to establish its maximum extent.
f. Slow to decline once the instigator and its primary reaction (Process A) has
ceased.
The emotion experienced consists of adding Process A and Process B and it

varies depending on the presence or absence of the instigator of the emotional
reaction and the evolution of Processes A and B.
The more repetitive the presence of the emotional reaction instigator, the weaker
Process A becomes while Process B gets stronger and lasts longer. Process B gets
stronger with use and weaker when not in use.
The dynamics of the affective reaction would be as follows: when a stimulus appears,
there is a fast and intense triggering of Process A (e.g. agitation). Process B (calm)
appears to be slower. Outwardly, State A (Phase 1) is displayed. If the stimulus that
produces the reaction is maintained, the intensity of Process A is reduced and the
Adaptation Phase (Phase 2) begins. When both Processes A and B (agitation and
calm) have the same strength, the Stability Phase (Phase 3) begins.
When the instigator disappears, Process A fades away quickly, while Process B
continues to be maintained for some time, which is when State B is really displayed
outwardly, the opposite to State A (Phase 4). In this example, it is the time when, after
the situation that produced the intense agitation is over, the organism is in a state of
hypoactivation and general lethargy. If the emotional instigator does not return,
everything goes back to normal little by little (Phase 5).
The standard pattern of affective dynamics has 5 phases:
limit of A
adaptation of A
stable level of A
limit of B and
decline of B and return to the baseline
As we have shown, Process B is strengthened by the repeated exposure to the

situation that produces the primary affective reaction, while Process A remains
unaltered. This explains the fact that when there is regular exposure to the emotional

instigator, the organism starts to show increasingly more typical reactions of State B or,
in other words, the affective reaction becomes accustomed, so that to produce an
emotion of the same intensity, stimuli of a greater magnitude is required.
The opponent-process theory of motivation consists of a homeostatic explanation,

which describes a useful process to maintain emotional stability and to avoid excessive
deviations. For this, when there are stimuli that provoke an emotion which generate a
reaction that makes the organism divert from neutrality (Process A), other responses
from the opponent sign (Process B) that re-establish the balance are required.
The emotions that appear outwardly at all times are the result of the combination of the
two processes. Nevertheless, not all emotions produce this opponent affective
response pattern. In this regard, Mauro (1988) produced happiness and sadness
through hypnosis, but the opponent phenomena were only noted in happiness, and not
in sadness.
According to Solomon (1980), the emotional reaction of some stimuli consists of

Process B, which is why there are no opponent processes. Therefore, the problem lies
in identifying which type of emotions produce Process A and which produce only
Process B. In humans, it seems that the emotions that cause autonomic activation
(such as happiness and anger) are those that produce opponent processes. Mauro
(1992) interprets that it is not that a weak Process A does not produce an opponent
process, but that it takes longer to vanish when the stimulus that elicited it has
disappeared, so the effects of the opponent process are compensated and there is no
affective post-reaction. The fact that it takes longer to vanish may be because the
effects that it produces are not so immediate, or because the stimulus that elicits it
does not suddenly disappear. In this sense, sadness may be a reaction to a loss, and
this loss does not disappear. A lot of the stimuli that produce emotional reactions are
psychological stimuli, rather than physical, which makes it difficult to determine when
they disappear for the individual. This would explain the fact that in these cases,
Process A would disappear slowly.
4.3.5.2 The excitation transfer theory

The excitation transfer theory (Zillmann, 1978, 1983) establishes that a determined
event produces sympathetic activation, which is generalised to different physiological
functions and diffused for a certain time. If other environmental events appear during
this period of time, the individual attributes the arousal experienced to these stimuli and
the response to the same may be increased. The following would be logical: a stimulus
produces a diffuse activation state that fades little by little. If a stimulus (supposedly
relevant) appears before it disappears which produces a determined activation, this
response is added to the effects of the previous arousal that was fading, so that the
individual learns to react more intensely to this new stimulus. This would explain why
the activation is important in aggressive response generation, the relation between this
and sexual crimes, increase in emotional response due to physical exercise, etc.
The basic postulates of this model are the following:

a) Emotional stimuli create a diffuse sympathetic activation state.

b) When two stimuli occur at the same time or close in time to each other, the effects
on this arousal are added together.
c) The individual interprets the arousal produced by the addition of both stimuli as
being the responsibility of the most prominent one (generally the latter).
The first point is a controversial issue as there are studies that defend the hypothesis of
specificity in the physiological reaction of emotion. With regard to arousal summation,
there are also contradictory research studies, but this could be precisely because not
all emotional stimuli produce the same physiological reaction. It would only make sense
to add the effects together if the same response was produced.
According to Mauro (1992), the opponent-process theory of motivation and the

excitation transfer theory are corollaries of a general model of affective dynamics
based on three general principles: homeostasis, affective consolidation according to
the regulation of opponent process and temporary variation of the emotional response.
In terms of homeostasis, the opponent-process theory of motivation is based on the

compensatory effects of the two branches of the autonomic nervous system. The
sympathetic reaction is responsible for the primary emotional reaction, while the
parasympathetic would notice the opponent process. There are two opponent
reactions, one immediate, nervous reaction, and the other slow, which is hormonal.
With regard to affective consolidation, the effects of two emotional stimuli may produce
an increase or decrease in the initial response, depending on whether the same
responses are produced and whether these are compatible.
In that way, we can predict the emotional response in different situations, taking both
theories into account (opponent process and excitation transfer), in accordance with
the moment (opponent process) as Process A or Process B can occur, and with the
stimuli that occur which produce an additional activation (excitation transfer).
Mauro (1988) demonstrated that when a CS was presented related to the state of
happiness at the same time as another associated with sadness, the effects
compensated each other and the subject did not show any emotional reaction. In the
same way, the effects can be added together if they both have the same direction.
4.3.6 Bodily expressions of emotions

A person's emotions are usually interpreted by bodily expressions, tone of voice and
the face. Psychologists study non-verbal communication and its emotional implication.
Technological advances have enabled greater precision in linking emotions to
registered facial expressions. Some gestures are determined by cultural factors and
others are common to everyone. Facial expressions are not only used to communicate
emotions, they also increase the emotion felt and send signals to the body so that a
response is consequently given.

Humans have 24 different muscles in the face. Some specific emotions or others are
expressed depending on how they are used. There are different smiles that express
different levels of joy. This helps us to express what we feel, which is often difficult to
explain with words. This is another form of communicating socially and feeling
integrated in a social group. It must be taken into account that the human being is a
classic example of a social animal.
The different facial expressions are universal; there is a similar language between
different cultures. Blind or deaf children can be observed to show the emotions they
experience in a very similar way to other people; they have the same facial
expressions. There are possibly some genetic and hereditary bases, given that a child
who cannot see cannot imitate the facial expressions of others. The expressions also
vary a little depending on culture, gender, country of origin, etc. Women are more
sensitive to perceiving facial expressions or emotional signals and this sensitivity
increases with age. Another example is oriental faces, especially Japanese, which
although they are quite inexpressive, they better express emotions at an intimate level.
Figure 44: Various facial expressions created by emotional states like affection,
anger, anxiety, despair, rage, disappointment, fear, happiness and love.
Facial expressions also affect the people we look at and alter their behaviour. When
observing someone who is crying, we become sad or serious and may also begin to

cry. On the other hand, anger, joy and sadness are quite often identified in the people
we observe. However, fear, surprise and aversion are less easily identified.
The table below presents a possible mapping of facial movements to affective

dimensions proposed by Smith and Scott (1997).
Table 12: A possible mapping between the facial actions and their meaning for
emotional state from by Smith and Scott. Smith and Scott (1997)
The aforementioned proposals are illustrate in the scheme below form Russell (2003).

Figure 45: Emotion and facial actions. Russell (2003)
Emotions also possess specific behavioural components apart from facial expressions,
which is the way they are externally expressed. This can be controlled to a certain
extent, depending on family and cultural learning. The specific behavioural components
humans use to externally express their emotions are:
Actions and gestures.
Distance between people.
Non-linguistic components of verbal expressions (non-verbal communication).
The figure that follows shows some examples of actions and gestures provoking
emotional states.
Figure 46: Different gestures that provoke emotional statements.
4.3.7 Measurement methods for emotions
4.3.7.1 Models and methods for measuring emotions

Recently, emotions have gained increasing attention concerning the way they can
affect human behaviour and decision making. The feeling of emotions can lead to
changes in the way people think and feel and are also important in forming memories
(Petty and Cacioppo 1986; Cahill and McGaugh 1995). Measurement of emotions
though, as well as correlating them with human behaviour, is complex and the

experience is limited. However, several methods for measuring, modelling and

simulating emotions have been applied over time and a review of these methods and
the experiences gained from past research activities can build a foundation for further
research in the area of understanding and simulating emotions.
One reason for the lack of research success in the area of emotions is the difficulty in
measuring and reconciling the many physiological and psychological responses that
are associated with the experience of emotion. Findings from research in other
contexts suggest that it is necessary to use multiple measures of emotion, including
self-reports by the research subjects themselves. Nine overall approaches for
measuring emotions are generally used:
1. Self report,
a. Verbal Self Report (VeSR)
b. Visual Self Report (ViSR)
2. Movement-to-movement ratings
3. Autonomic measures
4. Facial expressions
5. Electrodermal reaction (EDR)
6. Cardiovascular Responses
7. Eye-tracking
8. Speech recognition
9. Brain imaging.
In this chapter we will describe all the aforementioned categories, as well as the exact
methods that belong to each of these categories.
1. Self report
Self report is the most commonly used method for measuring emotions. The main
reason for this is that self-reporting involves relatively smaller expenses and does not
demand skills that are uncommon to find among consumer researchers. Scholars have
found that the use of scenarios in experimental research can generate discrete
emotional response (eg. Gopinath and Bagozzi, 1999; Roseman, 1991). However, self
report is difficult to apply in measuring emotions since emotions are often unconscious
or simply hard to define, causing bias to the reported emotions. Thus, even if self-
reporting is an obvious way of measuring emotion, this is not as straightforward as it
sounds. According to Russell (1974) and many others, emotions are multi-
dimensional, including the dimensions of pleasure (from pleasant to unpleasant) and

arousal (from sleepy to frenetic excitement), as they have been presented in previous
Chapter of the current Deliverable.
In self-reporting methods the challenge is choosing or developing a valid and reliable

measurement scale. Scales can be either theoretically driven or empirically driven.
Using an already developed scale can be problematic since the emotion-words on this
scale are not adjusted to the particular research focus. Nevertheless, using an already
developed scale, can lead to accepted and comparable results.
Additionally, a scale for measuring emotions can be either verbal or visual, the Verbal
self report (VeSR) and the Visual self report (VeSR) respectively.
2. Verbal self report (VeSR)
VeSR is one of the most expanded methods of measuring emotions. This method can
be applied in experimental designs, in questionnaires and in interviews. Research is
conducted using open ended questions or a battery of emotion items measured by
semantic differential or Likert scales. Approaches to emotions and thus to scales can
be empirically oriented (Edell and Burke, 1987; Holbrook and Batra, 1987; Richins,
1997; Schoefer and Diamantopolos, 2008) or theoretically oriented (Izard, 1977;
Mehrabian and Russell, 1974; Plutchik, 1980; Watson and Tellegen, 1985). The three
most common used measures of emotions in VeSR are:
o Emotions Profile Index (EPI).
o Differential Emotions Scale (DES).
o PAD-model (Pleasure-arousal-dominance).
o PANAS-model (Positive And Negative Affect Schedule).
o NERS score (Net Emotional Response Strength).
o ESRE (Emotion During Service Recovery Encounters) scale.
o POMS (Profile of Mood States)
The Emotions Profile Index (EPI) is a personality test designed to yield information
about certain basic personality traits and personality conflicts in an individuals life. It
was developed from Plutchik & Kellerman in 1974 and it reveals problematic areas and
provides the examiner with insights into the nature of the difficulties.
A circular profile is used to display the relative strengths of the basic personality traits.
This profile allows the examiner to compare these traits, to see their relative
importance for the individual, and to identify major conflicts between traits. The circular
profile thus provides a concise, dynamic description of some of the major forces acting
in the personality.
The EPI contains 62 forced-choice emotion descriptor pairs. It has been used in a
variety of contexts including clinics, mental hospitals, vocational guidance centres,

classroom demonstrations, counselling and psychotherapy, diagnostic evaluations and

research. The basic personality traits measured by the EPI may be described by the
following sets of bipolar terms:
o Timid VS Aggressive
o Trustful VS Distrustful
o Controlled VS Dyscontrol
o Gregarious VS Depressed
The person taking the test is simply asked to indicate which of two paired words is
more descriptive of him/her. The responses are transformed into scales representing
each of the eight Plutchiks primary emotions.
EPI is designed for adolescents and adults; the test provides results based on 1,000
normal men and women. Data are also given for special groups. The test takes 10 15
minutes to complete, and it can be hand scored in less than 5 minutes.
Another VeSR method for capturing emotions is Izards (1977) Differential Emotions
Theory (DES). DES is available in four forms, from which DES-II is the most used one.
DES II allows assessment of emotions experienced over extended time periods. DESII
also allows the researcher "to determine how often one experiences each of the
fundamental emotions" in a specific context or over a specified time period. It contains
30 adjective items that divided in three groups measures each of the ten fundamental
emotions, (Richins, 1997).
Mehrabien and Russells (1974) PAD-model (Pleasure-arousal-dominance) is founded

in environmental psychology and measures the individuals emotional responses to
their environment. The idea of the PAD model is that all emotions have two dimensions
and that all emotions fit into one of the three categories: Pleasure, arousal and
dominance. PAD contains 18 semantic differential items, six for each P, A and D.
PAD scale range from -1 to +1:
angry (-.51, .59, .25),
bored (-.65, -.62, -.33),
curious (.22, .62, -.01),
dignified (.55, .22, .61),
elated (.50, .42, .23),
hungry (-.44, .14, -.21),
inhibited (-.54, -.04, -.41),

loved (.87, .54, -.18),
puzzled (-.41, .48, -.33),
sleepy (.20, -.70, -.44),
unconcerned (-.13, -.41, .08),
violent (-.50, .62, .38).
Richins (1997) points out that the PAD-model of emotions validity for the measurement
of emotions in interpersonal interaction, cannot be assumed. Furthermore PAD
measures underlying dimensions, which means it cannot measure discrete emotions.
Critique specifically linked to PAD is that the scale has very low discriminate validity
(Groeppel-Klein, 2005).
One other model that should be mentioned is the PANAS-model (Positive And
Negative Affect Schedule). PANAS is a 20-item measure designed to assess Positive
Affect and Negative and it was explicitly made from Watson et al. (Watson & Tellegen,
1985; Watson, Clark & Tellegen, 1988; Watson & Clark, 1992). It measures positive
and negative affect as two mutually independent dimensions. Originally it was designed
to measure a mood scale but has been applied to measure emotional state in other
domains also, like product and service satisfaction by Mano and Oliver (1993) and
Dub and Morgan (1988) and post-purchase affect by Mooradian and Oliver (1997)
and finally negative effect in advertising by Huang (1997).
PANAS uses a 5-point emotional scale:
1 very slightly or not at all,
2 a little,
3 moderately,
4 quite a bit and
5 very much.
The figure below illustrates an example of the implementation of PANAS methodology,

from Crawford and Henry, 2004.

Figure 47: Implementation of PANAS methodology, from Crawford and Henry.
Finally, another methodology for emotional statement is the Profile of Mood States
(POMS) model of mood that was developed in 1971 and has been widely used for
measuring distinct mood states in medical patients, athletes, and normal population in
general. The POMS original scale contains 65 self-report items using the 5-point Likert
Scale. Six factors can derived from POMS, namely:
tension-anxiety,
depression-dejection,
anger-hostility,
fatigue-inertia,

vigor-activity, and
confusion-bewilderment
Participants can choose from 0 (not at all) to 4 (extremely). The test takes
approximately 3 to 7 minutes for healthy participants, and longer for the physically ill.
In spite of its popularity, the POMS is simply a preliminary scale (i.e., exploratory in
nature) and has never been validated since. Without validation, the results generated
from the numerous studies using the POMS were not persuasive.
The advantages of using VeSRs (Verbal Self Report) are that they are relatively
simple, cheap and quick to use, however there are some important limitations. They
involve a long list of emotion adjectives and the rating can cause fatigue in the
respondents and may damage the reliability. Furthermore verbal self report involves
cognitive processing, which may distort the original emotional reaction and lead to
biased results.
Generally, there are some important considerations to make when using self-report. It
is important to consider whether self-reporting is relevant for the purpose of the study,
and if found to be, how the scale should be constructed. Scale construction can be time
consuming and difficult. Sometimes existing scales can help but there are still aspects
to consider. The scale should be suitable for the context; it should include relevant
emotion words that are familiar to the participants and it should be noted that emotions
in a consumer context are likely to be less extreme than in general. Scales can be
unipolar or bipolar and if bipolar scales are found to be the most useful it is important
that these poles are in fact opposites. The scales should also be as short as possible,
especially if the research is not performed in a laboratory setting.
VeSR is also the most commonly used method for measuring future-oriented emotions.
3. Visual self report (ViSR)
ViSR has a lot in common with the VeSR techniques. In this case instead of using
emotion words the emotions are represented by cartoon-like figures or smileys. The
main approaches used are:
Self Assessment Manikin (SAM),
AdSAM
PrEmo
SAM (Lang, 1980), is the most frequently used ViSR instrument. It is based on
Mehrabian and Russels PAD-dimention (Pleasure-Arousal-Dominance). Here, instead
of rating a set of emotions, the respondents indicate which figures best represent their
emotional states. AdSAM (Morris et al., 2002), is a further development of SAM where
232 emotion words were scored on SAM and plotted in a two-dimensional space with
pleasure and arousal on the two axes. SAM has shown problems related to the
dominance dimension, which respondents seem to experience difficulties in relating to.

Furthermore the arousal pictures might have an interpretation issue related to reading
the figure.
Figure 48: SAM model illustration. (Lang, 1980)
PrEmo (Desmet, 2002), is the most recent visual measurement instrument and it
includes 14 animations of 1-2 seconds. Each animation represents a specific emotion.
Respondents indicate how strongly the target stimulus makes them experience each of
the 14 emotions represented by the puppets. With PrEmo more than one specific
emotion can be registered making it suitable to study mixed emotions. ProEmo further
makes it possible to measure more separate emotional aspects than SAM and due to
the animations contra the still drawings this can help the interpretation of the drawings.
Figure 49: ProEmo animations. (Desmet, 2002)

The advantages of using visual self report instead of verbal self report is that it is faster
and less boring for the participant, than verbal self report. Visual data are also suitable
for cross cultural research and research with children (Morris, 1995; Poels & DeWitte,
2006). However, as pointed out by Poels and DeWitte (2006), ViSR does not solve the
issue concerning the cognitive bias. It is still the perception of the emotion and not the
actual emotion that is being measured. The significance of this bias tendency depends
on the objective of the research and should only be considered a problem on validity
when measuring lower-order emotions.
4. Movement-to-movement ratings
Movement-to-movement ratings are mainly used for evaluation of the emotional state
for a specific stimulus. This method is mainly used in the advertising and it involves the
movement of a pencil on a piece of paper while viewing a commercial a so-called
warmth monitor. Movements are supposed to indicate how warm the respondents
feelings are at any given time, shifting hand to the left indicates boredom or
dissatisfaction, and shifting hand to the right indicates a positive emotional connection
(Aaker, Stayman, and Hagerty, 1986). Movement-to-movement ratings were used also
from Rossiter & Thornton (2004) to measure fear relief.
Figure 50: Overview of warmth monitor methodology.
Movement-to-movement ratings offer immediate and continuous measures of

emotional response. It is cheap, and easy to understand and use, however it is
important to notice that it only measures one dimension.
5. Autonomic measures
Psychological measurement has been used in emotional research from the 1920s,
mostly applied to measuring response to advertising (Bagozzi, 1991).
Autonomic measures rely on bodily reactions that are partially beyond an individuals
control. It therefore overcomes the cognitive bias linked to self report. Most autonomic
measures are conducted in a laboratory setting, which it is often criticised for, since it is
considered out of social context. However lab experiments are commonly used in both
psychology and in cognitive neuroscience as well as consumer research. The
advantages of using lab experiments are that you can control for different noise
parameters that would otherwise be difficult to detect. The technological development

have made it possible to apply some of the methods in a consumer context e.g. in
grocery stores.
6. Facial expressions
Measurement of facial expressions is commonly used in psychology. To applying this

method specific professional competences is required. All humans are capable to some
extent to read other peoples facial expressions and use them in social interaction,
however using the method for doing scholarly research calls for a more systematic and
scientific approach to reading facial expressions.
Two primary methods are used for the measurement of facial expressions. First of all
there is the Facial Action Coding System (FACS). FACS codes visible facial muscle
movement and links to specific emotions, see Figure 50. Using FACS, human coders
can manually code nearly any anatomically possible facial expression, deconstructing it
into the specific Action Units (AU) and their temporal segments that produced the
expression. As AUs are independent of any interpretation, they can be used for any
higher order decision making process including recognition of basic emotions, or pre-
programmed commands for an ambient intelligent environment. The FACS manual is
over 1000 pages in length and provides the AUs, as well as Dr. Ekman's interpretation
of their meaning.
Figure 51: Muscles for facial expressions. (Ekman, FACS)
FACS defines 32 AUs, which are a contraction or relaxation of one or more muscles. It
also defines a number of Action Descriptors, which differ from AUs in that the authors
of FACS have not specified the muscular basis for the action and have not
distinguished specific behaviours as precisely as they have for the AUs.
The second method is Facial electromyography (EMG). EMG technique measures

muscle activity by detecting and amplifying the tiny electrical impulses that are
generated by muscle fibbers when they contract. With EMG we can measure facial
expressions more precise than FACS since it can measure facial muscle activity even
though there are no changes in facial expressions.
Two areas where facial EMG techniques have been used are in Advertising Research
(Boll, Lang and Potter, 2001) and in Gaming (Ravaja Turpeinen Saari Puttonen
Keltikangas-Jrvinen, 2008)

Apart from the theoretical methods that the need of experts to validate the human facial
expressions, there are also two computerised methods, the Facial Expression Analysis
Tool (FEAT) and Facial Action Composing Environment (FACE), developed at the
University of Geneva in 1994.
The Facial Expression Analysis Tool (FEAT) (Wehrle, 1992/1996) is a connectionist

expert system that uses fuzzy rules, acquired from a FACS expert, to automatically
measure facial expressions. In this way, the FACS expertise is automatically
transformed into a network structure. The resulting network is able to do the
classification task, using FACS as the coding language. With FEAT we can precisely
analyze the dynamics of facial behaviour, including intensities and asymmetries.
Facial Action Composing Environment (FACE), allows one to synthesise in real time
three-dimensional animated facial expressions, including eye movements, on the basis
of the Facial Action Coding System
7. Electrodermal reaction
Electrodermal reaction (EDR) or Skin conductance measures activation of the

autonomic nervous system. Activation of the autonomic nervous system indicates
arousal (Ravaja, 2004). The EDR measure indicates the electrical conductance of the
skin related to the level of sweat in the eccrine sweat glands which is involved in
emotion-evoked sweating and is conducted using electrodes. Skin conductance can be
measured on the whole body but are more sensitive on the palms and soles of the feet
(Dawson, Schell and Filion, 2000).
EDR is a sensitive measure of arousal. However it requires a lot of experience and

sensitive equipment. It is therefore best carried out by experts. Furthermore EDR only
measures the occurrence of arousal not the valence of the arousal, which can be both
positive and negative. Another problem with using EDR are the individual variation and
situational factors such as fatigue, medication etc. (Hopkins and Fletcher, 1994), which
makes it hard to know what you are measuring. Furthermore measuring arousal may
not be the only purpose of the research. The equipment can be carried in a bag and
the study can thus be performed in various contexts.
EDR is considered interesting and useful to measure arousal but since it only
measures one dimension of the emotional experience, and the role of arousal in the
emotional experience is not yet fully understood, it is necessary to combine it with other
measures. Furthermore it should be considered whether it is worth the effort, since it is
time-consuming to perform the EDR experiments.
The use of EDR in a future-oriented perspective is to my best knowledge non-existing

in consumer research but it has been used in conjunction with heart rate measures in
psychology in connection with risky choice (Crone et al., 2004).
Some Typical Values of electrodermal reactions (EDR) components:
Skin Conductance level (SCL)= 2-20 S

Change in SCL= 1-3 S
Frequency of NS-SCRs= 1-3 per min
ER-SCR amplitude= 0.2-1.0 S
ER-SCR latency =1-3 S
ER-SCR rise time=1-3 S
ER-SCR half recovery time= 2-10 S
ER-SCR habituation =0.01-0.05 S per trial
8. Cardiovascular Responses
Heart rate can be used as an indicator of various phenomena: Attention, arousal and
physical effort. Phasic changes (short term changes) in heart rate indicate attention
and tonic changes (short term changes) indicate arousal (Poels and DeWitte, 2006).
Heart rate allows measuring the valence of arousal (Greenwald, Cook and Lang,
1989). Heart rate can be operationalized as number of milliseconds since the previous
heart beat (Lang, 1990, Poels and Dewitte, 2006)
Generally heart rate measures are considered to be a valid measure of arousal,

valence or attention. However results need to be interpreted with caution since several
phenomena can occur at the same time, affecting the heart rate in different direction
thus causing bias (Poels and DeWitte, 2006). Heart rate as indicator of emotional
response should not be used as single measurement method but could be combined
with.
9. Eye-tracking
Only few studies involving measurements of emotions with eye-movement have been
performed e.g. Treistman and Gregg (1979) examining eye-movement response to
catalogue advertisement measuring average viewing time as an indicator of high and
low performing ads. The concept of eye-tracking is linked to the measurement of
emotions in the sense that it measures attention and that attention is more attracted by
e.g. emotionally triggering images. Eye-tracking measures on attention can thus be an
indication of emotional reactions. However since eye-tracking is an unsure measure of
emotions it is not considered relevant to use it as a stand-alone method when studying
emotional aspects. I can however be useful to combine with other methods for some
problems.
Eye-tracking have mostly been applied to studies in advertising research where

attention play a large role. But can also be used in in-store decision-making (retail-
stores). It has been used to study attention toward health warnings in cigarette
advertisements among adolescents (Krugman et al. (1994) and to study attention wear-
out and the impact of repetition on advertising effectiveness (Pieters, Rosenbergen and
Wedel, 1999) and finally to measure the value of point-of-purchase marketing
(Chandon et Al., 2007)

10. Brain imaging
Brain imaging is a new method in consumer research. In this method emotions are
pointed out as an area of specific relevance. However the method is extremely
expensive, it requires expert knowledge and has severe technological limitations for
experimental designs. Furthermore knowledge within neuroscience is still relatively
young and therefore the complexity of the problems investigated must be relatively
simple. The use in consumer research is so far relatively limited and thus are the
examples of use related to measurement of emotions in consumer research. The most
commonly applied methods from neuroscience are the following:
Electroencephalography (EEG)
EEG is the oldest imaging method. It is also the least expensive and most widely
available brain imaging method and the data analysis is relatively less complicated
than for other brain imaging methods. It measures electronic activity on the outside of
the brain using scale electrodes and can only measure the outer areas of the brain,
which makes it unsuitable for measuring emotional aspects that are situated in the
limbic system in the mammalian brain. However it has a good temporal resolution
making it very suitable for some studies, especially if the focus is on when an action
occurs.).
Magnetoencephalography (MEG)
MEG measures magnetic fluctuations in the brain non-invasively. It has not been
applied much to consumer research and the equipment is not as common to find as
EEG. MEG is conducted using scale electrodes attached to the scalp like with EEG,
however to make this a better measure it is performed very sensitive to magnetic noise,
which means it must be conducted in a magnetic shield room (Hmlainen et al.,
1993). MEG has a good temporal resolution, like EEG, but a limited special resolution
although better than EEG. Considering only the results it is assessed to be better than
EEG, however, the method is also more expensive and more complex to analyse.
Furthermore it is less available.
Positron emission topography (PET)
PET is a nuclear medical imaging technique which produces a three dimensional

image or map of functional processes in the brain. The scanning is performed in a full-
body scanner. It measures blood flow in the brain using positron emissions after
radioactive contrast injection. Contrary to EEG and MEG, PET has a good spatial
resolution but this is on the expenses of the temporal resolution. PET is relatively costly
and data are complex to analyse. Most importantly it is invasive due to the injection of
radioactive contrast.
To measure future-oriented emotions PET has been used in neuroscience to measure

anticipated anxiety (Simpson et al., 2000) showing that at least this aspect of future-
oriented emotions can be measured by the technique.
Functional Magnetic Resonance Imaging (fMRI)

FMRI is a technique used for estimating neural activity non-invasively (unlike PET) and
with relatively good spatial and temporal resolution. fMRI has a better spatial resolution
than EEG and MEG but worse temporal resolution. The spatial resolution however is
much better than with PET. Specifically it measures the blood oxygen level dependent
(BOLD) signal that strongly correlates with neural activity (Yoon et al., 2006). Thus
fMRI can be used to estimate neural activity in active parts of the brain. The use of
fMRI in research has increased over the last twenty years (Jonides and Nee, 2006).
When using fMRI it is only possible to image slices of the brain, which means that you
only see part of the brain in each scan and you cannot see if there is activity in other
areas, therefore fMRI is not suitable for exploratory research. It is necessary to use
knowledge from cognitive neuroscience to build hypothesis on which areas will be
activated as well as construct a model, thus knowledge in neuroscience directs the
possible problems to be investigated.
fMRI is the most commonly used method for measuring future-oriented emotions in
neuroscience. It has to my best knowledge not yet been applied in consumer research,
however to study decision-making fMRI studies have contributed to the study of both
prospective outcomes (Kahn et al., 2002) and anticipatory affect in choice (Knutson
and Greer, 2008)
Another existent metric for identifying emotions is the Emotion recognition from speech.
Mylonakis et al. (2008) present a number of leading edge audio and video processing
technologies, which can be exploited to build robust ambient assisted living
applications for elderly groups and discuss approaches to emotion recognition from
speech. The problem of speech-based emotion detection lies in partitioning the speech
signal to homogenous segments, each conveying distinct emotion from a predefined
set of emotional categories, such as anger, fear, frustration, sadness, surprise and joy.
Features indicative of emotional state that are used in emotion detection algorithms
include
- pitch (tone),
- energy (loudness), contour statistics, spectral shape descriptors,
- speech rate,
- emotional keywords.
The most typical measurements for emotional and cognitive states are presented in the
table below, showing physiological variables and behavioural indicators.
Physiological Instrument Behavior related indicators

measurement
Electrodermal Electrodes Novelty; Intensity; Emotional
activity content,
and significance

Physiological Instrument Behavior related indicators

measurement
Heart rate/period Electrocardiograph Fear and anger, intensive attention
Blood pressure Sphygmomanometer Information processing load and

engagement in problem-solving
tasks; Anger, Stress
Blood volume Plethysmograph new or unexpected stimuli; fearful, or

anxious startled,
Pupillary Eye tracker fatigue, workload, novelty and

response emotional valence of the stimuli
Eye blink Electro-oculogram, Eye task demands, negative states, such

tracker as nervousness, emotional stress
and fatigue
Brain activity Electroencephalography relaxed states (alpha); novelty,

(alpha, beta, (EEG) complexity, and unexpectedness, as
delta, theta, well as during emotional excitement
gamma waves) and anxiety (alpha blocking); active,
busy or anxious thinking and active
concentration (beta); higher mental
activity: including perception,
problem solving, fear, and
consciousness (gamma)
Facial muscular Electromyogram emotional states

activity
Table 13: Physiological variables and behavioral indicators (Backs and Boucsein
2000)

4.3.8 Six basic emotions as cognitive and behavioural moderators
While some current primary theoretical arguments focus on whether or not basic
emotions exist and if their recognition is universal (and this is a controversial fact, as
we have revealed), it is clear that there are certain affective responses patterns that are
distinctive and generalised, and which demonstrate a series of common characteristics
in all human beings. These are the emotions of happiness, sadness, anger, surprise,
fear and disgust. We can even argue that they are characterised by a series of actual
physiological or motor reactions, as well as by the facilitation of specific behaviours that
may become adaptative.
In this section, we shall review some of the main characteristics of these emotions,
considered to be distinctive by majority consensus.
Happiness
Characteristics Happiness favours the reception and positive interpretation of the

various environmental stimuli. It is not fleeting in the same way as
pleasure is. It is intended to be a lasting emotional stability
(Delgado, 1992).
Instigators/ - Accomplishment, successful achievement of a person's goals.

Causes -Congruence between what a person wants and what they have,
between expectations and current conditions and comparison with
others (Michalos, 1986)
Physiological -Increase in activity in the hypothalamus, septum and amygdaloid

Activity nucleus (Delgado, 1992)
-Increase in heart rate, even if the cardiovascular reactivity is lower
than in other emotions, such as anger and fear (Cacioppo et al.,
1993).
-Increase in respiratory rate (Averill, 1969)
Cognitive -Facilitates empathy, which favours the appearance of altruistic

Processes behaviour (Isen, Daubman & Norwicki, 1987)
Involved -Favours cognitive performance, problem-solving and creativity (Isen
& Daubman, 1984), as well as learning and memory (Nasby &
Yando, 1982).
-This relation, however, is paradoxical, as very intense states of
happiness may slow down execution and even lead to overseeing
an important element in problem-solving and it may interfere with
creative thinking (Izard, 1991).
Function -Increase in the capacity to enjoy different aspects of life.

-Causes positive attitudes within a person and towards others.
Favours altruism and empathy (Isen, Daubman & Norwicki, 1987).
-Establishes connections and favours interpersonal relations (Izard,
1991).
-Feelings of vigorousness, competence, importance and freedom
(Meadows, 1975)
-Favours cognitive processes and learning, mental flexibility and
curiosity (Langsdorf, Izard, Rayias & Hembree, 1983).
Subjective Pleasant, desirable state, sense of well-being.

Happiness
Experience -Sense of self-esteem and confidence (Averill & More, 1993).
Table 14: Happyness
Anger
Characteristics Anger is the emotional component of the AHA syndrome

(Aggression- Hostility- Anger). Hostility refers to the cognitive
component while aggression refers to the behavioural. This
syndrome is related to psycho-physiological disorders, particularly
cardiovascular alterations (Fdez-Abascal & Martn, 1994a, b).
Instigators/ Aversive stimulation, physical, sensory and cognitive (Berkowitz,

Causes 1990).
-Conditions that cause frustration (Miller, 1941), interruption of
motivated behaviour, unfair situations (Izard, 1991), or attacks
against moral values (Berkowitz, 1990).
-Extinction of the operant, especially in continuous reinforcement
programs (Skinner, 1953).
-Immobility (Watson, 1925), physical or psychological restriction
(Campos and Stenberg, 1981).
Physiological -Substantial neuronal and muscular activity (Tomkins, 1963).

Activity -Intense cardiovascular reactivity (elevation in the indices of the
heart rate, systolic and diastolic pressure) (Cacioppo et al., 1993).
Cognitive -Focusing attention on the external obstacles that prevent goal

Processes achievement or are responsible for frustration (Stein & Jewett,
Involved 1986).
-Obnubilation, inability or difficulty to execute cognitive processes
efficiently.
Function -Energy mobilisation for self-defence or attack reactions (Averill,

1982).
-Elimination of obstacles that prevent goal achievement and cause
frustration. Even though anger does not always lead to aggression
(Lemerise & Dodge, 1993), it can at least be used to discourage
undesired reactions of other subjects and even to avoid a
confrontation.
Subjective -Feeling of energy and impulsiveness, need for intense, immediate

Experience action (physical or verbal) to actively solve the problematic situation.
-It is considered to be an aversive, unpleasant and intense
experience. Related to impatience.
Table 15: Anger

Fear
Characteristics Fear and anxiety are possibly the most-researched emotions and an
array of intervention techniques have been developed on them from
all theoretical orientations in psychology. The pathological component
is anxiety disorders which are related to an excessive and
inappropriate reaction of fear. It is one of the reactions that produces
most mental, behavioural, emotional and psychosomatic disorders.
The difference between anxiety and fear can be specified whereby
the fear reaction is produced when there is real danger and the
reaction is in proportion to the danger, while anxiety is a
disproportionate, intense reaction to the supposed danger of the
stimulus (Bermdez & Luna, 1980; Miguel-Tobal, 1995).
Instigators/ -Potentially dangerous situations or CS that produce CR of fear.

Causes Conditioned stimuli to a fear reaction may be completely varied and be
of course lacking in objective danger.
-New and mysterious situations, especially in children (Schwartz, Izard
& Ansul, 1985).
-Visual abyss (Gibson and Walk, 1960) in children, as well as height
and depth (Campos, Hiatt, Ramsay, Henderson & Svejda, 1978).
-Secondary evaluation processes that interpret a situation as being
dangerous (Lazarus, 1977, 1991a).
-Pain and pain anticipation (Fernndez & Turk, 1992).
-Loss of support (Watson, 1920) and, in general, sudden change in
stimulus.
Physiological -Acceleration of heart rate, increase in its conductance and fluctuation
Activity (Cacioppo et al., 1993).
Cognitive -Primary evaluation: threat. Secondary valuation: lack of suitable

Processes coping strategies (Lazarus, 1993).
Involved -Reduction in the efficiency of cognitive processes, obnubilation.
- Focusing nearly exclusively on the perception of the feared stimulus.
Function -Providing escape responses or avoiding the dangerous situation.

Fear is the most significant emotional reaction in negative
reinforcement procedures (Pierce & Epling, 1995).
-As nearly exclusive attention is paid to the feared stimulus, the
organism reacts quickly to this.
-It activates a large amount of energy. The organism may execute
responses that are a lot more intense than in normal conditions. If the
reaction is excessive, efficiency is reduced, as per the relation
between activation and performance (Yerkes & Dodson, 1908).
Subjective -This is one of the most intense and unpleasant emotions.

Experience - It causes apprehension, unease and discomfort.
-Concern, suspicion for a person's own safety and/or health.
-Sense of loss of control.
Table 16: Fear

Sadness
-While it is traditionally considered to be one of the unpleasant

Characteristics emotions, it is not always negative (Stearns, 1993). There is a great
cultural variation and there are even some cultures that do not have
words to define it.
Instigators/ -Physical or psychological separation, loss or failure (Camras &

Causes Allison, 1989).
-Disappointment, particularly if hope in something has been dashed.
-Situations of defencelessness, lack of prediction and control.
According to Seligman (1975), sadness appears after an experience
which has caused fear as sadness is the opposite process to panic
and frantic activity.
-Lack of reinforced activities and adaptative behaviour (Lewinsohn,
1974).
-Chronic pain (Sternback, 1978, 1982).
Physiological -High, sustained neurological activity (Reeve, 1994).

Activity -Light increase in heart rate, blood pressure and electrical skin
resistance (Sinha, Lovallo & Parsons, 1992).
Cognitive - Evaluation of loss or damage that cannot be repaired (Stein &

Processes Levine, 1990).
Involved -Focusing attention on the internal consequences of the situation
(Stein & Jewett, 1986).
-Sadness can lead to a cognitive process characteristic of depression
(cognitive triad, depression schemas and errors when processing
information), which, according to Beck, are the main factors in the
development of this emotional disorder (Beck, 1983).
Function Cohesion with other people, particularly with those who are in the
same situation (Averill, 1979).
-Decrease in rhythm of activity. Evaluation of other aspects of life
which were not considered before the loss (Izard, 1991).
-Communicating to others that a person does not feel well and this
can lead to help from other people (Tomkins, 1963), as well as
appeasement of aggressive reactions by others (Savitsky & Sim,
1974), empathy, or altruistic behaviour (Huebner & Izard, 1988).
Subjective -Dejection, melancholy, despondency.

Experience -Loss of energy
Table 17: Sadness
Surprise
Characteristics -While it is traditionally considered to be one of the unpleasant

emotions, it is not always negative (Stearns, 1993). There is a great
cultural variation and there are even some cultures that do not have
the words to define it.

Surprise
Instigators/ Unexpected stimuli.

Causes
Physiological -Physiological pattern characteristic of orientation reflex: decrease in
Activity heart rate.
-Momentary increase in neuronal activity.
Cognitive -Working memory and attention used to process the new information.
Processes -General increase in cognitive activity.
Involved
Function -Give the suitable behavioural and emotional reaction when faced
with new situations. Eliminate the residual activity in the central
nervous system that may interfere with the appropriate reaction when
dealing with new situation demands (Izard, 1991).
-Facilitate attention processes, exploratory behaviour and interest in
the new situation (Berlyne, 1960).
-Direct the cognitive processes to the situation which has arisen
(Reeve, 1994).
Subjective -Transitory state. It appears quickly and lasts temporarily until it
Experience changes and becomes a different emotional reaction.
-Mind temporarily going blank.
-Undefined affective reaction, although pleasant. Situations that
cause surprise are not remembered as being as pleasant as
happiness, but more pleasant than emotions like anger, sadness,
disgust and fear (Izard, 1991).
-Sense of uncertainty as to what is going to happen.
Table 18: Surprise
Disgust
Characteristics Disgust is one of the emotional reactions where physiological feelings

are most evident. The majority of disgust reactions are created by
interoceptive conditioning.
It is related to behaviour disorders, such as anorexia and bulimia, but
it can be the main therapeutic component of treatments based on
aversive conditioning, such as the fast smoking technique (Becoa,
1985).
Instigators/ -Unpleasant stimuli (mainly chemicals) that are potentially dangerous

Causes or annoying.
-Aversively conditioned CS. US are usually related to smell or taste.
Physiological -Increase in gastrointestinal reaction

Activity -Muscular tension
Cognitive Unknown
Processes

Disgust
Involved
Function -Providing escape responses or avoiding situations that are

dangerous or potentially harmful to health. The stimuli are often
related to what a person intakes so the basic quality is scent or smell
(Darwin, 1872/1984), even though the CS (Conditioned Stimuli) may
be associated with any other perceptive modality (visual scenes,
sound, etc.)
Despite the fact that some authors restrict the disgust emotion to
stimuli related to spoiled food or food that is dangerous for health
(Rozin & Fallon, 1987), it is true that this emotional reaction
is also produced by any other stimuli which are not necessarily related
to gastrointestinal problems. The disgust reaction can even be
produced as a result of food that is nutritious and in good condition.
-Promote healthy, hygienic and adaptative habits (Reeve, 1994).
Subjective -Need to avoid or stay away from stimulus. If the stimulus is related to
Experience smell or taste, this leads to unpleasant gastrointenstinal agitation,
such as nausea.
Table 19: Disgust
4.3.9 Emotions and elderly

Older people experience stronger direct feelings of anxiety in the form of fear or
being scared, whereas younger people tend to experience more guilt-related anxiety
(Kennard, 2005). Older people also report fewer experiences of depression, hostility
or shyness. So, the overall structure of emotion between older and younger people
seems similar, but the strength the relationship and frequency of emotional experience
does differ.
In an experiment older and younger people were monitored physiologically. Heart rate,
skin temperature, respiration and other biological factors were recorded whilst a variety
of pictures, all designed to provoke an emotional response, were shown. Although
younger and older people all reacted to the stimuli fairly similarly, the reactions were
often less marked in the elderly and particularly so when stimuli designed to provoke
the emotion of disgust were shown. (Kennard, 2005)
It appears that older people do experience less intense emotions. It may also help
to explain why they find the emotional dysregulation of early stage Alzheimer's so
troubling. In some situations the elderly may show very little or no emotion where some
might be expected; in other situations they may be moved to tears in a display of
emotion that may seem out of place.
Variation in emotional pattern is something that needs to be considered in any

assessment of the elderly by younger people. Caution against using phrases such as,

'flattened affect', or 'emotionally labile', should be exercised in the realization that what
is being observed is, in fact, perfectly normal.
Other two studies carried out by Sullivan and Ruffman (2004) show that even healthy
elderly adults were poor at recognizing certain emotions. In study one, an emotion
face morphed to express a new emotion. The elderly were impaired when
recognizing anger and sadness, whereas no differences were found between the two
age groups in recognizing fear or happiness, or in a task requiring reasoning about
non-emotion stimuli. In study two, the elderly were impaired when judging which of two
faces was more angry, sad, or fearful, but they were not impaired when judging other
emotions or when judging which of two beakers was more full. The elderly were also
impaired when matching emotion sounds to angry, sad, and disgusted faces, but not to
other emotions and not when matching non-emotion (e.g., machine) sounds to
machines. Elderly deficits were independent of performance on a task requiring basic
face processing (gender recognition). Overall, the results provide support for an age-
related decline in the recognition of some emotions that is independent of changes in
perceptual abilities, processing speed, fluid IQ, basic face processing abilities, and
reasoning about non face stimuli. Recognition of emotion stimuli might be mediated by
regions of the brain that are independent from those associated with a more general
cognitive decline.
A study (Deptula et al.,1993) comparing the relation between negative mood states and
memory in young and elderly subjects showed that the elderly group, but not the young
group, consistently exhibited significant correlations between their performance on
verbal recall measures and their ratings of their anxiety, depression, and withdrawal;
i.e., within the elderly group, higher levels of negative affective states were
associated with poorer memory.
These findings indicate that aging modulates the relation between emotional state
and memory functions, and they are consistent with the hypothesis that the elderly
are more vulnerable than the young to the adverse effects of negative emotional states
on memory. Therefore, even in normal elderly individuals without diagnosable
psychopathology, negative affective states (such as anxiety and depression) may
interfere with memory functioning.
4.3.10 Emotions and disabled

McClure et al. (2007) indicates that individuals with intellectual disability can reliably
identify their own emotions. In general, people with mild to moderate intellectual
disability can also recognize and accurately label facial expressions in others and
improve this skill with training.
Moores (2001) review and the few studies on this topic published after his review
specifically conclude: The ability to discriminate pleasant versus unpleasant emotions
is equally accurate for individuals with and without intellectual disability. However,
intellectually disabled individuals have more difficulty labeling specific emotions.
There is evidence that individuals with intellectual disabilities utilize a limited repertoire
of coping strategies when emotionally aroused (Benson & Fuchs, 1999).

4.3.11 Emotions modelling approaches

As it is already mentioned in previous chapters of the current Deliverable, the idea that
emotions play a critical role in some cognitive processes which were traditionally
considered as cold, like reasoning, planning, and decision making, has been put
forward in the last years by several researchers (LeDoux, 1996; Rolls, 1999, 2000). In
1994, Damasio proposes a mechanism by which emotional processes can guide (or
bias) behaviour, particularly decision-making. This hypothesis is the somatic-marker
hypothesis (SMH).
Recently the subject of emotion has attracted also the attention of the computer
science and artificial intelligence communities, and has emerged into a new area of
research, sometimes referred as affective computing (Picard 1997). Affective
computing was defined by Picard as computing that relates to, arises from, or
deliberately influences emotion. The definition encompasses the different categories of
affect, and includes both, systems that actively influence the user affective state, and
systems that capture the user affect.
Although there is already a number of cognitive appraisal models, such as Ortony,

Clore & Collins (1988) or Roseman, Antoniou & Jose (1996), which provide
conclusions about which emotion a subject should feel under given circumstances,
these symbolic representations do not explain what happens to the thinking process
itself as a result of these emotions. In order to understand and study this issue we need
to introduce emotions into cognitive architectures such as SOAR or ACT-R or AI. In this
chapter we present examples of implementation of emotional states into cognitive
architectures and Intelligent Agents.
It is becoming evident that cognitive models used to test different psychological

theories should take emotion into account (Belavkin, Ritter & Elliman, 1999). Many of
these cognitive models simulate subjects solving various puzzles and problems, or
consider children, as in Jones, Ritter & Wood (2000), whose emotions are easily
observable. Emotion accompanies any problem solving process indeed, but these
computer simulations say nothing about it. It seems that in cognitive science this
subject has not yet been studied deeply enough (Belavkin, 1999).
Reilly and Bates within the OZ project in 1992, created an environment for simulating
believable emotional and social agents and each of them had a set of goals that was
trying to achieve. These agents were able to express emotions after evaluating the
impact of an event on the agents goals. Ten year later, in 2002, Conati used a
probabilistic model, based on Dynamic Decision Networks to recognize the emotional
state of the user with educational games from possible causes of emotional arousal.
Velasquez, in 1998, studied the modelling of emotion based decision making to be

implemented in an emotional pet robot. The framework upon which Velasquez was
based is presented in the following figure.

Figure 52: Velasquez baseline framework. (Velasquez, 1998)
In contrast to what other models had proposed to that date emphasizing on cognitively
generated emotions, Velasquez consider both cognitive and noncognitive releasers of
emotion. The activation of emotional systems was triggered according to an equation
that considers both the excitatory (positive) and inhibitory (negative) input from other
emotional systems, well as its temporal decay.
Where Ai(t) is the activation of emotional system i at time t; Ai (t-1) is its activation at the
previous time step; () is the function that controls the temporal decay of the activation
of emotional system i; Rkl is the value of releaser k, and Wki, is its associated weight,
where k ranges over the set of releasers for emotional system i ; li is the strength of
the excitatory (positive) or inhibitory (negative) input from emotional system i, where
Al(t) is its activation value at time t; and f is a limiting function.
Belavkin, 2001, introduced emotions in a model implemented in ACT-R and used the
Nottingham Eye and Hand perception-action module (Baxter & Ritter 1996), as well
as the dancer model to interact with the task simulation. Belavkin showed how some
effects of emotions on problem solving can be modelled in cognitive architectures with
nondeterministic and statistical learning capabilities, like ACT-R. In such architectures,
changes of motivation (maximum cost, or goal value G) and randomness (noise
temperature ) can produce noticeable effects on the behaviour of models. This can be
used both for a better data matching and modelling the behaviour of subjects in highly
active or passive emotional states. The similarity between optimisation methods and
emotional reactions during problem solving, that Belavkin showed, suggests that
emotion makes a positive contribution to intelligence, as it implements powerful
heuristic methods already known and used elsewhere in AI.

Belavkin suggests that an emotional problem solver in general, follows a very powerful
optimisation method, in which:
o Positive emotions, experienced on successes during problem solving, are

accompanied by increase of the motivation (goal value G) and confidence (G/
ratio).
o Negative emotions correspond to a decrease in G and G/ (heating the system

up). Negative emotions occurring during problem solving can play a positive
role in overcoming possible problems. In a hill-climbing illustration these
problems are known as local maximum, plateau and ridge. The strategies used
to overcome these problems are the change of direction of the search (G
decrease) and random jump (noise temperature increase).
Silverman, in 2001, focused on improving the behavioural realism of computer

generated agents and attempted to reflect the state of the art in human behaviour
modelling with particular attention to value ontologies, emotion, and stress in game-
theoretic settings. The goal was to help those interested in constructing more realistic
software agents for use in simulations, in virtual reality environments, and in training
and performance aiding settings such as on the web or in embedded applications.
Silverman made a very interesting correlation of the emotions with the Stress
Components and fatigue and concluded to the following:
o Near Zero Event Stress: use initial task plan and dont call emotion model (no
situational construal). Ignore probabilities and apply criterion of optimism.
o High Event Stress: same as near panic.
o Fatigued (EF): reduce all positive goal- and preference-based positive emotions
(become timid). Ignore probabilities and apply criterion of pessimism.
o Time Pressure (TP): primary impact is to reduce probabilities of success for an

action.
Fum & Stocco (2004) present an interesting model of experimental data involving the
Gambling Task (GT) that elucidates some of the specific limitations of cognitive models
in ACT-R and some general issues in current artificial intelligence applications. The GT
is a well known research paradigm that has been used to investigate the role of
emotion in decision making (Bechara, Damasio, Damasio, & Anderson 1994; Bechara,
Tranel, & Damasio 2000). The task involves real world factors such as uncertainty, real
time decision making, and choices having personal consequences that could be
rewarding or punishing. At the beginning of each session of the GT the session
administrator gives the participants a certain amount of play money. The session then
proceeds through a series of trials. In each trial, the test administrator asks the
participants to choose cards from one of four decks, each with a complex schedule of
gains and penalties in the play money available to the participant. At the beginning of
the session, the administrators inform the participants that their goal is to maximize the
amount of play money they have left at the end of the session.

Gebhard, in 2005, developed ALMA, A Layered Model of Affect system used to model
emotions, moods and personality according to events and time passing. The goal of
this model is to provide realistic artificial emotions that can be then mapped to virtual
agents for use in virtual classrooms or games for example. In order to achieve this, a
three-dimensional space is used to represent the moods and emotions are used as
mood altering vectors. In ALMA, Gebhart used the OCC (Ortony, Clore and Collins)
model in order to realize a realistic modelling of human emotions. In ALMA the
emotions where mapped into a three-dimensional space called PAD-space. Emotions
are translated in the system as position modifiers into the PAD-space and are
generated according to the events, due to other agents or user, as well as, random
events, affecting the agent and their effects are decaying over time. These can be
altered to achieve a certain personal behaviour from the agent in the configuration.
These were mapped with the OCC Model using the table below.
Figure 53: Mapping of the OCC Model emotions and mood octants.
In another research Cochran, Lee and Chown, 2006, tried to develop a framework of
emotions, suitable for including emotion in an integrated cognitive architecture and
examining the impact of arousal on memory. In this research Cochran et al. connected
emotions to appraisals, i.e. fear was seen as a reaction to a prediction of danger. This
model, put forth by Chown and others (Chown, 2006; Kaplan, 1991), posits that

emotions are a fast, automatic assessment system consisting of three core parts or
mechanisms:
m1: an arousal system that provides a measure of importance,
m2: a pleasure/pain system that assesses valence, and
m3: a clarity/confusion system that provides a measure of competence.
The representation of the emotion according to this model is illustrated in the figure
below which depicts how these mechanisms can be viewed in a 3-state space and how
the model may drive these states over time in response to various stimuli.
Figure 54: Representation of emotions. (Cochran, Lee and Chown, 2006)
Continuing, Cochran, Lee and Chown tried to model the results of the a classic paired
associate experiment performed by Kleinsmith and Kaplan (1963), which revealed the
impact of arousal on memory by performing memory tests on subjects while monitoring
arousal levels, by utilizing a model that can predict both the low arousal and high
arousal results, in ACT-R.
Within ACT-R, memory is represented by chunks. Every chunk in ACT-R is associated

with an activation level. When a retrieval request is made of the memory system, the
chunk with the greatest activation among those that match a retrieval specification is
retrieved. However, there is a retrieval threshold that specifies the minimum activation
level of a chunk for retrieval. If the chunk with the highest activation among those that
match the request has an activation level less than the retrieval threshold, then a failure
to retrieve will occur.
The equation for the activation Ai of a chunk i is defined as:

Where:
Bi: The base-level activation.
: Noise.
The noise is composed of two components: a permanent noise associated with each
chunk and an instantaneous noise computed at the time of retrieval.
The equation for base-level activation for chunk i is:
n: Number of presentations for chunk i.
tj: Time since presentation j.
d: Decay parameter
It is obvious though, that arousal is not a parameter to the activation equation;

therefore, the existing ACT-R formula was adequate for modeling both the low arousal
and high arousal performance curves. Experiments showed that the current ACT-R
memory model is not adequate for predicting the recall results of the high arousal. In
order to do so, Cochran, Lee and Chown, created a memory model that increases
activation with time. This was accomplished by replacing the decay parameter with a
growth parameter (negative decay rate) and implementing a ceiling for activation after
45 minutes. This ceiling represents that the memory no longer continues to increase or
decrease (at least at the same rate) once it is deeply encoded and the short term
blocking effect has dissipated. The growth parameter generates an increase in memory
activation as time proceeds between encoding and attempted recall.
After implementing this experiment Cochran, Lee and Chown proposed the following:
1. The base level activation decay parameter (d) should be some function of
arousal at time of encoding. The decay rate should increase for lower arousal
levels and turn into a growth parameter (g = -d) for higher arousal levels.
2. the defined d(a) should be a linear function of arousal (a), a high arousal
threshold (ah), and a scale parameter (s):

ds: scaled decay (growth) parameter: s*d
a: arousal at the time of encoding;
an: nominal arousal
ah: high arousal threshold
Arousal (a) can vary between 0 (e.g., sleep) and 1 (full arousal). The exponent to tj
becomes a growth parameter when arousal is high (a > ah), and the decay accelerates
when arousal is low (a < ah). The equation defaults to the ACT-R activation equation
when arousal is set to its nominal level and decay is scaled to the inverse of the (1-
a/ah) term. The term outside the log function (a an) provides initial blocking for high
arousal levels.
In 2008, Pour Mohammad Bagher studied using fear in the Intelligent Agent system
simulation with fuzzy logic and more specifically, by implementing fuzzy inference
system using Java environment. He described the relation between three reasons of
fear emotion and the fear degree that a human may feels, and simulate an emotional
fuzzy intelligent agent via some fuzzy rules. 27 fuzzy rules were designed in which the
degree of the three reasons of fear was considered as input (or antecedents) and the
degree of captured fear emotion was considered as output (or conclusion). Pour
Mohammad Bagher developed a model that calculated fear intensity and it is illustrate
in the figure below.

Figure 55: Class diagram of the designed software and calculated fear emotion
intensity. (Pour Mohammad Bagher 2008)

4.4 Motivation as cognitive and behavioral moderator

4.4.1 Definitions of motivation
Motivation is linked to individual and sociocultural aspects and is present in all areas of
human existence. We speak about motivation to refer to the impetus that starts,
guides and maintains behaviour to reach a desired goal or objective (Ardouin,
Bustos, Gay y Jarpa, 2000; Cortese, 2005; Zornoza, 2005).
In short, motivation is what makes an individual act and behave in a certain way. It is a
combination of intellectual, physiological and psychological processes decided for a
particular situation, how quickly the individual reacts and in which direction energy is
channeled (Solana, 1993)(quoted by Lpez).
In Cuba, Diego Gonzles Serra (1982) defines human motivation as a complex

integration of psychic processes which, while in constant reciprocal transformation and
determination with the activity and external stimuli, aims at satisfying mans needs, and
in consequence regulates the direction and intensity of the behaviour.
Gonzlez and Mitjans (1989) say that human motivation cannot be reduced to a frame
of mind that instantaneously stimulates a behaviour based on the living experience of
need. The dynamic potential of motivation is linked with a relevant content for the
subject organized into concept, reflections, ratings, whilst also being the carrier of an
emotional burden
The most important element in motivation lies in how we feel emotionally in a given
situation. A fundamental element is the way we feel our emotions.
From the previous definitions, we can summarize that motivation is a complex variable
in which conscious, unconscious, cognitive and affective aspects intervene, as well as
being a differentiating factor for stamping strength and directionality on human
behaviour, the reason for which it has been the subject of study by different scientific
disciplines.
4.4.2 Types of motivation

Motivation is made up of all factors having the capacity to prompt, sustain and guide
behaviour towards an objective. We could talk about two types of motivation:
Primary motives: help us to satisfy basic needs, such as food, drink, heat and shelter.
These needs must be fulfilled in order to guarantee survival and may not be
easily altered by will.
Some of them are cyclical (eating and sleeping) and the intensity with how they
are felt increases and decreases on a more or less regular basis.
Secondary motivation: this type is acquired or learnt and the needs that are satisfied
may or may not be indirectly related to the primary reasons. Some of these may be:

Friendship.
Freedom.
Honour.
Power.
Wealth.
Fame.
Love.
Motivation can be also classified as positive or negative:
Negative motivation is what forces a person to do something through the use of

punishment, threats, etc.
Positive motivation is the constant desire to improve, guided by a positive spirit,

which some writers split up into intrinsic and extrinsic (Leonor, 1997).
o Intrinsic motivation is characterised as such because the person centres

their interest on work, thus showing an active role in achieving their ends,
aspirations and goals. An activity is carried out for the enjoyment
experienced whilst doing it and triggers off exploration, knowledge and
creativity. This type of motivation can be subdivided into intrinsic motivation
leading to the doing of things, and towards stimulating experiences; the
latter focuses on carrying out an activity in order to live sensations as
sensory pleasure, enjoyment, excitement and aesthetic experiences, whilst
the former (intrinsic motivation toward doing) is defined as joining in an
activity as a consequence of the satisfaction felt on trying to do or create
something.
o In the case of extrinsic motivation, motivators of an external nature are

taken into account, like the advantages offered by carrying out a task. In
other words, it is a means to reach an end and not the end in itself. There
are three types of extrinsic motivation:
1. External regulation. Behaviour is regulated through rewards and

punishments, such as Ill finish the report today because if not Ill be
disciplined.
2. Introjected regulation. The reasons for doing a determinate action are

internalized. It is not, however, a true self-determined internalization and
limits itself to the internalization of past, external contingencies, for
example, Ive got to finish off a certain project because last month I
didnt receive any stimulation as I stopped doing it

3. Identification. This is the extent to which the behaviour is considered

important for the person, who perceives that they themselves chose this
type of behaviour. Thus, the internalization of extrinsic reasons is
regulated by means of identification, for instance in the case of I chose
this profession because it is important for me
An example of intrinsic and extrinsic motivation for working purposes is shown in the
following table.
Type of motivation Motivations
Extrinsic motivation Motivation based on organisational policies.
Motivation based on establishing effective human

relationships.
Economic motivation.
Motivation based on working conditions.
Motivation based on prestige of company
Motivation based on proximity to home.
Motivation based on management style.
Intrinsic motivation Motivation based on doing and achieving.
Motivation based on content of activity performed.
Motivation based on maintaining stability.
Motivation based on amount of autonomy granted.
Motivation based on personal commitment.
Motivation based on personal growth.
Table 20: Intrinsic and extrinsic motivation for working purposes
4.4.3 Motivation theories and theoretical models

In an aim to comprehend human activity, Psychology has attributed motivation with the
role of explaining behavioural causes. Among the basic psychological processes,
perhaps motivational are the ones to be most closely related to action, regardless of
whether the theoretical framework taken be behavioural, cognitive or dynamic (Barber
y Mateos, 2000). In fact, the relation between causal explanation (motivation) and
resulting effect (behaviour) has often generated a certain air of confusion that is evident
in the circular nature of quite a few psychological definitions, where the concept of
motivation is inferred from the behaviour that must be explained by supporting itself on

the concept of motivation. Criticism of the circular explanation suggests that a scientific
theory must define the states (needs, desires, impulses, incentives) that are put
forward as behavioural reasons independently from the activities that are trying to be
explained (Wise, 1987).
Throughout the XX century, Psychology has developed multiple theories on human

motivation to the extent that in order to get a broad view of the dominant trends it is
necessary to introduce some organizing criteria. Many different motivational types
exist, however there is no doubt that the dualist tradition, which has prevailed
throughout the history of western thinking, has left an indelible mark on psychological
studies of motivation.
Following the dualist criteria, the classification that is proposed here in order to begin
the debate on the role of motivation in explaining human activity has its starting
point in the differentiation between reactive models and activation theories. A
classification means organizing and clarifying, yet it also implies standardizing, dividing
and simplifying, which leads to a loss in differentiating nuances and richness in content.
However, accepting risks is a condition that is necessary to advance knowledge.
4.4.3.1 Reactive Theories

Under this heading we can include clearly differentiated theoretical and epistemological
approaches. Nevertheless, between them they all share the concept of the subject as a
reactive being, whose actions respond to changes produced in the state of a specific
stimulating situation. The motivation is therefore set off as a reaction to a specific
emotion (fear), a biological need (hunger) or a psychological one (curiosity) and
also when in the presence of external stimuli (appetizing/aversive). In all of these
cases, the aim of the motivational behaviour always consists of satisfying a demand,
thereby reducing pressure. (See following figure)
Figure 56: Motivational reaction to external and internal stimuli

With regard to the initiating sources of the motivated behaviour, there are two remarks
concerning basic emotions and the inclusion of curiosity. The approach of emotions as
motivational agents is shared by almost all writers, although there are different ways
to show this link. Therefore, whilst Izard maintains that the central function of a basic
emotion is similar to that of a reason and consists of activating and guiding behaviour,
the Buck model illustrates motivational and emotional processes as two sides of the
same coin, attributing emotions the specific task of making successful adaptations
simple or difficult. On the other hand, the interpretation of curiosity as a
psychological need with motivational properties similar to those of hunger or thirst
should be explained in more depth, which will be done later.
Underlying the basis for this motivational representation is the notion of

homeostasis, a concept that derives from Physiology and was introduced by Cannon
to refer to the dynamic equilibrium that the internal environment maintains within a
limited variability
Psychophysiological research initially took interest in primary motivational

orientations, such as feeding, fleeing or avoiding pain, and in order to explain them
it developed both local theories (hunger or thirst) as well as others involving the
intervention of Central Motive States (Morgan, 1943). In each one of them a series of
behaviours aimed at maintaining homeostatic equilibrium are activated either by the
stimulation of specific peripheral points, or by activity from the excitatory centres of the
hypothalamus.
Hull and his schools (Hull, 1943, 1952; Spence, 1956) motivational formula represents
the most systematic and comprehensive attempt to transfer the homeostatic model to
the psychological explanation of human motivation. Hull develops the functionality of
motivational processes using two basic concepts, drive and incentives, and his
explanation is encapsulated within the general behavioural theory, which is based on
the principals of associative learning, which are characteristic of mediatory
behaviourism. In the sixties, the incorporation of Duffys physiological activation
concept (1962) provided Hulls drive with psychophysiological measurement, which
meant that its intensity could be measured irrespective of the resulting behaviour
(Suay, Salvador y Gonzlez, 1996).
This theory has also been subject to numerous criticisms. Among them, it is worth
pointing out his general claim for explaining complex human behaviours, related to
situations of frustration or to conflict-based reactions, from experimental designs
fundamentally based on animal research in which study is carried out solely on
motivational reactions when faced by situations of forced hardship. In spite of
everything, there is no unitary motivational model in present psychology to represent
what Hulls proposal signified in his day.
Reactive cognitive explanations: models based on Expectancy/Valence (Value)
The cognitive point of view, whose influence in the field of psychology started to rise
from the end of the fifties, focuses attention on the study of complex motivations

occurring in specifically human activities with respect to conscious behaviour of a

voluntary nature, such as is the case of achievement motivation or goal setting. A
large number however of the cognitive models developed to explain motivation take on
and reproduce the general scheme that is characteristic of the homeostatic concept.
The clearest example of transposing the homeostatic model to the field of human
cognition can be seen in a series of theories roughly labelled as cognitive consistency.
Although these theories have specific formulations and different scopes of application,
they all converge on the idea of presenting motivational interpretation within alternative
series of consistency/inconsistency as central features of psychic activity.
Festingers cognitive dissonance theory (1957), the psychological reactance theory

(Brehm, 1966), Bems self-perception model (1972) or the equity theory (Adams, 1965;
Homans, 1961), share a prototypal structure of the homeostatic model.
The general scheme of the drive reduction theory, which is applied in the Hull design
to the need to satisfy the desire for food in a situation of hunger, is used to explain the
functioning of the psyche to cope with dissonant thoughts, perception of inequality or
any other inconsistency between beliefs and behaviours. The motivational content we
want to learn about changes (basic motives versus secondary motives), as well as the
general design of the research (laboratory animal versus observation and recording of
human reactions). But, the explanatory homeostatic structure remains intact in the
drive reduction model and in the cognitive consistency theories. They both share a
mechanistic interpretation of human motivation based on the dynamic alternation
between initial equilibrium (consonance) and later disequilibrium (dissonant situation).
The subject is motivated therefore when faced by a specific stimulating situation
and reacts using predetermined behaviour in an aim to reestablish the previous
state.
Among the cognitive explanations for human motivation, one that has had the greatest
impact is based on highlighting the role that the expectancies and valences play with
respect to action-anticipating concepts in the level of effort (motivational tendency)
assigned to certain behaviours. A series of models, developed by Atkinson (1957),
Feather (1959) or Vroom (1964) and applied to diverse environments, can be grouped
together under the common denomination of expectancy/valence theories (E/V).
These theories share a rational vision of human beings with those of
consonance/dissonance and consider that behaviour is intentional to the extent that
it complies with a functional purpose. Likewise, they share a reactive approach to
motivation in so far as motivational tendencies are interpreted as specific reactions
triggered off by changes in the stimulating situation. In this case, the stimuli
(expectancies and valences) are internal, mentalist and anticipatory. Although
intentionality is taken as an inherent character of behaviour, it does not represent an
explicatory construct of human motivation. They keep the mechanist criteria by
ascribing all the responsibility for the dominant motivational tendency to expectancy
and valence.
One of the writers of these theories who most stands out is Vroom (Vroom, 1964),
although this has been completed by Porter-Lawler (Porter y Lawler, 1968). This theory
claims that individuals as thinking people have beliefs and harbour hopes and

expectations with respect to future happenings in their lives. The behaviour is a result
of choices between alternatives and these choices are based on beliefs and attitudes.
The aim of these choices is to maximize the rewards and minimize the pain
(Pinder, 1985[19]). Highly motivated persons are those who perceive certain goals
and incentives as valuable for them and, at the same time, they subjectively view
that the probabilities of achieving them is high. So, in order to analyse the
motivation it is necessary to know what they are looking for in the organisation and how
they believe they can achieve it (Laredo).The most important points of the theory are
(Galbraith, 1977):
All human effort is carried out with the expectation of certain success.
Subject trusts that if the expected performance is achieved, certain

consequences will follow for them. The expectation that the achievement of
the aims will be followed by undesirable consequences is called instrumentality
Every consequence or result has a determinate value for the subject, called
valence.
A persons motivation to carry out an action is greater, the greater the product
of the expectancies, depending on the instruments and the valence. Will I
perform enough? What will I achieve if I work hard? Is it worth it?
The relation between effort and performance depends on two factors: the
subjects abilities and their perception of the position.
Each person has a certain idea about the level of performance that they are
capable of achieving in the task.
People believe that whoever carries out the jobs best will receive the best
rewards.
A persons strength of motivation in a specific situation is equal to the product

between the value that the person allots to the reward and the expectancy of
possibly achieving it:
Strength of motivation = Value of reward * Probability of achieving
Some of the consequences could be:
The definition of standards, goals and objectives must respond to real

estimations. The aim is to define demands that are reachable but through
effort.
The rewards per achievement must be well aligned with the true
expectations. This requires a knowledge and awareness of people, their
culture, their interests, etc.

It is essential that people are convinced that the rewards they receive are
fair and that they are certain that a person who carries out a task badly will
not receive the same rewards as they do.
Reworking and widening of Expectancy/Valence (Value) parameters
When Professor Pedro Mateos (1996) reviews the evolution of concepts of motivation,
intention and action with regard to Psychology, he analyses in depth some
psychological proposals which, while maintaining the general philosophy for models
based on expectancy and valence, try either to widen their scope, or to complement
these two cognitive parameters with other motivational components. Examples of re-
working of E/V parameters can be found in proposals by Bandura (1977) or
Heckhausen (1977), where two types of expectancies are differentiated.
Banduras classification distinguishes between efficacy expectations and outcomes.

The first, referring to the perception of self-capacity to carry out a certain behaviour,
can be conceptually placed between the subject and the action. The expectancy of an
outcome, however, refers to the conviction that a determinate action will produce a
determinate outcome. They intervene as mediatory activities between the action and
the expected outcome, as shown in the following diagram:
Figure 57: Classification of expectancies proposed by Bandura (1977)
Heckhausen, on the other hand, classifies three types of expectancies:
i) situation-result (S-R)
ii) action-result (A-R)
iii)result-consequence (R-C).
The A-R expectancy practically coincides with the notion of Banduras outcome
expectancy. The S-R expectancy is defined as the belief that a situation in itself,
separate from the subjects behaviour, will lead to a determinate outcome. For
example, the conviction that determinate situations of poverty go together with school
failure. Lastly, the R-C expectancy consists of the belief that the expected result will act
as a mediating tool in order to reach some determinate consequences. If I finish my
degree (result) Ill be able to access a professional position (consequence)
Both Banduras notion of expectancy efficacy and Heckhausens R-C expectancies

were to produce important repercussions in the psychological interpretation of
motivation. Bandura develops the concept of self-efficacy (Bandura, 1982, 1995),
insisting on the motivating role that is played by self-perception as an agent capable of
carrying out determinate actions. We will come back to the notion of self-efficacy and
perception of self-capacity later when we talk about intrinsic motivation.

At the same time, the distinction between outcome and consequence posed by
Heckhausen makes it possible to interpret the results as instrumental steps aimed at
goals of a higher order that give important value to our actions. In this way, although it
may not be possible to directly influence the consequences and only the results, the
belief in the relation between results and subsequent consequences gains motivational
value. The relevance of future actions over present behaviour was to be emphasized
by Raynor (1981) through his conception of future orientation and by Gjesme through
the concept of temporary distance of goal
In addition to extending the traditional concept of expectancy, or to incorporating

differentiated types of anticipated assessments (social comparison, self-assessment,
extrinsic incentives, costs/gains, or cultural value), there have been other proposals
incorporating additional parameters to those of expectancy and valence to explain the
motivational tendencies of behaviour.
One of the most representative examples of the extension of the E/V model is found in
Atkinson and Burts dynamic action theory, (1970, 1978). Here, the dominant action
tendency is represented as the final outcome of counteracting instigating and
consuming strengths to cope with inhibitory forces, on the one hand, and resistance, on
the other. The idea of including in the motivational explanation both the tendencies
towards action and those of evasion is already present in Atkinsons (1957) previous
explanation about achievement behaviour. The novelty of this proposal is that the
incorporation of mediating parameters (instigating and consuming forces/inhibitory and
resistance forces) among the expectancies/valences and the resulting motivational
tendencies, raises the theoretical possibility of producing a change in behaviour and in
the underlying motivational structure, even when the subjects expectancies and
assessments of a vital situation are constant.
Figure 58: Scheme of explanatory parameters according to the dynamic action

theory
To sum up, from the sixties on there has been a proliferation of a series of models
aiming to explain the close relationships between motivational factors and
behaviour, or between knowledge, motivation and activity. In spite of the undeniable
progress represented by some of these cognitive proposals with regard to a more
active and self-regulating consideration of motivation (Bandura, 1982; Raynor, 1981),
the series of theories included in this block, classified as reactive, share a common
approach that can be outlined in the following graphical representation.

Figure 59: General scheme of motivaton following the reactive theories approach
4.4.3.2 Activation Theories

This approach has its origins in considering human beings as the causing agents of
their own actions. Motivation is interpreted as an activity that often manifests itself
spontaneously, without the need of being reduced to a mere reaction to specific
stimulating situation. Moreover, motivated behaviour is considered as propositive
since, to a large extent, its development is driven by plans, goals and objectives.
From a psychophysiological point of view, Hebbs (1955) concept of motivation,

defined as the tendency of any organism to produce organized activity, and
knowledge of neural substrate of arousal with regard to reticular formation and non-
specific thalamus nuclei, constitute two important endorsements to hold up the
conception of a human being as a causal agent. Later criticism of representing
activation as a unitary construct (Vila y Fernndez, 1990) would make it possible to
incorporate a systematic approach. This would take into account both the central and
peripheral neural structures as a modulator for neuroendocrine components, especially
the series of interactions that are set up between these components. Identification of
the brains motivation system based on the findings of Olds and Milner, have made it
possible to infer the existence of a neural circuit, whose function is to regulate the level
of activation and to provide the necessary impulse to act in a certain direction. (Suay,
Salvador y Gonzlez, 1996).
Unlike reactive approaches, human motivation in activation theories can basically

be characterised by spontaneous and propositive properties. Curiosity, the thirst
for exploration or a sense of self-determination of behaviour are, in themselves,
capable of psychological activation and the aim of this activity does not consist in re-
establishing the previous situation or in restoring the broken equilibrium, but in guiding
behaviour towards new, unknown and challenging paths. The motivational importance
of the intentionality and the weight exercised by will over propositive behaviour have
grown in strength in recent years, revolving round a general theory of the action that is
clearly separate from the traditional scheme concerning behaviour.

The development of intrinsic motivation and psychological research developed to

explain its origin show interest, above all, in spontaneous analysis as a characteristic
motivation trait, which is present in multiple human activities. The propositive nature,
which is in close harmony with the notions of intentionality and will, was to take up the
centre of attention in the motivational explanation of action models (Heckhausen, 1987;
Kuhl, 1985).
Intrinsic motivation
The definition of intrinsic motivation suggests that a large amount of human activity is
carried out for the pleasure involved, or for the interest it entails. The key question for
psychological research has revolved around awareness of what the underlying factors
are that make it possible to explain intrinsic motivation. Ultimately, what makes an
activity intrinsically motivating? The answers obtained have been diverse and, to a
certain extent, complementary.
Some researchers have shown interest in analyzing the specific properties of some
objects, which turn them into intrinsically motivating. So, for example, Berlynes studies
(1960) applied to the field of education converge in characteristics on novelty,
complexity and unpredictability, which certain activities possess and determine the
motivational interest. An intermediate level of each one of these properties awakens
the subjects interest and favors curiosity, the thirst to explore them. Whether an activity
should turn out to be moderately new, complex or unpredictable, depends somewhat
on a comparison of information obtained from different sources. In this sense, such
properties are defined as collations of stimuli patterns
To a certain extent, the reactive homeostatic scheme adheres to this explanation, in

the sense that the collative properties (novelty, complexity, unpredictability) of certain
stimuli arouse curiosity and lead to the exploration and handling of objects in order to
learn about them better. Similar to in Hulls classic scheme, the shortage of a certain
food substance generates hunger and leads to the search for food to satisfy the need.
In this sense, curiosity can be represented as a psychological need, which acts in a
similar way to the basic needs of hunger and thirst, as was previously mentioned
Csikszentmihalyi (1975) incorporates the notion of flow, whose most direct antecedent
could be found in the idea of optimum challenge, as a meeting point between the level
of difficulty of a task characteristic of an activity and the skills the person has for
resolving the problematic situation. Here, motivational activation does not only depend
on the novelty or intrinsic interest of the job in question, but on the correspondence
between the motivational activation and the personal resources a person has to cope
with the situation. The origin of intrinsic motivation does not only depend on the
collative properties that certain objects possess, but on a suitable balance between an
individuals competence and the challenge involved in the task. When the challenges
are greater than the individual competences a state of anxiety is generated due to an
excess of difficulty. If, on the contrary, the skills are far greater than the challenges, the
individual will get bored and therefore not be very motivated.
Although there is no doubt that the notions of flow, or challenge, incorporate, to a

certain point, the feeling of personal control over ones own skills and the interaction of

the subject with activity, the underlying motivational conception is still reactive, insofar
as the pleasure that leads to carrying out the activity comes from, at least in part, the
collative properties of the stimuli.
However, psychological research interested in intrinsic motivation has looked into its
origins inside the subject by means of self-perception as a person who is competent,
efficient and determined to act. The notions of self-competence (White, 1959), personal
causation (deCharms, 1968), self-determination (Deci y Ryan, 1985), self-efficacy
(Bandura, 1982) and personalised action (Nuttin, 1985) highlight the motivational
prominence of subjectivity and the role of the individual as a causal agent of their own
behavioural activity. Each of these concepts tries to accurately define the nature of
intrinsic motivation. However, in any of these explanations motivation, rather than
responding to specific determinate stimulating characteristics, provides opportunities
for people to put their competences and determinations to the test in an active and
intentional manner. The characteristics of dynamism and internal activation move away
from the notion of the intrinsic motivation of enjoyable activities derived, for example,
from a passive sensorial experience. As Reeve (1994) says, we might like going to the
cinema or listening to music, but these activities cannot be defined as being behaviours
activated by intrinsic motivation. It would be a different thing if such sensorial activities
aroused our interest to learn how to read music, play a musical instrument or to study
cinematography.
Maslows hierarchy of needs theory (Maslow, 1954) is perhaps the most classical
and widely known theory. He identified five different levels of motivational needs having
a pyramidal structure in which the basic needs are found at the bottom, and the higher
or rational needs are found above (Physiological, safety, social, esteem, self-
actualization). For Maslow, these categories of relationships are placed hierarchically in
such a way that a need is only activated after a lower level has been satisfied. Only
when a person manages to satisfy the lower needs do the higher needs gradually
come into play, and thus the motivation in order to satisfy them.
Figure 60: Marlow's hierarchy of needs

From the point of view under analysis here, with regard to the differentiation between
reactive theories and activation theories, it is this last consideration on intrinsic
motivation the one that arises in the actively and spontaneously motivated individual, in
a quest for self- assessment both as a competent person and with self-determination,
as far as the environment is concerned.
Action models
Together with the development of intrinsic motivation, in the eighties a series of

psychological models was formed incorporating intentionality and willpower as
relevant aspects in propositive human behaviour, delegating it a specific role in
motivational explanations. The two most representative are Kuhls (1985, 1986) action
control theory and Heckhausens (1987) Rubicon model of action phases. Both
explanations are directly linked with the German tradition of studying the will, from the
Wurzburgo School, and with the notion of intentionality, defined by Ach as the driving
force for a determinate tendency that triggers off its realization.
The two largest contributions of these models to support an active approach for
motivation and not simply a reactive one are:
Establishment of an important conceptual demarcation between motivational

tendency and achievement of objectives, from which theoretical and applied
consequences of great impact are derived, and
The idea that goal-setting behaviour is organized in a hierarchy, although

development of its structure and empirical founding of the hierarchy are yet to
be deciphered.
With regard to the first issue, both models include a differential analysis between the
processes intervening in decision making and those that act on carrying out the
proposed aim. In order to achieve a voluntary goal it is necessary, without a doubt, to
be very decided on wanting to achieve it. However, the intention, as a central
motivational concept representing the maximum level of commitment with the action,
does not guarantee success of the desired outcome. Experience in everyday life shows
the distance that exists between the intentions and the achievements. Almost all
students start an academic course with the firm resolution of passing a number of
subjects; however, the proportion of those who achieve it is usually much lower.
In particular, Kuhl highlights as insufficient the implicit assumption included in EV

theories on identifying motivated behaviour with goals. On the contrary, the starting
point for his model is to differentiate between intentionality and action. The proposal
hinges on two basic ideas concerning:
impulses, desires, assessments and other motivational tendencies are

determinates of the degree of commitment with the action. The highest
level can be found in the intention or firm resolution of achieving a goal, and

between intention and behaviour there is a series of interceding complex

volition process that will have to win out against different, difficult tendencies,
both external and internal, in order to achieve the goal set.
These last processes and strategies, more than the driving, affective and cognitive
parameters that determine the level of commitment with the action, are the ones that
Kuhls model is interested in studying. The figure shown below aims at representing the
basic approach of the writer.
Figure 61: Scheme of the Action Control Theory (Kuhl, 1985)
Heckhausens Rubicon model shows the same idea as Kuhl, with regard to the
differentiation between motivational and volition processes, in more detail. In his
representation, he includes four phases and two key moments of inflexion (the Rubicon
step) in the decision making process, as can be seen in the following diagram:
Figure 62: Scheme of Rubicon theory of action phases (Heckhausen, 1987)
The fourth role of the model also incorporates a new motivational component in the
sequence, which corresponds with the assessment the subject makes on the action
achieved; an idea which had already been sketched out previously using the notion of
expectancy of result consequence. Incorporation of the assessment in the
explanation on the motivated action for voluntary behaviour makes it possible, on one
hand, to establish a feedback circuit in the representation, as only the positive
assessment of the obtained results will enable the intention to be permanently
deactivated. On the other hand, assessment of the preceding action makes it implicitly
possible to assess the consequences, thus linking with criterion that the future can
have an influence over the present through anticipated goals and their assessment.
4.4.3.3 Integrating motivation, emotions and cognition: future trends

Emotional responses, through their physiological components and their tendency
towards action, prepare and motivate people to be able to relate to the events that elicit
these emotions. The emotions also often try to obtain their own rewards by attempting
to motivate behaviours that increase the likelihood of others occurring that cause
certain emotional states. However, not only do the emotions influence the motivation,
but there is a reciprocal influence.
The idea that a great deal of human motivation occurs spontaneously, which is in line
with the intrinsic interest to exercise ones own skills, or to put to the test the capacity to
intervene in the environment, constitutes a considerable advance and has substantially

enriched the role played by subjectivity in the development of psychological activity.

Another decisive point has been the inclusion of intentionality and free will when
assessing ones behaviour or that of others. Experience shows that neither behavioural
response nor an evaluation of the event is always the same when the actors bad or
good intention is assumed. Knowledge, motivation and assessment therefore exhibit
close interactions between each other.
Apart from a continuous interaction among these three psychological processes,

Kuhls model (1986) sets up specific links for each one with the social environment in
such a way that the priority relation of cognitive processes is that of representation,
whilst what specifically characterizes emotions is the extent to which humans tend to
assess any event, to a greater or lesser degree. Lastly, the specific characteristic of the
motivational processes is the extent of commitment with the action. The following figure
shows a scheme of this idea.
Figure 63: Relation of psychological processes with the world (Kuhl, 1986)
Nevertheless, cognitive psychology has dedicated much time in preferentially analysing

the crucial role that knowledge (dissonance, expectancies, causal attributions) and
willpower (regulatory mechanisms and ways to control an action) play in motivation and
human activity. This has reached the point of maintaining not only with theoretical
arguments, but also with empirical evidence - that such a mental factor as anticipating
future goals may be decisive in assessing the degree of effort and behavioural
reactions. The influence that knowledge has in the hottest talking point in Psychology
emotions has been rekindled again recently in the study on emotional intelligence,
of emotional quotient as a different and complementary factor to that of IQ (intelligence
quotient) and the development of instruments to measure these concepts.
The inverse path has however not been explored to a similar extent. It is often taken for
granted that motivational tendencies and emotions have an influence over cognitive
representations. It is usually taken that personal wishes, affectionate aspects, and even
the sense of self-efficacy affect the representation of knowledge and human
interaction. But the diverse ways in which such influence occurs has not been analysed
in depth or formalized.
There is also a scarcity of theoretical and empirical research about the close relations
and interactions existing between motivation and emotion, outside the

psychophysiological hypotheses already mentioned. To consider some basic emotion

(fear) as a primary system of motivation does not enable us unless we expand on the
model to appreciate the role played by multiple human emotions, such as envy,
jealousy, compassion, hate, tenderness. These are all affective aspects of a not so
primary nature, which phenomenological observation and experience itself point to as
determinant factors in psychic activity and need to be born in mind in a motivational
explanation of motivation
4.4.4 Physiological, cognitive and behavioural responses to motivation

There are several motivational processes at the base of organisms' feedback
mechanisms, among which include physiological homeostatic regulation, activation
regulation and cognitive homeostasis.
The root of the concept of homeostasis can be found in research undertaken in the
middle of the 19th century by French physiologist Claude Bernard, who is considered
the forefather of experimental medicine. If the idea of homeostasis was due to Bernard,
the concept was proposed by American physiologist Walter B. Cannon, who studied
the regulatory role of the autonomic nervous system in the self-regulation of several
physiological processes. The concept of homeostasis (derived from the Greek words
homeo and stasis - similar position) was used by Cannon (1915) to refer to the
constant conditions or stable statuses maintained in the organism through coordinated
physiological processes. According to the definition provided by Cannon, homeostasis
is a functional concept. The definition refers to a goal or to a result to be achieved; the
stable state.
The majority of researchers who study homeostasis are interested in the mechanisms
that enable stable states to be maintained. Almost all of them agree that these
mechanisms make up a negative feedback or negative regeneration model, which
implies that if there is a diversion in one direction, there is a reaction in the opposite
direction. They are negative regeneration self-regulatory systems that maintain
constancy in the human internal environment. Loss of stable state creates conditions
that influence the return to this state, in the same way that a constant temperature in a
home is controlled and regulated with a thermostat that controls a heating appliance,
producing hot air, and an air conditioner, producing cold air.
The basic idea of homeostasis is that it achieves the stable state, through regulatory
mechanisms that correct diversions with respect to a previous stable state. Stable
states are defended when deviations are detected in a controlled quantity and are
corrected by compensatory mechanisms that oppose change.
The homeostatic model explains action on the level of physical activity, in the
physiological aspect as well as the motor or behavioural aspect, inciting behaviours
including breathing, ingesting food and liquid substances, regulating temperature, and
avoiding potentially hazardous stimulant situations, which enables the organism to
adapt to external changes and which are essential to survival.
The theory contributed by Lewin (1952) may be considered a bridge between the first
motivational theories, based on the homeostatic model of reducing stress, based on
retroactive mechanisms (theories by Freud, de Hull and behavioural scientists) and

current theories in human motivation, based on proactive mechanisms. Of these, Lewin

maintains hedonism and homeostasis, but makes an important advance, projecting his
theory towards a cognitive orientation, which is indicated by the interest in
psychological requirements more than biological requirements.
In order to analyse the neuro-physiological foundations of motivation, the main data

available are organised under the following headings: 1) The reinforcement
motivational system, and 2) physiological activation.
1. The reinforcement motivational system
Physiological bases of reward and aversion: Today, psychophysiologists share the

belief that experiences of pleasure and displeasure are not located in small specific
structures in the brain, but that numerous structures in the limbic system and the brain
are involved. The three main limbic circuits are the mammillothalamic tract, the
hippocampal circuit and the amygdaloid circuit.- Neurochemical reward bases
Recent discoveries derived from the neurobiological application to the explanation of

motivated behaviour have enabled several models to be formulated. It is currently
believed that the main neurotransmitter circuits associated to the experience of
pleasure and reward are those of dopamine and endorphins.
Electrical stimulation is particularly effective in the medial prosencephalon band, which

covers the lateral hypothalamus.
The theory of the opponent process reveals the automatic action of a series of cerebral
counter-mechanisms in both positive and negative emotional experiences. Thus,
the sensations of pleasure and aversion can be acquired.
2. Physiological activation
Cerebral structures and activation systems: The two most important cerebral structures
involved in activation physiology are reticular formation and the hypothalamus.
Reticular formation performs a primordial function in activation through its two
components: The reticular activation system and descending reticular formation. The
hypothalamus firstly regulates a series of hormones in the endocrine system. It also
regulates the autonomic nervous system, controlling sympathetic and parasympathetic
responses. The anterior hypothalamus relaxes the organism, whereas the posterior
hypothalamus activates it.
The optimal activation level: The activation level is relevant, to say the least, in a
double aspect: In relation to execution or accomplishment and the hedonic tone
experienced by the subject.
4.4.5 Measurement techniques for motivation

There are several scales that measure the motivation scales of different domains and
for specific users. They are mainly focused on extracting the intrinsic and extrinsic
motivational measurements. The most relevant amongst them:

Academic Motivation Scale (EME) (Robert J. Vallerand et al, 1992b): A

questionnaire for measuring the intrinsic, extrinsic, and motivation in education.
based on the tenets of self-determination theory and is composed of 28 items
subdivided into seven sub-scales assessing three types of intrinsic motivation
(intrinsic motivation to know, to accomplish things, and to experience
stimulation), three types of extrinsic motivation (external, introjected, and
identified regulation), and a motivation.
Elderly motivation scale (EMS-72) (Robert J. Vallerand et al, 1992):This scale

assesses the intrinsic and extrinsic motivation of elderly individuals. It assesses
4 constructs: intrinsic motivation, self-determined and non self-determined
extrinsic motivation, and motivation. It is subdivided in 6 life domains for which
elderly individuals are asked to complete 4 items (one item per construct) for
each of 3 different situations within each life domain. Thus, there are a total of
12 items per life domain and a grand total of 72 items for the whole scale. Each
item is assessed on a 7-point scale. The questionnaire is suited to look at
personal attitude toward six different life domains: our health, biological needs
(eating, sleeping), relations with others (family, friends, and people in general),
religion, leisure and information (news).
Work Motivation Scale (Brady & Reinink, 2001).The WMS was designed as a
self-report instrument to be used in career development, pre-employment,
human resources, vocational counselling, job development, work adjustment,
job satisfaction, job retention, and disability management. It assists individuals
in career development and planning by helping them understand their work
motives and values and apply that understanding to their career choices and
preferred work environment. The Work Motivation Scale is a revision of the
Work Orientation Values Survey (WOVS), published in 2002.
Bernard et al. (2008) report the development of the Assessment of Individual

MotivesQuestionnaire (AIM-Q), a new instrument based on an evolutionary
psychology theory of human motivation. It provides multitraitmultimethod
(MTMM) assessment of individual differences on 15 motive scales. A total
heterogeneous sample of N =1 251 participated in eight studies that evaluated
the homogeneity, internal consistency, testretest reliability, and MTMM
convergent and discriminant validities of the AIM-Qs three methods. methods.
These studies generally support the overall strategy of assessing individual
differences in multiple evolutionary-based motives with multiple methods.
Task engagement is evaluated in the Dundee Stress State Questionnaire

(DSSQ; Matthews et al., 1999; Matthews et al., 2002; Helton et al, 2009), see
also the section on Stress.

4.4.6 Motivation as cognitive moderator
4.4.6.1 Motivation affecting memory

An experience carried out by Erikson (2002), postulates that motivation only affects
memory when it is presented at the time of encoding (working memory) and has no
significant effect when presented at recall, or during the process of consolidation. De
Fockert, et al.(2001) suggests that motivation can only affect active processes of
memory. Other studies confirm that motivation affects working memory (Heinrich,
1968).
4.4.6.2 Motivational affecting on attention

Shifting attention in space may be an important factor in determining how the organism
will perceive and interpret environmental information, and thus it may represent an
early influence on goal-directed behaviour. Derryberry and colleagues, (1994, 1995)
have shown that both experimental manipulations of motivational state and individual
differences on trait dimensions of neuroticism and extroversion can influence
attention and information processing. Normal volunteers, after fasting for 16 to 22
hr, were more likely to allocate attention to food-related words than neutral words on a
dot probe attention task. This effect was not observed in nonfasting participants,
suggesting that changes in attention took place following the induction of a hunger
state. Finally, literature indicates that selective attention to threatening information
supports and sustains maladaptive patterns of information processing
characteristic of anxious and depressive states. These studies suggest that selective
attention is sensitive to affective-motivational states. From this perspective, it is
plausible that biases in attention may occur in response to stress.
4.4.7 Motivation and elderly & disabled

Little research has been performed regarding elderly and disabled.
According to Thomas P. et al, 2008, loss of motivation is frequent in aged people
and sometimes is difficult to assess. It is associated with depression but not
exclusively. It overlaps with frontal brain aging consequences. Moreover, somatic
problems can interfere with a loss of motivation; loss of motivation is frequent with
exhausting chronic diseases, cognitive impairment and it facilitates the social
withdrawal. In elderly, loss of motivation blunts the intensity of moral pain. Depressions
with conative disorders can move in a dysexecutive syndrome. Even if, owing to the
treatment, the cognitive disorders disappear, several years after, an authentic dementia
can be diagnosed.
With aging process, elderly must cope with losses, less or more balanced: affective,
physical and financial losses, triggering the reject of any personal commitment. Both
loss motivation and executive disorders have to be systematically assessed.
Loss of motivation differs from a mood disorder.
Lack of motivation and social dependency in old people result from a failure of the
psychological mechanism of elaboration. Loss of motivation appears to be an
acquired and dynamic process resulting from the psychological difficulties experienced
by the individual due to the loss of autonomy and an increasing dependency which is
not accepted although, paradoxically, searched for. This ambivalence results in loss of

motivation and psychic regression which belongs to the non cognitive disorders of
elderly people.
Vallerand & OConnor (1989) suggest a theoretical framework for research on the
psychology of motivation in the elderly. This framework posits the existence of 4 types
of motivation (intrinsic, self-determined extrinsic, nonself-determined extrinsic, and
amotivation) that are assumed to have a number of consequences for adaptation and
well-being. The 4 types of motivation can be reliably measured and are related to other
important aspects of the lives of elderly people. The 4 motivational scales are:
amotivation,
nonself-determined extrinsic motivation,
self-determined extrinsic motivation,
intrinsic motivation.
They list correlation coefficients between the 4 kinds of motivation in elderly and they
are:
residential constraint,
self-esteem,
depression,
life satisfaction,
meaning in life,
general health,
psychological status,
activeness.
4.4.8 Motivation modeling approaches
A generic connectionist model of motivation is proposed in (Portegys, 1999) and might

be used for simulation purposes.
According to Geen (1994), motivation refers to the initiation, direction, intensity and
persistence of human behaviour.
Intensity of motivation may be thought of as the momentary magnitude of motivational

arousal (Brehm & Self, 1989). Where the magnitude of motivational arousal concerns
the total amount of effort a person would make to satisfy a motive, and this effort could
be spread over time, the intensity is the magnitude at a point in time. Thus, where a
high level of motivational arousal is spread over a long period, the intensity of

motivation could always be low. However, where the magnitude of motivational arousal
is high and must be concentrated within a brief period, the intensity of motivation must
be great. It is the difference, for example, between moving 100 pounds of books one
book at a time or all at once.
While the factors that determine the magnitude of motivation, or the total effort one is
willing to make, are not the central concern of the review of (Brehm & Self, 1989), they
nevertheless must be understood because they set a boundary on the intensity of
motivation. Brehm & Self (1989) adopt the simple position that whatever factors affect
the effort one is willing to make to satisfy a motive are in fact the determinants of the
magnitude of motivation. In general, these factors are internal states such as needs
(e.g. food deprivation), potential outcomes (e.g. acquisition food, experience of pain),
and the perceived probability that some behaviour, if successfully executed, will satisfy
the need, produce or avoid the outcome. As in typical expectancy-value models of
motivation, Brehm & Self (1989) assume that needs and/or potential outcomes vary in
magnitude or value, and that the magnitude of motivation is a multiplicative function of
need, value of the potential outcome, and the perceived probability that a properly
executed behaviour will produce the desired effect. This model, which will be
recognized as a simple version of many present-day theories of motivation, depicts
what Brehm & Self (1989) call potential motivation. The reason it is called potential as
opposed to actual motivation is that it is not a sufficient set of conditions or the
specification of motivational arousal.
The direct function of motivational arousal is not the satisfaction of needs or the
avoidance or acquisition of potential outcomes. Rather, it is the production of
instrumental behaviour. When the difficulty of instrumental behaviour surpasses ones
capacities or outweighs the value of the potential gain (need reduction, outcome
attainment, or outcome avoidance), there will be little or no mobilization of energy. The
greater the potential motivation, the greater is the amount of energy that a person will
be willing to mobilize.
Because the function of motivational arousal is the production of instrumental

behaviour, measures of arousal deemed most reflective of motivation involve the
sympathetic nervous system, which prepares the organism for activity. Specifically,
cardiovascular changes in response to beta-adrenergic stimulation have been linked
with effortful coping (Obrist et al 1978). Laboratory attempts to document the
relationship between task difficulty and motivational arousal has focused on
cardiovascular reactivity (Brehm & Self, 1989) as measured by changes in:
heart rate (HR),
systolic blood pressure (SBP),
diastolic blood pressure (DBP).
These changes have been noted not only during coping behaviour, but also
immediately prior to instrumental activity, indicating that motivational arousal varies in
anticipation of task demand.

Kreibig (2010b) gives a brief outline of the physical components of cardiovascular

response (CVR) in emotion. She then reviews theoretical positions on the organization
of CVR in emotion and its motivational function. Finally, she considers a systematic
integration of emotion, motivation, and CVR. CVR precedes emotion, is a response
component of emotion, and persists after the occurrence of an emotion. A similar
relation is true between emotion and motivation: motivation can be seen as a cause of
emotion, as one of its major aspects, and as one of its consequences (Frijda, 2000).
The effort invested in goal-striving, i.e., motivational intensity, depends on

subjective task difficulty and the importance of success (Kreibig et al., 2010;
Brehm and Self, 1989; Wright and Kirby, 2001): motivational intensity is proportional to
difficulty (the difficulty law of motivation, Ach, 1935), while importance of success
marks the upper limit of what a person is willing to do in order to succeed (i.e., potential
motivation). Beyond this point, success is either viewed as impossible or to require
more effort than warranted by its importance and hence, all effort is withdrawn. These
predictions apply in case of clearly defined and known performance standards (i.e.,
fixed difficulty), where individuals can estimate how much effort is required to
succeed in a task. When a task calls for a subjects best performance (i.e., unfixed
difficulty), individuals cannot estimate how much effort is required to succeed and
hence motivational intensity is predicted to correspond to the maximum of justified
motivation or the importance of success (Richter & Gendolla, 2009). In relation to
Obrists (1976, 1981) active coping approach to cardiovascular adjustments,
motivational intensity has been operationalized by means of b-adrenergically
influenced cardiovascular indices, particularly:
pre-ejection period (PEP),
systolic blood pressure (SBP).
Typical results demonstrate shortened PEP and increased SBP, andwith less
consistencyincreased diastolic blood pressure (DBP) and heart rate (HR) with
increasing difficulty level (Kreibig et al., 2010): Low cardiovascular reactivity is
observed for tasks that are easy, impossible, or not worth investing the necessary
effort; high cardiovascular reactivity is observed in conditions of high or unspecified
difficulty, interpreted to index importance of success (e.g., Richter and Gendolla, 2009;
Richter et al., 2008). Conditions of high self-involvement as contrasted to low self-
involvement have been demonstrated to increase cardiovascular reactivity for difficult
and unfixed tasks, indicating an increase of importance of success (e.g., Gendolla et
al., 2008). Other research has also related
electrodermal activity (EDA) to effort mobilization and
skin conductance response rate (SRR) to task engagement
(e.g., Gendolla and Richter, 2005; Pecchinenda and Smith, 1996).
A generic connectionist model of motivation is proposed in (Portegys, 1999) and could

be used for simulation purposes.

For workplace purposes, Klein (1989) proposed a model that integrates several
different motivational theories and previous control theory. The proposed model is
dynamic, parsimonious, and focuses on self-regulation and the underlying cognitive
mechanisms of motivation. In explicating this model, numerous hypotheses are
derived regarding (a) the nature of goals and feedback; (b) cognitive, behavioural, and
affective reactions to goals and feedback; and (c) the role of attributions, expectancies,
and goal hierarchies in determining those reactions.
In the personal healthcare and well-being domain, a motivation model (based on the
work of (Atkinson et al., 1960) for rehabilitation of people with physical and/or
cerebral handicaps is considered in Geelen & Soons (1996). According to (Atkinson
et al., 1960), motivation is not a stable personality trait. Achievement motive behaviour
is determined by three variables: the subjectively estimated chance of a successful
completion (assuming that action is undertaken), the perceived value of a positive
result, and the perceived costs of taking the action (Geelen & Soons, 1996).
The motivation model by Atkinson is the following:
Motivation = F (Expectancy x Motive x Incentive).
This model is applicable only when an individual knows that his effort will be evaluated
(by himself or others) in terms of some standard of performance, and that the
consequence of this action will be either a success or a failure.
Birren and Schaie (1985) posit that Atkinsons model is suitable in the rehabilitation
setting. They reformulate the model as follows:
M = (P(Os) x V) : C,
where M is motivation for rehabilitation,
P(Os) is the subjective perception of the patient of the likelyhood of success,

assuming that he or she makes the required effort,
V is the individually perceived utility or value of a successful rehabilitation,
and C the perception of the cost of the outcome.
Maclean et al. (2002) attributed motivation to stroke patients on the following basis:
behaviour (proactivity was equated with motivation, passivity with lack of

motivation)
compliance with rehabilitation (compliance was seen as indicative of motivation,

noncompliance as a lack of motivation).
The determinants of motivation were located partly in personality factors but also in
social factors.

5 Psychological and Behavioural VERITAS

Abstract User models
In the previous chapter, different psychological states (stress, fatigue, emotions and
motivation) that specially affect elderly and disabled people have been collected and
analysed. For each facet, the definitions and categories have been gathered, as well as
the descriptive models. An analysis of the stimulus and causes that trigger a
psychological state and the physiological, cognitive and behavioural human responses
that are triggered by the psychological state has been also performed.
In addition, we have study how psychological states affect on cognitive attributes

(reaction time, memory, attention, etc), analysing relevant psychological studies in
order to extract quantitative and qualitative metrics, parameters and rules.
An analysis of computational models (overlays in ACT-R and other cognitive

architectures, mathematical models, etc) of the P&B states and facets has been also
produced, focusing when possible on psychological and behavioural models for
VERITAS users and domains.
The Psychological and Behavioural Abstract User model is a compilation of all this
information, in which for each psychological state, the category, description, types, ICD
classification, causes, age-relation, ICF functional limitations and the physiological,
cognitive and behavioural responses are addressed. The objective of the AUM is to
present the next layer of detail of the framework, a level at which it is meaningful to
taxonomize and compare different contributions from the literature. To do so we have
adopted a common semantic and structure, that is also consistent with the physical and
cognitive Abstract User Models.
However, specific measurements and metrics of how a psychological state affects

people and more specifically elderly and disabled havent been easy to find.
Quantitative values and parameters do not always exist, and the information found is
mostly qualitative. In addition, literature often provides imprecise or even contradictory
data, so a careful revision analysis had to be performed in order to obtain a consistent
Abstract User Model. The same happens with reliable computational and
mathematical models that can be used as overlays of cognitive architectures, either
they do not exist or they are incomplete.
Since specific metrics and values are difficult to gather in this domain, in a second step
the affected attributes, existent computational & theoretical models and rules or
possible alternative measurements have been also compiled. The purpose of the
following year will be to fine tune all the information gathered in order to create the final
Virtual User Model with concrete P&B models that contain models, standards,
guidelines and rules for the older people and people with disabilities.
In this chapter, the tables with the Abstract User Model information and the approach
that has been followed for creating the P&B ontology have been included.

5.1 Psychological and Behavioural Abstract User Models tables

The following table includes a first analysis concerning the main dimensions of each psychological facets: category, description, type of
the psychological state, ICD classification if its available, the causes-instigators-stimuli that trigger the psychological states, relationship of
the psychological state with aging, functional limitation when there are, and the physiological, cognitive and behavioural/emotional
responses that are triggered in presence of the psychological state. The table (Table 21) is fulfilled with information extracted from
previous chapter and gives a full overview of the dimensions of the psychological states, although it does not include specific parameters,
models and metrics, which are left for a second analysis (see Table 22, Table 23, Table 24, Table 25).
P&B Short Type ICD Instigators- Age- Functional Physiological, Cognitive and Behavioural/
Category description classification Stimuli-Causes related limitations (ICF Emotional Responses
(If applicable) Classification)
Physiological Cognitive Behavioural/
(If applicable) Response Response Emotional
Response
Stress The adaptive Eustress: Experienced right No -b151 Digestive Adrenaline Alertness Favours
response of Positive before you have functions release. increases. selection of
the body Short Term the need to exert appropriate
when facing Stress physical force: big -b555 Endocrine Heart rate Senses behaviour
various game, competition gland functions increases heighten.
stressors. Is (Heart beats (Perception) Exciting
b3108 Voice faster) mood/emotio
a body functions Increase
protection n
Small arteries processing
against -b830 Other contraction. information. Improves
physical & functions of the (Thickens performance
emotional skin blood) Increases
pressure and search of Pleasant
danger. -b4100 heart Pumping blood information. anxiety
Follows an rate into muscles
Inverted-U and organs.
b420 Blood
model
pressure Blood
concerning
functions pressure
cognitive and
task- b730 Muscle increases
performance, power functions Interrupts
depending
s5803 Adrenal functioning of

Response
on the stress gland digestive

arousal system.
b156 Perceptual
functions Sweating
b144 Memory Muscle tension
functions
Rapid
b152 Emotional breathing
functions
Voice changes
Acute F43.0 Acute Immediate No b1646 Problem- Activates the Increases If Extreme
Stress: stress reaction perceived threat, solving sympathetic mental activity arousal:
Negative either physical, division of the Disorientatio
Short Term F43.1 Post- emotional or autonomic Stress impairs n.
Stress: traumatic stress psychological nervous prefrontal cortex
disorder b1648 Higher- (PFC )cognitive Freeze
(Degree level cognitive system
depending function
F43.2 functions, other Increase level Agitation
on Adjustment specified Difficulty to
arousal). of cortisol Confusion
disorders retrieve long-
Model: Increase level term memories Befuddled in
R45.7 State of adrenaline
inverted emotional shock b1301 Affects on severe crisis
U-shaped and stress Motivation Increased memory
curve. heart rate consolidation.
Z73.3 Stress, b144 Memory
not elsewhere functions Increase Affects on
classified breathing rate attention:
reduces the
b1441 Long- Higher blood attentional
term memory pressure focus.
b1442 Retrieval Pumping blood

of memory into muscles

Response
b1565 and organs.

Visuospatial
perception Decrease fine
motor
b1440 Short- coordination
term memory performance
b140 Attention Hypothalamus
functions of the brain
which secretes
b152 Emotional Adrenocorticot
functions rophic
b1560 Auditory releasing
perception Hormone ARH
b3108 Voice Breakdown of

functions glycogen
stores in liver
b114 Orientation and muscle
functions
Formation of
b147 more glucose
Psychomotor
functions Increases
strength of
skeletal
muscles
Decreases
blood clotting
time
Reduces
intestinal
movement
Inhibits tears,

Response
digestive
secretions.
Relaxes the
bladder
Dilates pupils
Increases
perspiration
Affetcs
speech: pitch
and voice
intensity
Long Term F43.1 Post- Constant stressors Yes: b1646 Problem- Decalcification Decreases Anxiety,
Stress/ traumatic stress in a long period of 30% solving of bones concentration nervous
Chronic disorder time: elderly (calcareous
stress have deposits) Affects memory: Depression
F43.2 Person moving had declarative, (loneliness,
Adjustment from city. b1648 Higher- Negative spatial,etc frustration..)
chronical level cognitive
disorders ly high action on
Work stress functions, other immune Decreases Fear
F43.8 Other cortisol specified attention to
Work changes levels. system. Worried
reactions to details
severe stress endocrine Pacing
system Posttraumatic
Z73.3 Stress, b1301 stress
Motivation response Increase of
not elsewhere decreases alcohol
classified b144 Memory Release of memory task- taking
functions corticosteroids dependent on
. hippocampus disturbances
b1441 Long- of sleep
blood pressure Deteriorates

Response
term memory is raised declarative Forgetfulnes

verbal memory s
b1442 Retrieval muscles
of memory tiredness Facilitates Poor
procedural judgment
b1565 digestive memory learning
Visuospatial difficulties (ie:fear) Reduced
perception creativity
dizziness Impairs spatial
b1440 Short- information Less
term memory reduced fantasizing
hippocampal memory.
b140 Attention volume Affects synaptic Preoccupatio
functions plasticity. n
33% Elderly
b152 Emotional users with high Deficits in Blocking
functions cortisol levels: hippocampus-
Hippocampus Reduced
b1560 Auditory dependent productivity
averaged 14% memory tasks
perception smaller) Errors in
b3108 Voice Impairment of math and
levels of stress hippocampus-
functions hormones rise grammar
dependent
b114 Orientation with aging learning and
functions memory
b147 Diminished
Psychomotor problem solving
functions
Elderly with
b1645 stress: Worst
Judgement memory
performance
b172 Calculation and worst
functions performance in
tasks involving
b1340 Amount

Response
of sleep hippocampus.
Elderly with
diabetes
causing
cognitive
impairments and
higher cortisol
levels were
significantly
associated with
poorer working
memory (letter-
number
sequencing) and
processing
speed (digit
symbol test)
Fatigue Body and Physical Lifestyle alterations Yes: b1340 Amount Concentration Deterioration of Labor force
mind fatigue (increased activity, increase of sleep of ATP in sustained decreases
exhaustion nutritional d nerve cells attention level
that occurs deficiencies, sleepine b198 Mental decreases Performance
as a result of increased need for ss functions, other Deterioration of drops
work or energy). during specified Increase of complex (longer
effort. It is the day, urea content in information response
b420 Blood blood. processing time, longer
characterize and also pressure
d by the fatigue, lapses)
functions Hypertension Slower Reaction
inability to tirednes time
perform s, and b440 Formation of
physical lack of Respiration the gamma Deterioration of
tasks with energy, functions amino butyric Procedural
the usual are acid in motor Memory
force and a normal b730 Muscle centres

Response
relented signs of power functions increases

speed in the an (protective
rational advance b4100 heart rate function)
processes d age b7602
which may Significant
Coordination of alterations in
cause voluntary
memory lung breathing
movements and blood
failure.
b151 Digestive circulation
functions systems
b5500 Body Reduction of

temperature muscles
tension
b140. Attention
functions EEG (increase
of sleepiness
b164 Higher- is associated
level cognitive with increased
functions theta and
alpha activity )
b750 Motor
reflex functions Slower Eye
Movement
b144 Memory (0,1-0,6 Hz)
functions
Palpitations
Gastric pain,
ulcers, colic
Feeling of
suffocation
Lower Body
Temperature

Response
Pain in the
lymph nodes
located under
the armpits
and in the
neck
Mental Sleep disturbance Yes b5451 Mineral Decalcification Decreases Laxity

fatigue balance of bones concentration sensations
(calcareus
b4350. Immune deposits) Affects memory: Mood
Lifestyle alterations response declarative, changes
(increased activity, Negative spatial,etc
nutritional d2401 Handling action on labour force
deficiencies, stress immune Difficulty decreases
increased need for system. recollecting
energy). b140 Attention information, Performance
functions High level of speaking drops
Stress or clumsiness (longer
b144 Memory response
functions Emotion]
Making more time, longer
d330 Speaking mistakes than lapses)
normal.
b610 Urinary Inability to
excretory focus
functions Learning
b152 Emotional difficulties
functions
d198 Learning
and applying
knowledge,
other specified

Response
Negative Fear Anticipation of a Yes: Global mental Migraines Raise of Stress

emotions threat or danger functions (b110- attention:
that causes Older b139) Stomach Focusing nearly Depression
anxiety, uncertainty people Pains exclusively on
experien b151 Digestive Anxiety
or insecurity. Trembling the perception of
ce functions the feared Insomnia
Conditioned stimuli stronger High blood stimulus.
to a fear reaction direct b755 Involuntary Panic
movement pressure
may be completely feelings Raise of
varied and be of of reaction Temperature perception Avoids
course lacking in anxiety functions changes danger:
objective danger. in the Because the
b420 Blood Facial visual system Motivates
form of pressure
-New and changes has limited towards
fear or functions
mysterious capacity, protection
being
situations, Acceleration of emotions such
scared. b5500 Body It causes
especially in heart rate, as fear may play
temperature increase in its apprehensio
children (Schwartz, Alzheim an important
conductance n, unease
Izard & Ansul, er & b140 Attention role in guiding
and fluctuation and
1985). elderly: functions the selection of
(Cacioppo et discomfort.
emotion relevant input
-Visual abyss al b210 Seeing al., 1993). (LeDoux, 2000;
(Gibson and Walk, dysregul functions hman &
1960) in children, ation. Mineka, 2001).
as well as height d330 Speaking
So, fear can
and depth Higher guide the
d177 Making
(Campos, Hiatt, levels of organism by
decisions
Ramsay, negative influencing: (I)
Henderson & affective attention
Svejda, 1978). states in processes, (II)
elderly

Response
-Secondary are awareness of

evaluation associat the visual input,
processes that ed with and (III) affective
interpret a situation poorer processing
as being memory independent of
dangerous awareness.
(Lazarus, 1977,
1991a). Increased
Reaction time
-Pain and pain
anticipation Speed up the
(Fernndez & Turk, decision making
1992). procedure
-Loss of support Reduction in the

(Watson, 1920) efficiency of
and, in general, cognitive
sudden change in processes,
stimulus. obnubilation
Brain has
evolved to react
more quickly to
potentially
threatening
situations.
(BBC news,
October 15,
2007)
Speech and
language
distortion.

Response
Sadness -Physical or Yes b140 Attention Debility Reduce reaction Depression

psychological functions time
separation, loss or Alzheim Headaches Insomnia
failure (Camras & er & b152 Emotional Reduce
elderly: functions High, attention Panic
Allison, 1989). sustained
emotion Fatigue
-Disappointment, al b1343 Quality of neurological With sadness,
particularly if hope dysregul sleep activity increases in Loneliness
in something has ation. (Reeve, 1994). limbic-paralimbic
b1340 Amount blood flow Pessimism
been dashed. of sleep -Light increase
Higher (subgenual
-Situations of levels of in heart rate, cingulate, Motivates
defencelessness, negative blood pressure anterior insula) towards new
lack of prediction affective and electrical and decreases personal
and control. states in skin resistance in neocortical reintegratio
According to elderly (Sinha, Lovallo regions (right n.
Seligman (1975), are & Parsons, dorsolateral
sadness appears associat 1992). prefrontal,
after an experience ed with inferior parietal)
which has caused poorer were identified.
fear as sadness is memory
the opposite (Helen S.
process to panic Mayberg et al.
and frantic activity. Reciprocal
Limbic-Cortical
-Lack of reinforced Function and
activities and Negative Mood:
adaptative Converging PET
behaviour Findings in
(Lewinsohn, 1974). Depression and
Normal
-Chronic pain Sadness)
(Sternback, 1978,
1982). Focusing
attention on the

Response
internal
consequences
of the situation
(Stein & Jewett,
1986).
-Unpleasant stimuli Yes Increase in NA Motivates us

(mainly chemicals) gastrointestina to reject that
that is potentially Alzheim l reaction. which lies
dangerous or er & ahead.
annoying. elderly: Muscular
emotion tension Nausea
-Aversively al
conditioned CS. dysregul
US is usually ation.
related to smell or
Aversion/Disgust taste. Higher
levels of
negative
affective
states in
elderly
are
associat
ed with
poorer
memory
Anger Confronting Yes Global mental Headaches Reduced Insomnia

frustrating or functions (b110- attention:
adverse events Alzheim b139) Migraines Focusing Leads us
and to those er & attention on the towards
elderly: b420 Blood

Response
interpreted as emotion pressure Trembling external destruction

premeditated, al functions obstacles that
unjustifiable and dysregul Intense prevent goal
avoidable. ation. b140 Attention cardiovascular achievement or
functions reactivity are responsible
Aversive Higher (elevation in for frustration
stimulation, levels of d177 Making the indices of
decisions (Stein & Jewett,
physical, sensory negative the heart rate, 1986).
and cognitive affective b114 Orientation systolic and
(Berkowitz, 1990). states in functions diastolic Reduced
elderly pressure) perception
-Conditions that are d330 Speaking (Cacioppo et
cause frustration associat al., 1993). Increased
(Miller, 1941), ed with b1343 Quality of Reaction time
interruption of poorer sleep
motivated Speed up the
memory decision making
behaviour, unfair b1340 Amount
situations (Izard, of sleep procedure
1991), or attacks Loss of
b156 Perceptual
against moral orientation
functions
values (Berkowitz,
1990). Speech and
language
-Extinction of the distortion.
operant, especially
in continuous
reinforcement
programs (Skinner,
1953).
-Immobility
(Watson, 1925),
physical or
psychological
restriction
(Campos and

Response
Stenberg, 1981).
Emotions Joy Fun, euphoria, Alzheim b140 Attention -Increase in Favours Behaviour is
with Positive gratification, er & functions activity in the cognitive motivated.
Valence contentment; elderly: hypothalamus, performance,
provides a emotion b156 Perceptual septum and problem-solving Motivates a
sensation of well al functions amygdaloid and creativity desire to
being and security. dysregul nucleus (Isen & reproduce
ation. (Delgado, Daubman, that which
Accomplishment, 1992) 1984), as well made us feel
successful as learning and good).
achievement of a -Increase in memory (Nasby
person's goals. heart rate, favours
& Yando, 1982) interpersonal
even if the with mild
-Congruence cardiovascular relations.
between what a arousal.
Positive reactivity is Feelings of
person wants and lower than in Very intense
emotions what they have, vigorousnes
other states of s,
between emotions, happiness may
expectations and competence,
such as anger slow down importance
current conditions and fear execution and
and comparison and freedom
(Cacioppo et even lead to (Meadows,
with others al., 1993). overseeing an
(Michalos, 1986) 1975).
important
-Increase in element in -Sense of
respiratory problem-solving self-esteem
rate (Averill, and it may and
1969) interfere with confidence
creative thinking (Averill &
(Izard, 1991). It More, 1993).
may also reduce

Response
attention and
perception by
same reasons.
Surprise Shock, Alzheim b114 Orientation Physiological Loss of Motivates

astonishment, er & functions pattern orientation. towards
confusion. Surprise elderly: characteristic orientation
is very transitory. emotion of orientation Working
al reflex: memory and
Unexpected stimuli dysregul decrease in attention used to
ation. heart rate. process the new
information.
-Facilitate
attention
processes,
exploratory
behaviour and
interest in the
new situation
(Berlyne, 1960).
-Direct the
cognitive
processes to the
situation which
has arisen
(Reeve, 1994).
-General
increase in
cognitive
activity.

Response
Motivation The impetus Intrinsic Homeostatic drive: Yes, b4100 heart rate Sympathetic Expectancies Reduce
that starts, motivation Internal Stimuli: loss of Nervous and value pressure.
guides and (internal emotions, motivatio b420 Blood System: attributed to
maintains motivators): biological needs, n is pressure behaviours. maximize
behaviour to psychological frequent functions Heart rate the rewards
reach a needs. in elderly (HR) and
Functions of the minimize the
desired goal (Thomas cardiovascular Definition of
or objective. Expectations P.,2008) pain
system (b410- which (Pinder,
Valences . b429) Systolic Blood behaviours
Pressure 1985
Intrinsic comply with
Marlows pyramide: (SBP) functional
Physiological, factors:
purposes and
Safety,Belonging, residenti which not. Depending
Esteem, Self- al Diastolic Blood on achieving
actualisation constrai Pressure expectations
nt, (DBP) : positive
Cognitive and negative
self- Dissonance emotional
esteem, experiences.
Cardiovascular Thus, the
depressi Response Self perception sensations
on, (CVR) of self-efficacy of pleasure
life and aversion
satisfacti can be
on, Electrodermal acquired.
activity (EDA)
meaning
in life,
general Skin
health, Conductance
Response
psycholo Rate (SRR)
gical

Response
status,
activene
ss. (SRR)
Extrinsic Appetising/Adverse No b4100 heart rate Sympathetic Reduce

motivation elements Nervous pressure.
(external b420 Blood System:
motivators) pressure maximize
Expectancies
functions Heart rate the rewards
and value
(HR) Systolic and
Functions of the attributed to
Blood minimize the
cardiovascular behaviours.
Pressure pain
system (b410- (SBP) (Pinder,
Definition of
b429) 1985
which
Diastolic Blood
behaviours
Pressure
comply with
(DBP)
functional
Cardiovascular purposes and
Response which not.
(CVR)
Cognitive
Electrodermal Dissonance
activity (EDA)
Self perception
Skin of self-efficacy
Conductance
Response
Rate (SRR)
Table 21 Psychological and Behavioural Abstract User Models

After this first analysis, a more profound study taking into account existent models, parameters and metrics has been also performed. This
second analysis is reflected in the following tables, one for each psychological state. The first column represents the psychological facet.
Second column reflects the existent metrics founded from psychological studies: affected attributes, quantitative parameters and
qualitative rules. The third column lists the most representative models of the psychological state. When possible, the related rules and
parameters are referenced. In the last column the existent measurement techniques for psychological states responses: questionnaires,
scales, physiological measurements, etc are listed. The questionnaires and metrics are referenced when available.
P&B Existing Metrics from Psychological analysis Computational Models: Rules and parameters Questionnaires and other
STATE methodology to measure them
(Qualitative or Quantitative)
(Specify for different domains and users
Affected Values, rules Rules Metrics if available)
attributes (AA)
(specify if values change for a (specify if rules change for (Found params,
specific user), rules) a specific user) wanted params, ACT-
Bibliography. R params)
Stress AA: Reaction Time STRESS MODEL by Janis and Mann Stressors Scale (full
(Different information section 4.1.6.1)
Levels of For young and non stressed: (Irving, L. J., Mann, L. 1977).
Arousal) Social Readjustment Rating Scale
mean auditory reaction times = 140-160 msec Rule: Stress vs Performance/Decision (SRRS), Thomas Holmes, Richard
Effectiveness Rahe (1967)
visual reaction times = 180-200 msec
iSTRESS: Life Experiences Survey (LES),
(Galton, 1899; Woodworth and Schlosberg, 1954; (Sarason et al., 1978) Hassles Scale
Fieandt et al., 1956; Welford, 1980; Brebner and 1 unconflicted adherence, select mt = mt-1 (Kanner et al., 1981)
Welford, 1980)
2 unconflicted change to next Sm (next state) Score of 136+: Very High
time to touch is intermediate, at 155 msec Level of Stress,
(Robinson, 1934). 3 m = 1, M whichever is Best Reply (vigilance)
Score of 116-135: High Stress,
Reaction time follows an inverted-u 4 near panic so, m = 1 of 1, M (if highly
model(Welford, 1980; Broadbent, 1971; Freeman, experienced users: Recognition Primed Decision Score 76-115: Average stress,
1933): making. Also, defensive avoidance occurs at this
level for non experts). Score 56-75: Low stress,

attributes (AA)
If level of arousal (stimuli/stressors) is lower or > 4 panic, so Sm = avoid the stressor (hyper- Score 51-55: Very low stress
greater than the optimum, then Reaction time vigilance)
increases.
Metrics
If level of arousal (stimuli/stressors) is optimum, then Subjective Stress Scale (SSS) for Mildly
Reaction time decreases (Stressor level, pressure time, risks) Intellectually Handicapped (Paul
Bramston and Gerard J. Fogarty, 1995):
Neurons responding to the aversive visual stimuli Output: Performance/Decision Effectiveness
happened within about 120 ms [Adolphs] For each item, the response categories
No quantitative metrics. are: Event not experienced, no stressful
If aversive (threatening) visual stimuli, then Visual_ event, some stress, moderate stress, a
reaction_time=120 great deal of stress.
Elderly: Automotive:
Simple reaction time increases slowly until the 50s The State Driver Stress Questionnaire
and 60s, and then lengthens faster as the person is intended to assess the state
gets into his 70s and beyond (Welford, 1977; Jevas experience of driver stress; thus, it was
and Yan, 2001; Luchies et al., 2002; Rose et al., designed to be administered verbally in
2002; Der and Deary, 2006). actual driving situations Half of the
Stress Arousal Checklist items indicated
Luchies et al.(2002) also reported that this age effect positive mood (relaxed, contented,
was more marked for complex reaction time peaceful, comfortable, and calm) and
tasks. the other half indicated negative mood
ACT-R OVERLAYS
(tense, bothered, nervous, uneasy, and
AA: Sustained attention distressed).
(e.g., Gluck, Gunzelmann, Gratch, Hudlicka, & Ritter,
2006).

attributes (AA)
A dynamic model of stress and sustained attention WORKING MEMORY Workingplaces

can be represented by an inverted-U: an input stress
can vary between extreme values of underload and Rule 1:Modify the decay rate of working memory Job Stress Questionnaire (JSQ; Caplan
overload (hypostress and hyperstress). (Hanckock, objects. et al., 1975; Harris et al., 1999) is a 13-
1986c). item questionnaire designed to reflect a
Rule 2: Increase the declarative memory retrieval four dimensional "Person-Environment
Increasing discomfort (increasing negative stress) threshold parameter. Fit" model of job stress. Each of the
is followed by a rapid decreased in psychological dimensions reflects an aspect of the job
adaptability (attentional resource capacity). Rule 3: Decrease the base-level activation of all environment which, in interaction with a
Eventually, this is followed by a similar decrease in memory elements when stressed particular individual, is hypothesized as
physiological adaptability. being stress producing. These four
ATTENTION: dimensions are: (a) Workload, (b) Role
Conflict, (c) Role Ambiguity, and (d)
Rule 1: Wickens- WM overlay: Working memory Utilization of Skills, which represent four
capacity decreases. common sources of stress and do not
represent all types of stress.
Rule 2: Another way to implement an overlay that
decreases attention to the task is to create a A different version of JSQ
secondary task. This secondary task simulates worry. (http://www.stress.org/topic-
The rules that create this secondary task might be workplace.htm) has 11 items. If the JSQ
seen as architectural productions. This overlay score is between 10-30, the person
decreases performance rate on the serial subtraction handles stress well; 40-60, moderately
task. well; 70-100, he or she is encountering
problems that need to be addressed
PERCEPTION: and resolved.
Wickens-Perceptual Tunneling options: Cognitive-affective correlate scales

attributes (AA)
Rule 1. Modify the visual attention latency (full information section 4.1.6.2)
parameter.
Derogatis Stress Scale: DSP
Visual-attention-latency increases > 0.085. (Derogatis, 1980):
TSS=n t-scoresn
Rule 2. Decrease the default distance (from the SSS scores:

screen) parameter.
SSS= 0: no stress
Rule 3. Increase the saccade time parameter. (ACT-
R + EMMA, Salvucci, 2001) SSS 42 low subjective stress
Rule 4. Decreasing the perceptual width field 43 SSS 55 moderate

parameter. subjective stress
COGNITIVE TUNNELING: SSS >55 high subjective stress
Cognitive tunneling- (Wickens et al, 1998.) SSS=100: extremely highly

stressed
Rule 1: Declarative and procedural retrieval
thresholds in ACT-R can be modified to represent a Physiology stress response (full
greater reliance on well-known and well-practiced information section 4.1.6.3)
knowledge.
i) electrodermal techniques (EDA)
declarative retrieval threshold0. (eccrine sweat gland), (ii)
electromyographic techniques (EMG),

attributes (AA)
procedural retrieval threshold0. that measure muscle tension. The

muscle group frontalis and related facial
Rule 2: Noise in the procedural rule application muscles are prime targets of the stress
process can be decreased (i.e., activation noise and arousal and therefore the prime target
expected gain noise), which would lead to only the for measurement. (iii) Cardiovascular
most well-practiced materials being retrieved and measures, such as heart rate,
applied. peripheral blood flow, and blood
pressure. (ECG)
activation noise: ans> 0
Automotive & Workplaces
Expected gain noise EGN is modified. physiology
Silverman Model (Silverman, 2010), These techniques are often used for
detecting stress in drivers (Healey &
iSTRESS(t) = f{ES(t), TP(t), EF(t)} Picard, 2005) or in users at home
through ubiquitous intelligent sensing
systems Choi et al. (2006). Picard & Du
(2002).
ES(t) = =t- [positive-events(y) - adverse-events(y)]
t
/ Palmer et al. (2004) present a model of

work stress. The individual symptoms
0 <= ES(t) <= 1.0, ES can vary over the interval 0,1 are raised blood pressure, sleep and
with 0 being over-confident and unstressed, 0.5 is gastrotestinal disturbances, increased
neutral, and 1.0 is totally stressed. alcohol and/or caffeine and/or nicotine
intake, increased irritability & negative
TP = TI /( TI + TA) emotions, back pains, tension,
palpitations and headaches.

attributes (AA)
0 <= TP <= 1.0 Psychological responses
EF(t) = EF(t -1) + { i=1,I wi[PMFi(t)/(PMFi(t) + FTi)]/I} - Stanford Acute Stress Reaction
{EF(t-1)/10 if Questionnaire (SASRQ) (Cardena &
Spiegel, 1993; Shalev, Peri, Canetti, &
0=PMFi(t)} Schreiber, 1996).Ranges of Values per
different Studies.
0 <= EF(t) <= 1.0
- Acute stress over 0.8
- Threatening patient .87
- Vicinity of shooting .93 (.72.88 for

subscales)
- Sexual abuse survivors .93 (.78

.86 for subscales)
- Emergency rescue .8 (.64.98 for

subscales)
- Catastrophic flooding .95 (.82.88

for subscales
Dundee Stress State Questionnaire

(DSSQ; Matthews et al., 1999;
Matthews et al., 2002; Helton et al,

attributes (AA)
2009).
Elderly
Elders Life Stress Inventory (ELSI): The

ELSI is a self-report life event scale,
consisting of 30 items, each rated on a
six-point scale (05), particularly
developed for the elderly. Two types of
scores are derived from the scale, one
reflecting number of life events
experienced during the last year and
the other measuring the subjective
impact of those life events. The scale
has satisfactory psychometric
properties (Aldwin, 1991).
Acute AA: long term memory ( declarative and Previous models ( for different types of stress Stanford Acute Stress Reaction
Stress procedural) arousal) can be used here as well. Questionnaire (SASRQ) (Cardena &
Spiegel, 1993; Shalev, Peri, Canetti, &
High levels of cortisol impaired memory retrieval, but STRESS MODEL by Janis and Mann Schreiber, 1996).Ranges of Values per
only old, not recent, memories (McGaugh ). different Studies.
(Irving, L. J., Mann, L. 1977).
Retrieval of old memories performance decreases. - Acute stress over 0.8
Rule: Stress vs Performance/Decision

attributes (AA)
AA: Explicit episodic memory: Spatial Effectiveness - Threatening patient .87
Luethi, Meier and Sandi (2009) induced acute stress - Vicinity of shooting .93 (.72.88 for
with Trier social stress test (Kirschbaum et al., 1993). subscales)
Spatial episodic memory is increased. (Higher scores iSTRESS:
in Spatial memory test (t-values=2.07) - Sexual abuse survivors .93 (.78
1 unconflicted adherence, select mt = mt-1 .86 for subscales)
AA: Explicit episodic memory: Verbal
2 unconflicted change to next Sm (next state) - Emergency rescue .8 (.64.98 for
Explicit verbal memory was previously documented to subscales)
be negatively affected by stress and high cortisol 3 m = 1, M whichever is Best Reply (vigilance)
levels (Lupien et al., 2005; Sauro et al., 2003; Wolf, - Catastrophic flooding .95 (.82.88
2006), with strong evidence indicating that retrieval 4 near panic so, m = 1 of 1, M (if highly for subscales
processes are particularly vulnerable (de Quervain et experienced users: Recognition Primed Decision
al., 2000, 2003; Het et al., 2005; Roozendaal, 2002). making. Also, defensive avoidance occurs at this Dundee Stress State Questionnaire
Stress and glucocorticoids may have opposing level for non experts). (DSSQ; Matthews et al., 1999;
effects on explicit memory consolidation and Matthews et al., 2002; Helton et al,
retrieval, with enhancing effects on consolidation > 4 panic, so Sm = avoid the stressor (hyper- 2009).
and impairing effects on retrieval (e.g. Beckner et vigilance)
al., 2006; Roozendaal, 2002; Sandi and Pinelo-Nava,
2007; Smeets et al., 2008). Metrics
If Acute_stress & emotional_stimuli (Stressor level, pressure time, risks)
then { Explicit episodic memory: Verbal_ Output: Performance/Decision Effectiveness

consolidation is enhanced.

attributes (AA)
Explicit episodic memory: Verbal_retrieval is No quantitative metrics.

impaired.}
Luethi, Meier and Sandi (2009) induced acute stress

with Trier social stress test (Kirschbaum et al., 1993).
Explicit episodic memory (verbal) did not change.
If Acute_stress (TSST) & Neutral_emotional_stimuli,

then No changes on Explicit episodic memory:
Verbal.
Implicit/ Procedural Memory
Luethi, Meier and Sandi (2009)
If Acute_stress (TSST) & Negative_emotional_stimuli
then { Implicit Memory retrieval impaired; Classical

conditioning_scores t-value= -2.28/0.56), Implicit
ACT-R OVERLAYS
memory consolidation ( storage) enhanced;}
(e.g., Gluck, Gunzelmann, Gratch, Hudlicka, & Ritter,

2006).
If Acute_stress (TSST) & Neutral_emotional_stimuli,
then, No changes on Implicit Memory ;
WORKING MEMORY

attributes (AA)
AA: Working memory Rule 1:Modify the decay rate of working memory
objects.
Lupien and Schramek, 2006. Luethi, Meier and Sandi
(2009). Rule 2: Increase the declarative memory retrieval
threshold parameter.
Induced working memory impairment.
Rule 3: Decrease the base-level activation of all
Lower reading span. (t-value=-2.09) memory elements when stressed.
Working memory size decreases. ATTENTION:
AA: Attention Rule 1: Wickens- WM overlay: Working memory

capacity decreases.
Decrease: Hancock (1986), Wickens et al. (1998)
AA: Focus of attention & Decision Making decreases attention to the task is to create a
secondary task. This secondary task simulates worry.
With acute stress, stress narrows the focus of The rules that create this secondary task might be
attention Under time pressure, the subject adopts a seen as architectural productions. This overlay
simpler mode of information processing in which decreases performance rate on the serial subtraction
alternatives are not explored fully and certain task.
important cues are used to determine the decision.
PERCEPTION:
If High Level Acute Stress (Emergency, Threatening
situation) then {Attention_focus narrows: individual Wickens-Perceptual Tunneling options:
focuses just on critical issues and elements;
Individual_behaviour= risk-avoiding; Amount of Rule 1. Modify the visual attention latency

attributes (AA)
attended tasks decreases; Eliminates nonessential parameter.

information; } (Pittsburgh Research Laboratory)
Visual-attention-latency increases > 0.085.
AA: Emotions
Rule 2. Decrease the default distance (from the
M. A. Ellenbogen et al., 2002. A stressful challenge, screen) parameter.
affects negatively.
Lowering of mood and increase of anxiety in R + EMMA, Salvucci, 2001)
response to the negative stressor.
Rule 4. Decreasing the perceptual width field
If stressor= negative Then lowering mood, parameter.
If stressor= positive Then no change or increase

mood.
COGNITIVE TUNNELING:
If stressor= neutral, Then mostly change to negative
mood. Cognitive tunneling- (Wickens et al, 1998.)
If acute stress & negative-emotional visual stimuli, Rule 1: Declarative and procedural retrieval
Then Increase level of: anxiety, depression, hostility thresholds in ACT-R can be modified to represent a
and dysphoria. greater reliance on well-known and well-practiced
knowledge.
(Morelli and Burton, 2009).
declarative retrieval threshold0.
AA: Selective attention

attributes (AA)
U.S. Army Research Laboratory (Morelli and Burton, procedural retrieval threshold0.
2009). Selective attention performance on a
multiple object tracking task after stress-inducing Rule 2: Noise in the procedural rule application
exposure is decreased. process can be decreased (i.e., activation noise and
expected gain noise), which would lead to only the
AA: Divided attention most well-practiced materials being retrieved and
applied.
Dual tasks= (simultaneous auditory and visual
stimuli) continuous performance tests. activation noise: ans> 0
(Petrac et al., 2009). Increased environmental Expected gain noise EGN is modified.
stress relates to decreased divided attention
performance in auditory vigilance when Silverman Model (Silverman, 2010),
performing dual tasks.
Bohnen et al. (1990),
t
Subjects who exhibited high cortisol levels (acute /

stress) during a session of continuous mental tasks
had poorer divided attention following task 0 <= ES(t) <= 1.0, ES can vary over the interval 0,1
performance. with 0 being over-confident and unstressed, 0.5 is
neutral, and 1.0 is totally stressed.
AA: Sustained attention (vigilance)
Vigilance decrement becomes significant within

the first 15 minutes of attention, (Teichner,(1974)) TP = TI /( TI + TA)

attributes (AA)
If task demand conditions are high, decline in 0 <= TP <= 1.0

detection performance can occur more quickly
(time < 15 minutes)( Helton et al. (2007) EF(t) = EF(t -1) + { i=1,I wi[PMFi(t)/(PMFi(t) + FTi)]/I} -
{EF(t-1)/10 if
Vigilance tasks (continuous attention)= stressors;
0=PMFi(t)}
(Gluckman et al.,1988)
0 <= EF(t) <= 1.0
If acute stress= noise (acoustic stressor above 90
db SPL)), then sustained attention performance
decreases. (Kryter,1970)
Sustained attention is degraded as thermal

homeostasis of the observer is disturbed.
Loeb and Jeantheau, 1958).
Performance on low demand tasks is facilitated in

the presence of low-level varied noise.
AA: Sustained attention, emotions, & fatigue
If stressor= vigilance task
In a study with subjects (Trackray, Bailey and

Touchstone, 1977), these where asked to rate
themselves on five mood dimensions before and after

attributes (AA)
a vigilance task. They reported they were more

strained and less attentive after the task. Subjects
also reported higher levels of fatigue and
drowsiness after the task than before.
Elderly:
Lower detection rates and higher false alarm rates

with age when comparing groups between 19 and 27,
40 and 55, and 70 and 80 years old (Parasuraman, R.
& Giambra, L. (1991).
Decreased performance for elderly (decreases

beyond age 40). (Deaton, J. & Parasuraman, R.
(1993)).
AA: Perception
Note perceptual narrowing (or tunneling) as a major

effect of stress. (Wickens & Hollands, 2000)
If stress, then effective visual perceptual_field

attributes (AA)
becomes smaller.
AA: Decision making
Decision-making process follows an inverted-U

arousal-performance model of stress. (Kowalski-
Trakofler* and Charles Vaught)
Problem solving and decision making in demanding

real-world situations can be susceptible to acute
stress effects(Massa et al, 2002, pg 1).
If acute stress, then simple strategies are favoured

over more complicated ones: reduce and select the
information being attended to and processed,
focusing on critical issues. (Limited number of options
considered)
Stress induces a tendency to offer solutions before

all decision alternatives had been considered and
to scan such alternatives in a non-systematic fashion.
(Keinan et al.) and can (not necessarily) lead to
errors, poor performance and bad decisions.
Unfamiliar circumstances are relevant regarding

deficient decision making.

attributes (AA)
Negative acute stress is likely to have a particularly

detrimental effect on decision in the use of analytical
and creative strategies.
If eustress (positive stressors)
Positive effects of stressors are increased motivation

and energy, faster reactions, clearer thinking and
improved memory retrieval.
The critical appraisal of demands and coping

resources is based on a host of factors: experience,
training and personality. Familiar circumstances are
relevant regarding effective decision making.
AA: Speech and language
The average intensity of speech is observed which

increases in noise (Lombard reflex), with anger or
some types of high workload. (stressful situations).
mainly vowels and semivowels show a significant
increase in intensity while consonants not.
Mean word duration is a significant indicator of

speech in slow, clear, angry, Lombard (noise) and
loud conditions.

attributes (AA)
(Steeneken and Hansen, 2002).
Pitch is the most widely considered parameter of

stress evaluation. Ex:
Pitch (Neutral)= 14515 Hz;
Pitch (Lombard)= 16024 Hz
Pitch (Angry)= 25395 Hz
Different stressors types affects differently on the

different speech production subsystems:
If physical stressor: affects on articulator movements.
If physiological stressor, affects on muscular actions
If perceptual stressor, affects on generation of

articulatory targets
If psychological stressor, affects on linguistic

programming and ideation.
Acute AA: Selective attention Previous models ( for different types of stress Stanford Acute Stress Reaction
stress: arousal) can be used here as well. Questionnaire (SASRQ) (Cardena &
positive, Rapid attentional disengagement from negative words Spiegel, 1993; Shalev, Peri, Canetti, &

attributes (AA)
neutral & occur in response to stress-induced negative affect, STRESS MODEL by Janis and Mann Schreiber, 1996).Ranges of Values per
negative as an adaptive means to regulate emotional arousal. different Studies.
M. A. [Ellenbogen et al., 2002] (Irving, L. J., Mann, L. 1977).
Integrated - Acute stress over 0.8
with If stressor is negative & (negative emotional stimuli) Rule: Stress vs Performance/Decision
Emotions then mood is lowered & shift attention_time Effectiveness - Threatening patient .87
increases (from negative to positive stimuli)
iSTRESS: - Vicinity of shooting .93 (.72.88 for
If low dysphoria (positive/neutral emotions) subscales)
1 unconflicted adherence, select mt = mt-1
{& positive stressors then Shift_attention_time is - Sexual abuse survivors .93 (.78
higher: 38010 ms} 2 unconflicted change to next Sm (next state) .86 for subscales)
{& neutral condition then Shift_attention_time 3 m = 1, M whichever is Best Reply (vigilance) - Emergency rescue .8 (.64.98 for
decreases: 37010 ms;} subscales)
4 near panic so, m = 1 of 1, M (if highly
{& negative stressor then Shift_attention_time experienced users: Recognition Primed Decision - Catastrophic flooding .95 (.82.88
decreases: 35010 ms} making. Also, defensive avoidance occurs at this for subscales
level for non experts).
(Low dysphoria more efficient shifting attention with
negative stressors than positive stressors) > 4 panic, so Sm = avoid the stressor (hyper-
vigilance) Dundee Stress State Questionnaire
If high dysphoria (negative emotions) (DSSQ; Matthews et al., 1999;
Metrics Matthews et al., 2002; Helton et al,
{& positive & neutral stressors then 2009).
Shift_attention_time is lower: 36510 ms} (Stressor level, pressure time, risks)

attributes (AA)
{& negative stressor then Shift_attention_time Output: Performance/Decision Effectiveness

increases: 39210 ms}
No quantitative metrics.
(High dysphoria is more efficient shifting attention with
positive stressors than negative stressors)
If positive& neutral stressor, low dysphoria person

is less efficient (38010 ms) in shifting attention than
a high dysphoria person (37010 ms).
If negative stressors, low dysphoria person is

substantially more efficient (35010 ms) in shifting
attention than a high dysphoria person (39210 ms).
Elderly:
Elderly with chronic stress and high levels of cortisol

ACT-R OVERLAYS
(stress) at the time of assessment, have significantly
poorer performance on tasks of selective
attention than participants whose levels of cortisol (e.g., Gluck, Gunzelmann, Gratch, Hudlicka, & Ritter,
2006).
declined over the four years (lower chronic stress).
(Lupien et al., 1994). Seeman et al. (1997) reported

similar results from the subsample of women in their
WORKING MEMORY
cohort of elderly individuals (aged 7079 years).
Rule 1:Modify the decay rate of working memory

attributes (AA)
objects.
Rule 2: Increase the declarative memory retrieval

threshold parameter.
Rule 3: Decrease the base-level activation of all

memory elements when stressed
ATTENTION:
Rule 1: Wickens- WM overlay: Working memory

capacity decreases.

decreases attention to the task is to create a
secondary task. This secondary task simulates worry.
The rules that create this secondary task might be
seen as architectural productions. This overlay
decreases performance rate on the serial subtraction
task.
PERCEPTION:
Wickens-Perceptual Tunneling options:

attributes (AA)
Rule 1. Modify the visual attention latency

parameter.
Visual-attention-latency increases > 0.085.

screen) parameter.

R + EMMA, Salvucci, 2001)

parameter.
COGNITIVE TUNNELING:
Cognitive tunneling- (Wickens et al, 1998.)
Rule 1: Declarative and procedural retrieval

thresholds in ACT-R can be modified to represent a
greater reliance on well-known and well-practiced
knowledge.
declarative retrieval threshold0.

attributes (AA)
procedural retrieval threshold0.
Rule 2: Noise in the procedural rule application

process can be decreased (i.e., activation noise and
expected gain noise), which would lead to only the
most well-practiced materials being retrieved and
applied.
activation noise: ans> 0
Expected gain noise EGN is modified.
Silverman Model (Silverman, 2010),

t
0 <= ES(t) <= 1.0, ES can vary over the interval 0,1
with 0 being over-confident and unstressed, 0.5 is
neutral, and 1.0 is totally stressed.
TP = TI /( TI + TA)
0 <= TP <= 1.0

attributes (AA)
EF(t) = EF(t -1) + { i=1,I wi[PMFi(t)/(PMFi(t) + FTi)]/I} -

{EF(t-1)/10 if
0=PMFi(t)}
0 <= EF(t) <= 1.0
Chronic AA: Long Term Memory ACT-R OVERLAYS Elderly

(Sustained)
stress Chronic over-secretion of cortisol prevents the brain (e.g., Gluck, Gunzelmann, Gratch, Hudlicka, & Ritter, Elders Life Stress Inventory (ELSI): The
from creating new memories or accessing already 2006). ELSI is a self-report life event scale,
existing memories. (animal research by Diamond ; consisting of 30 items, each rated on a
R.M. Sapolsky ). WORKING MEMORY six-point scale (05), particularly
developed for the elderly. Two types of
Time to learn a new pattern increases. Rule 1:Modify the decay rate of working memory scores are derived from the scale, one
objects. reflecting number of life events
It prevents the brain from laying down a new memory, experienced during the last year and
or from accessing already existing memories. Rule 2: Increase the declarative memory retrieval the other measuring the subjective
threshold parameter. impact of those life events. The scale
AA: Working Memory has satisfactory psychometric
Rule 3: Decrease the base-level activation of all properties (Aldwin, 1991).
(Kazushige Mizoguchi et al, 2000) memory elements when stressed.
Induces impairment of spatial working memory. ATTENTION:

attributes (AA)
Chronic stress impairs the maintenance of a novel Rule 1: Wickens- WM overlay: Working memory
short-term memory, impairs the transfer of newly capacity decreases.
acquired information from WM to LTM, thereby
affecting learning. Rule 2: Another way to implement an overlay that
decreases attention to the task is to create a
Stress affects working memory with respect to skill in secondary task. This secondary task simulates worry.
arithmetic problem-solving, reading and reading The rules that create this secondary task might be
comprehension, and the 11 retrieval of information seen as architectural productions. This overlay
from long term memory (Ashcraft and Kirk, 2001). decreases performance rate on the serial subtraction
task.
PERCEPTION:
Elderly:
Wickens-Perceptual Tunneling options:
1/3rd of elderly are chronically affected by stress.
deficits in hippocampus-dependent memory tasks. Rule 1. Modify the visual attention latency
parameter.
Higher cortisol levels are significantly associated with
poorer working memory (letter-number sequencing) Visual-attention-latency increases > 0.085.
and processing speed (digit symbol test).
Seeman et al. (1997) reported similar results from the screen) parameter.
subsample of women in their cohort of elderly
individuals (aged 7079 years): exhibit declines in Rule 3. Increase the saccade time parameter. (ACT-
memory performance. R + EMMA, Salvucci, 2001)

attributes (AA)
parameter.
AA: Episodic Memory COGNITIVE TUNNELING:
Elderly: Chronic stress may exacerbate age-related Cognitive tunneling- (Wickens et al, 1998.)
decline on episodic memory tests that require greater
executive resources and more integrated and Rule 1: Declarative and procedural retrieval
elaborative processing. (VonDras et al. 2005) thresholds in ACT-R can be modified to represent a
greater reliance on well-known and well-practiced
Young and mid-life adults who report high levels of knowledge.
everyday stress (PSS and ELSI) approximate the
Logical Memory Scale performances of individuals 40 declarative retrieval threshold0.
or more years older who report low levels of stress.
procedural retrieval threshold0.
Seeman et al. (1997) reported similar results from the
subsample of women in their cohort of elderly Rule 2: Noise in the procedural rule application
individuals (aged 7079 years): exhibit declines in process can be decreased (i.e., activation noise and
memory performance. expected gain noise), which would lead to only the
most well-practiced materials being retrieved and
AA: Selective attention applied.
Liston et al colleagues (2009). Chronic stress disrupts activation noise: ans> 0

the brains ability to shift attention.
Expected gain noise EGN is modified.
Shifting attention ability decreases during that
period. Task performance decreases.

attributes (AA)
AA: Divided attention
Decreased ability to shift focus (Bruce McEwen et al):

Time to shift attention increases. Decreased
performance in tasks: time to catch on to a new
pattern increases significantly.
AA: Perception and emotions
Danigno-Subiabre et al. (2005), chronic stress impairs

both visual and acoustic perception.
More deleterious effects in the subcortical auditory

system than in the visual system and may affect the
aversive system and fear-like behaviours.
If chronic stress, then

Condition_avoidance_responses (CAR %) decrease
when using auditory or visual stimuli.
Stress produces a more dramatic decrease in

acoustic conditioning than in visual conditioning.
Auditory stimuli (acoustic negative stressor) can

trigger aversive behaviours (fear, anxiety), perhaps
more likely than visual stimuli (Azrin, 1958; Reed et

attributes (AA)
al., 1996; Macedo et al., 2005).
Table 22: Metrics, computational models and measurement techniques for stress: Abstratct User Model Cont.

P&B Existent Metrics from Psychological Computational Models: Rules and parameters Questionnaires and other
STATE analysis methodology to measure them
(Qualitative or Quantitative) (Specify for different domains and users

if available)
Affected Values, rules Rules Metrics
attributes
(AA) (specify if values change (specify if rules change for a (Found params, wanted
for a specific user), rules) specific user) Bibliography. params, ACT-R params)
Fatigue AA= Sustained attention ACT-R OVERLAYS Fatigue state can be induced by sleep
deprivation (TSD). The effect on
Deterioration of sustained attention SUSTAINED ATTENTION reaction time and errors freq. is
level measured by Psychomotor Vigilance
Rule 1: Test (PVT)
AA= Declarative memory
E = PG C + [Gunzelman et al, 2005 , 2009] Automotive:
Reduce the activation of declarative
sleep deprivation may lead to poorer performance on a task eyelid movement (percentage of eyelid
that requires sustained attention and rapid responses to closure over the pupil, PERCLOS, and
Knowledge, leading to longer retrieval
frequent signals. (G is set at 1.87 as baseline condition; G average eye closure speed, AECS
times and
is set to 1.77, 1.72, and 1.68 to represent the effects of 1, 2, tracking of gaze , speech
occasional retrieval failures and 3 days of TSD respectively). G can be also modeled
through a biomathematical model (Sleep Deprivation.. Walter Reed Serial Addition/Subtraction
Gunzelm 2009, pag 896-897 Task (SAST). This task refers omission
errors and slower down in fact retrieval
Performance rate decreases
Salvucci at al. Myer et al. show the time for a production-

rule in ACT R firing increases from 50 ms for a younger
person to 56.5 ms for an older person a 13% increase.
Rule 2 :
Ai =bi+ somma j WjSji Dip + s
Table 23: Metrics, computational models and measurement techniques for fatigue: Abstract User Model Cont.

P&B STATE Existent Metrics from Psychological Computational Models: Rules and parameters Questionnaires and other methodology
analysis to measure them
(Qualitative or Quantitative) (Specify for different domains and users if

available)
attribute
s (AA) (specify if values change for (specify if rules change for a (Found params, wanted
a specific user), rules) specific user) Bibliography. params, ACT-R params)
Emotions AA: Memory ACT-R OVERLAYS MODELS Emotional Scales
Cochrane, B., Lee, F.J., & Chown, E. (e.g., Gluck, Gunzelmann, Gratch, Hudlicka, & Ritter, 2006). ESRE (Emotion During Service
(2006). Modeling Emotion: Arousals Recovery Encounters) scale
impact on memory.
Profile of Mood States (POMS) model of
High arousal during encoding can mood
facilitate long term retention, but it is also
associated with an inability to retrieve POMS metrics: 0 (not at all) to 4
information for a short period of time (up (extremely).
to about 30 minutes) following the original DECISION MAKING
encoding. Feeling of Emotions Scales
(Belavkin et al. 1999)
AA: Declarative Memory The Self-Assessment Manikin (SAM)
Rule1: When more information becomes known about the task,
subjects become more confident, which corresponds to higher G/
Fum, D. & Stocco, A. (2004). Memory,
ratio and lower noise.
emotion, & rationality: An ACT-R
interpretation for gambling task results. In
MOTIVATION
Proceedings of the sixth International
Conference on Cognitive Modeling (pp.
Positive emotions, experienced on successes during problem
106-111). Pittsburgh, PA: Carnegie
solving, are accompanied by increase of the motivation (goal value
Mellon University/University of Pittsburgh.
G) and confidence (G/ ratio).
The memory affects on emotions.
Negative emotions correspond to a decrease in G and G/ (heating
the system up). Negative emotions occurring during problem solving
AA: Problem Solving
can play a positive role in overcoming possible problems.


available)
attribute
(Belavkin et al. 1999) Inverted U Curve Model Pleasure-Arousal-Dominance (PAD) :
Emotion makes a positive contribution to Memory Performance and Arousal PAD scale range from -1 to +1:
problem solving, since it implements
powerful heuristic methods already known Cochrane, B., Lee, F.J., & Chown, E. (2006). Modeling Emotion: angry (-.51, .59, .25),
in AI and mathematics, and hence it is Arousals impact on memory When arousal is high, tasks that
important for intelligence. require a relatively high cognitive workload (complexity) or are less bored (-.65, -.62, -.33),
well-learned are less likely to be pursued. However, high levels of
arousal can increase the likelihood that physically demanding tasks curious (.22, .62, -.01),
(e.g. running) will be pursued.
dignified (.55, .22, .61),
elated (.50, .42, .23),
hungry (-.44, .14, -.21),
inhibited (-.54, -.04, -.41),
loved (.87, .54, -.18),

PAD-Space MODEL:
puzzled (-.41, .48, -.33),

Thierry Bock, Artificial Emotions 2 : An Implementation of OCC,
Including Middle-term Dynamics
sleepy (.20, -.70, -.44),
Personality Traits
unconcerned (-.13, -.41, .08),
Pleasure := 0.21*Extraversion + 0.59*Agreeableness +


available)
attribute
0.19*Neuroticism violent (-.50, .62, .38).
Arousal := 0.15*Openness + 0.30*Agreeableness - 0.57*Neuroticism Physiology measures
Dominance := 0.25*Openness + 0.17*Conscientiousness + Emotion recognition from speech:

0.60*Extraversion - 0.32*Agreeableness
Voice changes.
FUZZY RULES MODELS
Changes in metabolism:
L. Pour Mohammad Bagher, Intelligent Agent System Simulation
Using Fear Emotion, Conference on Machine Intelligence (ICMI Positron emission topography (PET)
2008), 2008
Functional Magnetic Resonance Imaging
(fMRI).
Parameter: Fear Intensity Magnetic fluctuation:

Emotion calculation component uses the degrees of fear parameters
which are set by user to calculate fear intensity of agent by using the Electric fluctuation:
implemented fuzzy rules as the result.
Facial Action Coding System (FACS) MODEL
Eye-position and eye-movement
Movement of facial muscles indicating emotional reaction. indicating attention:
The FACS manual provides the Action Units, as well as Dr. Ekman's Eye-tracking


available)
attribute
interpretation of their meaning. Cardiovascular Responses.
Facial Action Coding System (FACS) by Paul Ekman, Wallace V. Phasic changes (short term changes) in
Friesen, and Joseph C. Hager, 2002 heart rate indicate attention and tonic
changes (short term changes) indicate
AdSaM MODEL arousal
The nine points of AdSaM are: Enthusiastic, Warmed, Comfortable,

Ambivalent, Indifferent, Apprehensive, Sullen, Troubled and
Alarmed.


available)
attribute
PrEmo Model
Conductance of the skin indicating arousal Model:
Electrodermal reaction (EDR)
Typical Values of electrodermal reactions (EDR) components:
- Skin Conductance level (SCL)= 2-20 S
- Change in SCL= 1-3 S
- Frequency of NS-SCRs= 1-3 per min


available)
attribute
- ER-SCR amplitude= 0.2-1.0 S
- ER-SCR latency =1-3 S
- ER-SCR rise time=1-3 S
- ER-SCR half recovery time= 2-10 S
- ER-SCR habituation =0.01-0.05 S per trial
Cardiovascular Responses Model
Phasic changes (short term changes) in heart rate indicate attention

and tonic changes (short term changes) indicate arousal.
Movement-to-movement ratings system
Shifting hand to the left indicates boredom or dissatisfaction, and

shifting hand to the right indicates a positive emotional connection:


available)
attribute
ALMA Fum & Stocco (2004) MODEL
A Layered Model of Affect system used to model emotions, moods

and personality according to events and time passing.


available)
attribute
Table 24: Metrics, computational models and measurement techniques for emotions: Abstratct User Model Cont.

P&B Existent Metrics from Computational Models: Rules and parameters Questionnaires and other
STATE Psychological analysis methodology to measure
them
(Specify for different
Affected Values, rules Rules Metrics domains and users if
attributes available)
(AA) (specify if values (specify if rules change for a (Found params, wanted
change for a specific user) Bibliography. params, ACT-R params)
specific user),
rules)
Motivation AA=Selective ACT-R OVERLAYS MOTIVATORS

attention
ACT-R does not have an explicit Motivational subsystem, Bernard et al. (2008) report
A motivational state and however motivation can be implemented using the Goal the development of the
individual differences on trait subsystem (example: Belavkin, R. V. (2001). The role of Assessment of Individual
dimensions of neuroticism and emotion in problem solving. In C.Johnson (Ed.), Proceedings MotivesQuestionnaire
extroversion can influence of the AISB'01 Symposium on Emotion, Cognition and (AIM-Q),
attention and information Affective Computing: In ACT every rule, in addition to its
processing. symbolic representation, has also so called expected gain E INFOTAINMENT
and a rule with the highest gain wins the competition in a
selective attention is Academic Motivation
conflict set. Expected gain E is calculated by the following
sensitive to affective- Scale (EME) (Robert J.
equation:
motivational states. Vallerand et al, 1992b): A
E = PG C + () questionnaire for
AA=Memory measuring the intrinsic,
where P is expected probability of achieving the goal if the extrinsic, and amotivation in
Motivation only affects rule fires, G is the value of the current goal in time units, C is education.
memory when it is presented expected cost of that rule in time units (it represents how
at the time of encoding long will it take to achieve the goal if that rule fires), and () WORKPLACES
(working memory) and has is a random variable representing noisy or non-deterministic

them
specific user),
rules)
no significant effect when part of ACT-R conflict resolution mechanism; Work Motivation Scale
presented at recall, or during (Brady & Reinink, 2001).
the process of consolidation. MOTIVATION ATKINSON MODEL FOR
REHABILITATION ELDERLY
The motivation model by Atkinson is the following: Elderly motivation scale

(EMS-72) (Robert J.
Motivation = F (Expectancy x Motive x Incentive). Vallerand et al, 1992):
This model is applicable only when an individual knows that Physiological Response
his effort will be evaluated (by himself or others) in terms of
some standard of performance, and that the consequence of Brehm & Self, 1989)
this action will be either a success or a failure. measures:
Birren and Schaie (1985) posit that Atkinsons model is heart rate (HR),
suitable in the rehabilitation setting. They reformulate the
model as follows: systolic blood
pressure (SBP),
M = (P(Os) x V) : C,
diastolic blood

them
specific user),
rules)
- where M is motivation for rehabilitation, pressure (DBP).
- P(Os) is the subjective perception of the patient of the Kreibig (2010b) gives a
likelyhood of success, assuming that he or she brief outline of the physical
makes the required effort, components of
cardiovascular response
- V is the individually perceived utility or value of a (CVR) in emotion. She then
successful rehabilitation, reviews theoretical
positions on the
- and C the perception of the cost of the outcome. organization of CVR in
emotion and its
motivational function.
Finally, she considers a
systematic integration of
emotion, motivation, and
CVR.

them
specific user),
rules)
CLARION ARCHITECTURE
The CLARION cognitive architecture does have a

motivational subsystem (R. Sun, Motivational representations
within a computational cognitive architecture . Cognitive

them
specific user),
rules)
Computation, Vol.1, No.1, pp.91-103. 2009
Motivation model Galbraith, 1977):
The model postulates:
- All human effort is carried out with the expectation of

success.
- Subject trusts that if the expected performance is

achieved, certain consequences will follow for them.
Every consequence or result has a determinate value
for the subject, called valence.
- The relation between effort and performance depends

on two factors: the subjects abilities and their
perception of the position.
- Each person has a certain idea about the level of

performance that they are capable of achieving in the

them
specific user),
rules)
task.
- People believe that whoever carries out the jobs best will
receive the best rewards.
- Strength of motivation = Value of reward * Probability

of achieving
Motivation model Geen (1994)
Motivation refers to the initiation, direction, intensity and

persistence of human behaviour:
- Intensity of motivation= momentary magnitude of

motivational arousal (Brehm & Self, 1989).
- Magnitude of motivational arousal= total amount of

effort a person would make to satisfy a motive;
- Persistence = effort extended over time;

them
specific user),
rules)
Factors determining magnitude of arousal (potential

motivation)
- Needs (Vary in value)
- Potential outcomes (Vary in value)
- Perceived probability of success.
relationship between task difficulty and motivational arousal

have focused on cardiovascular reactivity (Brehm & Self,
1989) as measured by changes in:
heart rate (HR),
systolic blood pressure (SBP),
diastolic blood pressure (DBP).
Kuhls model (1986)

them
specific user),
rules)
interaction among these three psychological processes:

cognitive, emotional and motivational.
Table 25: Metrics, computational models and measurement techniques for motivation: Abstract User Model Cont.

5.2 Towards an ontologies description

5.2.1 Introduction and methodology
In this section we describe a first approach towards describing the Behavioural and
Psychological Abstract User models in an ontological way. An ontology is a formal
representation of knowledge as a set of concepts within a domain, and the
relationships between those concepts (Wikipedia, Ontology). In order to define the
ontology the designer must be familiar with the target domain and fully understand
what aspects of the domain are to be modelled.
The ontology must cover all the dimensions of the psychological states that VERITAS
is focused in: stress, fatigue, emotions and motivation. To accomplish this task, the
following approach was followed:
Figure 64: Ontology methodology
First, an in depth study and research on every psychological state was done. This
allows a deep understating of the domain and more importantly to determine the
different types, pathologies, causes and answers that are related to the corresponding
psychological state.
Second, as a result of the previous step, the following outcome was produced:
A formal definition of every psychological state as well as a taxonomy of their

attributes, causes and responses.
A table with the responses that every psychological state produces for every
type of stimuli. This includes the metrics and rules (whenever possible) to
quantify them (Table 21).
This generated a descriptive model of the psychological states that is of the interest of
VERITAS project and is ready to be modelled as a formal ontology.
Third, taking the output of the previous step and using Protg [ref] an ontology is
being defined. The model will be then reviewed and tested to meet the requirements of
the project by creating training tests that describe real life examples of VERITAS
domains.

In the following section, a first approach towards a formal description of the P&B
Abstract User Models is presented. The final ontologies will be developed during the
following year.
5.2.2 Formal and explicit description approach:

The starting point is the class PsychologicalState (Figure 64). This class represents a
psychological state: its definition, cause, classification and what effects it has on the
user. It contains the following properties:
hasCause: points to an instance of the Cause class; it describes the cause that
triggers the current psychological state.
hasICD: points to an instance of the ICD class, which contains a description

based on the International Classification of Diseases.
hasICF: points to an instance of the ICF class, which contains a description

based on the International Classification of Functioning.
affectsCognitiveFunction: points to an instance of the CongnitiveFunction class,

which describes what cognitive function and in which measure is affected by
this psychological state.
hasShortDescription: a String describing the psychological state in a few words.
Figure 65: PsychologycalState attributes
The Cause class defines the type of cause that is causing the current psychological
state (Figure 65). A hierarchy of classes is proposed. It has two children: Stressor and
OtherCause Here we define different levels of hierarchy to define the stressor type in

detail. The class OtherCause is left as an example of other possible causes to be

defined afterwards.
Figure 66: Cause hierarchy
In order to better classify the nature of the current psychological state, the attribute
hasICD was defined. This attribute points to an instance of the class ICD, which
defines the ICD classification hierarchy. The following figure (Figure 66) shows a
portion of it:
Figure 67: ICD hierarchy
In the same manner, the attribute hasICF links to an instance of the ICF classification
to define what functionality is being affected by the current psychological state. In the
following figure (Figure 67) we can see an example of the ICF hierarchy classes:
Figure 68: ICF hierarchy
Since both the ICF and ICD classification are being used in other ontologies (physical,
behavioural) externalizing these definitions in a common ontology is here proposed.

Then, the psychological, physical and behavioural ontologies would use it by importing
it, ensuring that each ontology uses the same classification.
A certain psychological state affects one or more cognitive functions. This is defined in
the CognitiveFunction class, which may be defined as a High level cognitive function
or a Basic cognitive function (Figure 68). Each one contains a hierarchy to enable a
more detailed definition.
The CognitiveFunction class has an instance of the class Parameter, used to define the
source of the cognitive function value that is being set.
Figure 69: CognitiveFunction hierarchy
The Parameter class defines the value that produces the current psychological state
and the source (Figure 69). The Parameter class is modelled as a hierarchy of classes.
Depending on the source of the CognitiveFunction it will be an instance of one of the
following classes:
ModelParameter class: the source is well known computational model such as

ACT-R, EPIC or SOAR.
QuestionnaireParameter: the source is a questionnaire
MeasurementParameter class: any other kind of source (numeric measurement

or qualitative value)
Every Parameter class has the following attributes:
isQuantitative (Boolean): weather the information is quantifiable.
hasValue (String): the parameter value.

hasFunction (String): an expression that defines the values behaviour in

complex situations (equations).
hasMaxValue (String): The maximum possible value.
hasMeasureUnit (String): The unit of measure (seconds, percentage, etc)
hasMinValue (String): The minimum possible value.
Figure 70: Types of Parameter
The memory has been defined in the Memory class (Figure 70). The memory class is
defined as a hierarchy of classes:
Long term memory. Which is divided in three categories: semantic, episodic

and procedural.
Short Term memory: Corresponds to the working memory.
Every memory instance has the following attributes:
hasOperation: points to an instance of the Operation class, which has been

limited to either Consolidation or Retrieval.
hasMemoryType: points to an instance of the MemoryType class, which has

been limited to either Spatial or Verbal.

Figure 71: Types of Memory

6 Conclusions and next steps

The aim of this document is to describe the Behavioural and Psychological Abstract
User Model of VERITAS users, elderly and disabled people.
The first problem we have faced comes with the different definitions of what
behavioural and psychological models are: on one hand, a behavioural model is an
executable model defining how the character should react to stimuli from its
environment, whereas a psychological model is considered to be a synonym of a
cognitive model, which is oriented to describe how the human mind works. Taking into
account that WP1.4 is in charge of creating the Cognitive Abstract User Model for
elderly and disabled, in WP1.5 we have finally focused on modelling how different
psychological facets affect these users, analysing the physiological, cognitive and
behavioural responses, considering the P&B facets as moderators of human body
functions.
A first analysis of the computational models for modelling of more than 100 literature
sources has been performed. The resulting state of art of P&B computational models
such as ACT-R and EPIC, shows that cognitive architectures are extensively used in
psychology studies to model human behaviour. These architectures per se model
human cognition, allowing model, parameters and architectural variation for
generalization to new situations, users and domains. In cognitive architectures,
psychological facets should be therefore modelled as cognitive and behavioural
moderators or overlays of the architecture. Overlays are based on psychological
theories and studies about cognitive and behavioural moderators that can be applied to
cognitive architectures and simulate the effect of the psychological facets on embodied
cognition.
This approach is consistent with the theories of how the human mind works: a stimulus
(either internal or external, is perceived by the sensory system and processed by the
different cognitive functions in order to produce a motor response. If this stimulus is
perceived by the individual as emotional, threatening or stressing, the human body
triggers out a set of responses that moderate human cognition and behaviour when
performing any task.
Despite the fact that the analysis shows that ACT-R with overlays is one of the best
approaches for modeling the influence of psychological states on human cognition and
behavior, it may be unsuitable for VERITAS simulation purposes, which is more
focused on the physical and functional limitations of the users. However, and taking
into account the simulation approach followed in VERITAS, based on constraints, ACT-
R is useful for obtaining information relative to cognitive and behavioural parameters
that can be used in a constraints approach. Further analysis towards integrating the
models in a simulation scenario should be performed in order to ensure the suitability
of these models for VERITAS.
Elderly and disabled people face multiple life and body changes that lead them to
experiment distress, negative emotions, fatigue and lack of motivation. These
psychological states have a great impact on their quality of life and affect the overall

performance of these groups when performing a task. This is what in VERITAS we

have focused on modelling these psychological facets as moderators of human
performance.
A deep analysis of each psychological facet has been performed, focusing on the
physiological, cognitive and behavioural human responses. More than 200
psychological studies have been thoroughly analysed, with the result of a detailed
description of each facet, which shows the different dimensions of the psychological
states:
Description of each facet;
Categorisation of the facets, using when possible ICD classification.
Stimulus or causes that originate the state on individuals;
Theoretical models of the facets;
Physiological, cognitive and behavioural responses, with the corresponding ICF

classification.
Measurement methodology and metrics of these responses;
Study of the effect of the P&B state on each specific cognitive function relevant
for VERITAS users as described in WP1.4 (memory, attention, reaction time,
decision making), extracting parameters, and qualitative and quantitative
values.
Analysis of the studies for elderly and disabled
Study of the existent computational models and approaches that can be used,
focusing on VERITAS users and domains.
The different dimensions of the P&B states constitute the Psychological and
Behavioural Abstract User Model useful for VERITAS virtual user models description.
The resulting taxonomy, which is a first approach towards ontologies, proposes a

reference categorisation of the mentioned states that can be extended and therefore
used that in domains, research outside VERITAS.
Taking into account the elevate number of papers analysed, the diffused state of the art
regarding psychological states and the poor knowledge and studies carried out with
elderly and disabled up to date concerning these topics, the present study represents a
reference analysis of those psychological and behavioural states that have a great
influence amongst the elderly and disabled. It goes one step further providing an
extensive overview of the latest psychology studies concerning stress, fatigue,
emotions and motivation and processing all the information by cross-referencing
different topics.

Moreover, the structure and the methodology used for gathering and processing all the
information allow verification with users, future refinements and development of new
models for other psychological facets.
The integration of emotions, stress, fatigue and motivation on cognitive models will
allow creating more human-like models. From an application and product development
perspective, this knowledge is essential to ensure the acceptability and accessibility of
future products and services. These models are defined based on real users, on how
they feel and behave when confronted to certain positive and negative events. Adding
these features in virtual models for simulations use opens a large number of
possibilities for improving the cycle of products development.
Achievements:
Study of more than 100 generic behavioural and cognitive computational models
for VERITAS users and domains: The state of art shows that (1) a cognitive
architecture such as ACT-R is the most promising approach for modelling
psychological theories of human mind and (2) for modelling psychological and
behavioural facets the most suitable approach are overlays/ cognitive and
behavioural moderators.
Analysis of ACT-R and overlays: more than 30 literature sources have been
reviewed. This analysis shows the pros and cons of adopting this approach in
VERITAS. A further analysis considering the needs of VERITAS simulation
approach should be performed.
More than 200 psychological studies and experiences about stress, fatigue,
emotions and motivation have been examined, with special attention on the effects
of the P&B states on VERITAS users:
o Regarding stress, over 50 studies concerning physiological and cognitive

responses, around 100 references of stress as cognitive moderator and
over 30 references of measurement techniques have been collected.
o Concerning fatigue, over 40 papers dealing with physiological and cognitive

responses and 11 methods for measuring fatigue have been analysed and
included in the tables.
o In order to include the emotional states in the AUM, we have taken under
consideration more than 20 scientific papers of theoretical description and
responses of emotions. Following the general introduction about emotions,
a thorough presentation of the existing methods for measuring emotions has
been realised. Around 23 papers leaded to the presentation of 16 different
methods for measuring emotional states.
o Regarding motivation, over 50 references of motivational theory have been

reviewed. The analysis of motivation has not been so straightforward, there

is a lack on operational models for motivation and moreover to find how

different cognitive disabilities impact specifically on what we call
"motivation". Despite this, 4 relevant scales for motivation have been
included, as well relevant studies about motivation affecting on memory and
attention have been included in the AUM.
Analysis of more than 150 theoretical and computational P&B VERITAS users and
domains.
Also for each specific psychological state, relevant parameters, rules and metrics
have been extracted. Regarding stress and fatigue, around 11 and 20
computational models have been selected and included in the AUM respectively.
Concerning emotion, an analysis of 19 computational models has been presented,
including the relevant models in the tables. Finally, around 15 models of motivation
have been analysed.
Creation of an Abstract User Model that represents the multiple dimensions of the
four psychological states: stress, fatigue, emotions and motivation.
Development of a formal description of the tables that serves as static first

approach for developing the Abstract User Model ontologies.
Despite the fact that the domain of psychological states is quite a new income in
the field of simulations and human behaviour, we have achieved a thorough and
holistic presentation. This document includes all the information considered to be
related to VERITAS concept and needs. This fact might have lead in having too
much information gathered. However, this information is to be clarified in the
second year of the project, when the appropriate information that totally match the
project needs will be selected and implemented.
Next steps
Validation of extracted parameters, rules and metrics with real users testing,
using the Multisensorial Platform. Elderly and disabled users will be
participating in the tests.
Selection of the relevant parameters, values and rules for VERITAS purposes.
Development of the final modelling approach to be used in VERITAS: By

merging the Abstract User Models with the user interaction models (task
models), the final and parameterized Behavioural and Psychological User
model will be created.
Integration with the cognitive and physical models in order to create a complete
Virtual User Model.

7 References
A.Sloman, The Computer Revolution in Philosophy: Philosophy, science and models of
mind, Harvester press and Humanities press, 1978.
Aaker, David A.; Stayman, Douglas M. and Hagerty, Michael R. (1986), Warmth in
Advertising: Measurement, Impact and Sequence Effects, Journal of Consumer
Research, 12 (mar)
Ach, N. (1935). Analyse des Willens [Analysis of the will]. In: A.E. Abderhalden, (Ed.),
Handbuch der biologischen Arbeitsmethoden [Handbook of Biological Working
Methods], vol. 6. Urban & Schwarzenberg, Berlin, Germany.
Aldwin, C. (1994). Stress, Coping and Development. New York: The Guilford Press.
Anderson, J. & Bower, G. (1973). Human Associative Memory. Washington, DC:

Winston.
Anderson, J. R. (1983). A spreading activation theory of memory. Journal of Verbal

Learning and Verbal Behavior, 22, 261-295
Anderson, J. R. (2007) How Can the Human Mind Occur in the Physical Universe?
New York: Oxford University Press
Anderson, J. R. (2007). The image of complexity. In The 29th Annual Conference of

the Cognitive Science Society. Nashville, Tennessee, USA.
Anderson, J. R. (2007). Using brain imaging to guide the development of a cognitive

architecture. In W. D. Gray (Ed.), Integrated Models of Cognitive Systems (pp. 49-
62).New York, NY: Oxford University Press.
Anderson, J. R., & Lebiere, C. (1998). The atomic components of thought. Hillsdale,
NJ: Lawrence Erlbaum Associates.
Anderson, J., Bothell, D., Byrne, M., Douglass, S., Lebiere, C., & Qin, Y. (2004). An
integrated theory of the mind. Psychological Review, 111(4):10361060.
Andreassi, J. L. 2000. Psychophysiology: Human behavior and physiological response.

4th ed. Mahwah, NJ: Lawrence Erlbaum Associates.
Armitage, C. J., & Conner, M. (2000). Social cognition models and health behaviour: A
structured review. Psychology & Health, 15, 173-189.
Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour a
metaanalytic review. British Journal of Social Psychology, 40, 471-499.
Arnold, M. (1970). Feelings and emotions. New York: Academic Press.
Arnold, M. (1984). Memory and the brain. Hillsdale, NJ: Erlbaum.

Ashcraft, M. & Kirk, E. P. (2001). The relationships among working memory, math
anxiety, and performance. Journal of Experimental Psychology: General, 130, 224-237.
Atkinson, J.W., Bastian, J.R., Earl, R.W., Litwin, G.H. (1960). The achievement motive,
goal setting, and probability preferences. J. Abn. Soc. Psychol. 60: 27-36.
Azrin 1958 N.H. Azrin, Some effects of noise on human behavior, J Exp Anal Behav 1
(1958), pp. 183200.
Backs, R. W., & Boucsein. W. (Eds.). (2000). Engineering psychophysiology: Issues

and applications. Mahwah, NJ: Erlbaum.
Bagozzi, R. P.; Gopinath, M. and Nyer, P. U. (1999), The Role of Emotions in

Marketing, Journal of the Academy of Marketing Science, Vol. 27, no. 2
Bagozzi, Richard P. (1991), The Role of psychophysiology in consumer research, In:

Robertson, T. S. and H. H. Kassarjian (Eds), Handbook of Consumer Behavior, New
Jersey: Prentice Hall
Barry G. Silverman. "More Realistic Human Behavior Models for Agents in Virtual
Worlds: Emotion, Stress, and Value Ontologies" Center for Human Modeling and
Simulation (2001).
Basia Belza, PhD, RN, Department of Biobehavorial Nursing and Health Systems, Box
357266, University of Washington, Seattle, WA 98195-7266
Bechara, A., Damsio, A. R., Damsio, H., Anderson, S. W. (1994). "Insensitivity to

future consequences following damage to human prefrontal cortex". Cognition 50 (1-3):
715
Bechara, A., Tranel, D., & Damasio, H. (2000b). Characterisation of the decision-
making deficit of patients with ventromedial prefrontal cortex lesions. Brain, 123, 2189
2202.
Beckner, V. E., Tucker, D. M., Delville, Y., and Mohr, D. C. (2006). Stress facilitates
consolidation of verbal memory for a fi lm but does not affect retrieval. Behav.
Neurosci. 120, 518527
Belavkin, R. V., Ritter, F. E., & Elliman, D. G. (1999b). Towards including simple
emotions in a cognitive architecture in order to fit children's behaviour better. In
Proceedings of the 1999 Conference of the Cognitive Science Society.
Benson, B. A., & Fuchs, C. (1999). Anger arousing situations and coping responses of
aggressive adults with intellectual disability. Journal of Intellectual & Developmental
Disability, 24(3), 207214.
Berkun, M. (1962). Experimental studies of psychological stress in man. Psychological

Monographs, 76 (Whole No. 534).

Bernard, L., Mills, M., Swenson, L, Walsh, R (2008). Measuring motivation

multidimensionally: Development of the Assessment of Individual Motives
Questionnaire (AIM-Q). Assesssment, 15, 16 - 35.
Birren, J.E. & Schaie, K.W. (1985). Handbook of the Psychology of Aging. New York:
Van Nostrand Reinhold Company.
Biswas, P. & Robinson, P. (2008a). Automatic evaluation of assistive interfaces.

Proceedings of the 13th international conference on Intelligent user interfaces, pp. 247-
256.
Biswas, P. & Robinson, P. (2008b). A Motor-Behaviour Model for Physically

Challenged Users. Cambridge Workshop on Universal Access and Assistive
Technology, Cambridge, pp. 5-9.
Biswas, P. et. al. (2005). User Model To Design Adaptable Interfaces For Motor-
Impaired Users, In Proceedings of the Tencon 05 IEEE Region 10 Conferences,
Australia, 2005, pp. 1801-1806
Blandford, A., Butterworth, R., Curzon, P. (2004). Models of interactive systems: a case
study on programmable user modelling, International Journal of Human-Computer
Studies, vol. 60, pp. 149200.
Bodenburg, S. & Technow, U. (1992). Neuropsychologisce therapie am computer mit

alsten menschen. [Using computers in neuropsychological treatment with the elderly.]
Zeitschrift Fuer Gerontologie, 25(4), 255-258.
Bohnen, N., Houx, P., Nicolson, N. and Jolles, J., 1990. Cortisol reactivity and cognitive
performance in a continuous mental task paradigm. Biol. Psychol. 31, pp. 107116
Bolls, P. D.; Lang, A. and Potter, R. F. (2001), The Effect of Message Valence and
Listener Arousal on Attention, Memory and Facial Muscular Responses to Radio
Advertising, Communication Research, 28 (5)
Bonetti, D., & Johnston, M. (2008). Perceived control as a predictor of the recovery of
individual-specific walking behaviours following stroke: testing psychological models
and constructs. British Journal of Health Psychology, vol. 13, pp. 463-478.
Bosch, X. (2000). Stress hormone impairs long-term retrieval of memorised

information. The Lancet, 355, 1078.
Brady R. & Reinink, 2001. Work Motivation Scale. Ed.D.
Bramston, P., & Bostock, J. (1994). Measuring perceived stress in people with
intellectual disabilities: The development of a new scale. Australia and New Zealand
Journal of Developmental Disabilities, 19, 149-157.
Bramston, P., & Fogarty, G.J. (1995). Measuring stress in the mildly intellectually
handicapped: The factorial structure of the Subjective Stress Scale. Research in
Developmental Disabilities, 16, 117-131.

Brando et al 1994 M.L. Brando, S.H. Cardoso, L.L. Melo, V. Motta and N.C.
Coimbra, Neural substrate of defensive behavior in the midbrain tectum, Neurosci
Biobehav Rev 18 (1994), pp. 339346.
Brebner, J. T. 1980. Reaction time in personality theory. In A. T. Welford (Ed.),

Reaction Times. Academic Press, New York, pp. 309-320.
Brehm, J.W. & Self, E.A. (1989). The intensity of motivation. Ann. Rev. Psychol.
40:109-131.
Broadbent, D. E. 1971. Decision and Stress. Academic Press, London.
Buchanan, T. W., and Lovallo, W. R. (2001). Enhanced memory for emotional material
following stresslevel cortisol treatment in humans. Psychoneuroendocrinology 26, 307
317.
Butler, R. & Lewis, M. (1973). Aging and mental health: Positive sychosocial
approaches. St. Louis: The C.V. Mosby Company.
C.H. Chen. (1996). The Fuzzy Logic and Neural Network handbook. Mcgraw-Hill
Computer Engineering Series.
Cacioppo, J.T., Petty, R.E., and L.G. Tassinary. 1988. Communication, social
cognition, and affect: A psychophysiological approach. Pp. 219-245 in
Communication, social cognition and affect, eds. L. Donohew, H.E. Sypher and E.T.
Higgins. Hillsdale, NJ: Lawrence Erlbaum Associates.
Cahill, L. and McGaugh, J. L. (1995). A novel demonstration of enhanced memory

associated with emotional arousal , Consciousness and Cognition 4, 410-21
Callister, R., Suwarno, N., & Seals D. (1992). Sympathetic activity is influenced by task
difficulty and stress perception during mental challenge in humans. The Journal of
Physiology. 454 (1), 373-387
Cannon WB, 1927, "The James-Lange theory of emotion", Am J. Psychology; 39:106-

124.
Cannon, W. B. (1914). The emergency function of the adrenal medulla in pain and in
the major emotions. American Journal of Physiology, 33, 356372.
Caplan, R., Cobb, S., French, J., Van Harrison, R., & Pinneau, S. (1975). Demands
and worker health: Main effects and organizational differences. Washington, DC: U.S.
Government Printing Office.
Cardena, E., & Spiegel, D. (1993). Dissociative reactions to the Bay Area Earthquake.
American Journal of Psychiatry, 150, 474478.
Cassel, J. (1974). Psychosocial Processes and Stress. : The Behavioral Sciences and
Preventive Medicine. Washington, DC: Public Health Service.

Cattell, R. B. (1972). The sixteen personality factor. Champaign, IL: Institute for
Personality and Ability Testing.
Chan, M., Hariton, C., Ringeard, P., & Campo, E. (1995). Smart house automation
system for the elderly and the disabled. In IEEE International Conference on Systems,
Man and Cybernetics, pp. 15861589.
Chandon, P. et al. (2007), Measuring the value of point-of-purchase marketing with

commercial eye-tracking data, Working Paper Series INSEAD Business School,
2007/22/MKT/ACGRD
Charles, D, McGlinchey, S. (2004). The past, present and future of artificial neural
networks in digital games. In: Mehdi Q, Gough N, Natkin S, Al-Dabass D (eds)
Proceedings of the 5th international conference on computer games: artificial
intelligence, design and education. The University of Wolverhampton, pp 163169
Choi, J., Shin, D., & Shin, D. (2006). Ubiquitous Intelligent Sensing System for a Smart
Home. In: D.-Y. Yeung et al. (Eds.): SSPR&SPR 2006, LNCS 4109, Berlin: Springer,
pp. 322 330, 2006.
Cochran, R. E., Lee, F. J., & Chown, E. (2006). Modeling emotion: Arousal's impact on
memory In proceedings of the 28th Annual Conference of the Cognitive Science
Society (pp. 1133-1138). Vancouver, British Columbia, Canada.
Cochrane, B., Lee, F.J., & Chown, E. (2006). Modeling Emotion: Arousals impact on
emory. In the Proceedings of the 26th Annual Meeting of the Cognitive Science Society
Cohen, F., & Lazarus, R. S. (1979). Coping with the stresses of illness. In G. Stone, F.
Cohen, &N. Adler (Eds.), Health psychology (pp. 217254). San Francisco: Jossey-
Bass.
Cohn, J.F. and Kanada, T. (2007), Automated facial image analysis for measurement
of emotion expression. In: Coan and Allen (Eds), The Handbook of Emotion, Elicitation
and Assessment, New York: Oxford University Press
Cohn, J.F. and Kanada, T. (2007), Automated facial image analysis for measurement
of emotion expression. In: Coan and Allen (Eds), The Handbook of Emotion, Elicitation
and Assessment, New York: Oxford University Press
Cook DB, O'Connor PJ, Lange G, Steffener J., Neuroimage, 2007, Functional
neuroimaging correlates of mental fatigue induced by cognition among chronic fatigue
syndrome patients and controls, Volkov (1990, op. quoting Garca Manso et al.)
Crawford, J. R., & Henry, J. D. (2004). The Positive and Negative Affect Schedule
(PANAS): Construct validity, measurement properties and normative data in a large
non-clinical sample. British Journal of Clinical Psychology, 43, 245-265.
Crone, Eveline A. et al. (2004), Heart rate and skin conductance analysis of
antecedents and consequences of decision making, Psychophysiology, 41

Czaja, S. (1996). Aging and the acquisition of computer skills (pp. 201-220). In Rogers,
W., Fisk, A., et al., Aging and Skilled Performance: Advances in Theory and
Applications. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Dagnino-Subiabre A, Terreros G, Carmona-Fontaine C, Zepeda R, Orellana JA, Daz-

Vliz G, Mora S, Aboitiz F. Neuroscience. 2005;135(4):1067-74. Epub 2005 Sep 13.
Damasio, A.R., Tranel, D. & Damasio, H. (1991). "Somatic markers and the guidance
of behaviour: theory and preliminary testing" (pp. 217229). In H.S. Levin, H.M.
Eisenberg & A.L. Benton (Eds.). Frontal lobe function and dysfunction. New York:
Oxford University Press.
Darwin, C. (1998). The expression of the emotions in man and animals (3rd ed.). New
York: Oxford University Press.
Davidoff, Linda. 1980, 2a. Introduccin a la Psicologa, McGraw-Hill. Mxico.
Davidoff, Pretty. 1980, 2a. Introduction to Psychology, McGraw-Hill. Mexico.
Davies, D.R. & Tune, G.S. (1969) Human Vigilance Performance. New York: American
Elsevier
Davitz, J. (1964). Auditory correlates of vocal expression of emotional feeling. In J.

Davitz (Ed.), The communication of emotional meaning (pp. 101-112). New York:
McGraw-Hill.
Dawson, M. E, Schell, A. M. and Filion, D. L. (2000), The Electrodermal System, In:

Cacioppo, Tassinary and Filion (Eds), Handbook of Psychophysiology, Cambridge
University Press
Deaton, J. & Parasuraman, R. (1993) Sensory and Cognitive Vigilance: Effect of Age
on Performance and Subjective Workload.
Deppe, M., Schwindt, W.; Kugel, H.; Plassmann, H. And Kenning, P. (2005): Nonlinear
responses within the medial prefrontal cortex reveal when specific implicit information
influences economic decision-making, Journal of Neuroimaging, 15
Deptula D, Singh R, Pomara N., 1993. Aging, emotional states, and memory. Am J
Psychiatry. 1993 Mar;150(3):429-34. Geriatric Psychiatry Center, Nathan S. Kline
Institute for Psychiatric Research, Orangeburg, N.Y.
Der, G., and I. J. Deary. 2006. Age and sex differences in reaction time in adulthood:
Results from the United Kingdom health and lifestyle survey. Psychology and Aging
21(1): 62-73.
Derryberry, D., & Tucker, D. M.1994. Motivating the focus of attention. In P. M.

Niedenthal & S. Kitayama Eds. The hearts eye: Emotional influences in perception and
attention ~pp. 167196. San Diego: Academic Press.
Desmet, P. (2002), Designing Emotions, Delft: Technische Univeriteit

Diagnostic and Statistical Manual of Mental Disorders , 4th ed., text revision (DSM-IV-
TR). Washington, DC: American Psychiatric Press Inc., 2000.
Dinerstein, J., Egbert, P.K., de Garis, H., & Dinerstein, N. (2004). Fast and learnable
behavioral and cognitive modeling for virtual character animation. Comp. Anim. Virtual
Worlds 2004; 15: 95108
Dinges, D.F. (2004). Critical research issues in development of biomathematical

models of fatigue and performance. Aviat Space Environ Med, 75(3, Suppl.):A181191.
Doran et Sustained, 2001, Attention performance during sleep deprivation: evidence of

state instability, Archives of Italian Biology
Doron and LeDoux 1999 N.N. Doron and J.E. LeDoux, Organization of projections to
the lateral amygdala from auditory and visual areas of the thalamus in the rat, J Comp
Neurol 412 (1999), pp. 383409.
Dorrian er al Psychomotor vigilance performance: Neurocognitive assay sensitive to

sleep loss, 2005 Sleep Deprivation. Clinical issues, Pharmacology and Sleep Loss
effects
Dub and Morgan (1998), Trend effects and gender differences in retrospective
judgement of consumer emotions,Journal of Consumer Research, 23
Dustman, R., Emmerson, R., Steinhaus, L., Shearer, D., & Dustman, T. (1992). The
effects of videogame playing on neuropsychological performance of elderly individuals.
Journal of Gerontology: Psychological Sciences, 47(3), 168-171.
Edell, Julie A. and Burke, Marian C. (1987), The Power of Feeling in Understanding
Advertising Effects, Journal of Consumer Research,14 (4)
EHealthMD: What is stress? URL:

http://www.ehealthmd.com/library/stress/STR_whatis.html (accessed May 13, 2010).
Ekman, P. (1972). Universals and Cultural Differences in Facial Expression of Emotion.

In J. Cole ed. Nebraska Symposium on Motivation. Lincoln, Nebraska: University of
Nebraska Press: 207-283
Ekman, P. (1982). Emotion in the human face. Cambridge, UK: Cambridge University
Press.
Ekman, P. (2003). Darwin, deception, and facial expression. Annals of the New York
Academy of Sciences, 1000, 105221.
Ekman, P., & Friesen, W. V. (1967). Head and body cues in the judgment of emotion: A
reformulation. Perceptual and Motor Skills, 24, 711724.
Ekman, P., & Friesen, W. V. (1975). Unmasking the face: A guide to recognizing
emotions from facial clues. Englewood Cliffs, NJ: Prentice-Hall.

Ekman, P., & Friesen, W. V. (1978). Facial action coding system: A technique for the
measurement of facial movement. Palo Alto, CA: Consulting Psychologists Press.
Ekman, P., Friesen, W. V., & Ellsworth, P. (1982). What emotion categories or
dimensions can observers judge from facial behavior? In P. Ekman (Ed.), Emotion in
the human face (pp. 39-55). New York: Cambridge University Press.
Ekman, P., Friesen, W. V., & Hager, J. C. (2002). Facial action coding system. Salt
Lake City, UT: A Human Face.
Ellenbogen et al., 2002 M.A. Ellenbogen, A.E. Schwartzman, J. Stewart and C.D.
Walker, Stress and selective attention: the interplay of mood, cortisol levels, and
emotional information processing, Psychophysiology 39 (2002), pp. 723732.
El-Nasr, M.S., Ioerger, T.R., Yen, J. (1999). PETEEI: A PET with Evolving Emotional
Intelligence, Proc. 3rd Annual Conference on Autonomous Agents, 1999, pp. 9 - 15.
Elzinga, B. M., and Bremner, J. D. (2002). Are the neural substrates of memory the
final common pathway in posttraumatic stress disorder (PTSD)? J. Affect. Disord. 70,
117.
Encyclopedia of childrens health. http://www.healthofchildren.com/M/Munchausen-

Syndrome.html.
Erikson J. 2002. Motivation and Memory Processes. Hanover College.
Eriksson, M., & Papanikotopoulos, N.P. (1997). Eye-tracking for detection of driver
fatigue, Proceedings of the 1997 IEEE Conference on Intelligent Transportation
Systems, Boston, 9-12 November, pp. 314-319.
Ernoehazy W. Jr, MD, FACEP, Medical Director, Emergency Department, Ed Fraser

Memorial Hospital, Florida. Munchausen Syndrome.
http://emedicine.medscape.com/article/805841-overview
Eskelinen, M. (2001). The gaming situation. Game Studies: The International Journal of
Computer Game Research,1. Retrieved August 10, 2001,
http://cmc.uib.no/gamesstudies//0101/eskelinen/.
Eunice Y. Yuen, Wenhua Liu, Ilia N. Karatsoreos, Jian Feng, Bruce S. McEwen, and
Zhen Yan (2009) Acute stress enhances glutamatergic transmission in prefrontal cortex
and facilitates working memory. published ahead of print July 29, 2009,
doi:10.1073/pnas.0906791106
European Society for Traumatic Stress Studies, December 2010

https://www.estss.org/severely-stressful-events-and-dementia-a-study-of-an-elderly-
greek-demented-population-2/
European Transport Safety Council. (2001). The role of driver fatigue in commercial
road transport crashes. Brussels. URL: http://www.etsc.eu/oldsite/drivfatigue.pdf

Everly, G.S and Girdano, D. y. (1986). Controlling stress and tension. 2 Edicin.
Englewood Cliffs, NJ: Prentice-Hall.
Everly, GS, Jr. & Lating, J. (2002). Clinical Guide to the Treatment of the Human Stress
Response. NY: Kluwer.
Ewart, C. K. (1992). The role of physical self-efficacy in recovery from heart attack. In
R. Schwarzer (Ed.), Self-efficacy: thought control of action (pp. 287-304). London:
Taylor Francis.
Eysenck, H.J. & Rachman, S. (1965). The Causes and Cures of Neuroses, London:
Routledge and Kegan Paul.
E-zine Artciles, December 2010, http://ezinearticles.com/?Depression-As-a-

Disability&id=1226663
Fatigue Among Elders Tied to Other Ills, 2008, Study Shows Lack of Energy Could
Signal Health Problems Beyond Just Normal Aging Process, By Kelley Colihan
WebMD Health News
Feldman M. Playing Sick?: Untangling the Web of Munchausen Syndrome,

Munchausen by Proxy, Malingering, and Factitious Disorder. New York: Brunner-
Routledge, 2004.
Fieandt, K. von, A. Huhtala, P. Kullberg, and K. Saarl. 1956. Personal tempo and
phenomenal time at different age levels. Reports from the Psychological Institute, No.
2, University of Helsinki.
Fisk JD, Ritvo PG, Ross L, et al, 1994, Measuring the functional impact of fatigue:
Initial validation of the fatigue impact scale
Fitts, P.M. (1954). The information capacity of the human motor system in controlling
the amplitude of movement. Journal of Experimental Psychology, 47, pp. 381-391.
Flin, R. (1997). Sitting in the Hot Seat. Leaders and teams for Critical Incident
Management.
Flin, R. (2004). Crisis and Disaster Management. Distance Learning Manual. University
of Portsmouth.
Fordyce, W. E., Fowler, R. S., & Delateur, B. (1968). An application of behaviour

modification technique to a problem of chronic pain. Behaviour Research and Therapy,
6, 105-107.
Frank Morelli; Pamela A. Burton (2009) Military Psychology, Volume 21, Issue 1
January 2009 , pages 81 97
Freeman, G. L. 1933. The facilitative and inhibitory effects of muscular tension upon
performance. American Journal of Psychology 26: 602-608.

Freitas-Magalhes, A., & Castro, E. (2009). Facial Expression: The effect of the smile
in the Treatment of Depression. Empirical Study with Portuguese Subjects. In A.
Freitas-Magalhes (Ed.), Emotional Expression: The Brain and The Face (pp. 127-
140).
Frijda, N. H. (1986). The emotions. Cambridge: Cambridge University Press. (Dutch

transl.1988)(Italian transl. 1990).
Frijda, N.H. (2000). The psychologist's point of view. In M. Lewis & J. M. Haviland-
Jones (Eds.), Handbook of emotions. New York: Guilford Press, pp. 59-74.
Fum, D. & Stocco, A. (2004). Memory, emotion, & rationality: An ACT-R interpretation
for gambling task results. In Proceedings of the sixth International Conference on
Cognitive Modeling (pp. 106-111). Pittsburgh, PA: Carnegie Mellon
University/University of Pittsburgh.
Functional and Economic Impact of Sleep Loss and Sleep-Related Disorders, Unmet
Public Health Problem, 2006, Institute of Medicine (US) Committee on Sleep Medicine
and Research; Colten HR, Altevogt BM, editors. Washington (DC): National Academies
Press (US)
Funge, J. (1998). Making Them Behave: Cognitive Models for Computer Animation.
Ph.D. Thesis, University of Toronto
Gajos, K. Z., Wobbrock, J. O., & Weld, D. S., (2007) Automatically generating user
interfaces adapted to users motor and vision capabilities, In proceedings of UIST
2007.
Galata et al., 1999 Galata, G., Bucklin, R. and Hanssens, D. (1999), On the Stability of
Store Format Choice, Working Paper, November.
Galata, A., Johnson, N. & Hogg, D. (1999). Learning Structured Behaviour Models
Using Variable Length Markov Models. IEEE InternationalWorkshop onModelling
People, Corfu, Greece September 20th 1999, pp. 95-102.
Galton, F. 1899. On instruments for (1) testing perception of differences of tint and for
(2) determining reaction time. Journal of the Anthropological Institute 19: 27-29.
Gandevia, S.C. (1992). Some central and peripheral factors affecting human
motoneuronal output in neuromuscular fatigue. Sports medicine (Auckland, N.Z.) 13
(2): 9398.
Geelen, R.J.G.M. & Soons, P.H.G.M. (1996). Rehabilitation: an everyday motivation

model. Patient Education and Counseling, 28: 69-77.
Geen, R. (1994). Human motivation: A psychological approach. Wadsworth Publishing.
Gellhorn, E., & Loofbourrow, G. (1963). Emotions and emotional disorders. NewYork:
Harper & Row.

Gendolla, G.H.E., Richter, M. (2005). Ego-involvement and mental effort:

cardiovascular, electrodermal, and performance effects. Psychophysiology, 42: 595
603.
Gerber A.J., Posner J., Gorman D., Colibazzi T., Yu S., Wang Z., Kangarlu A., Zhu, H.,
Russel, J., & Peterson B.S. (2008). An affective circumplex model of neural systems
subserving valence, arousal, and cognitive overlay during the appraisal of emotional
faces. Neuropsychologia, 46 (8): 2129-2139.
Gevarter, W.(1978). Psychotherapyand the brain.

Unpublishedpaper,NASA,Washington,DC.
Ghaffari, M.; Genaidy, A.M. (2005). A Neural Network Model for Human Performance
Optimization in the Workplace, Neural Engineering,. Conference Proceedings. 2nd
International IEEE EMBS Conference on, pp.648-651.
Giannini, A.J. (1992) Fatigue,Chronic. In RB Taylor Difficult Diagnosis--2. Philadelphia,

WB Saunders Co., p. 156.
Gluck, K. A., Gunzelmann, G., Gratch, J., Hudlicka, E., & Ritter, F. E. (2006). Modeling
the impact ofcognitive moderators on human cognition and performance. In
Proceedings of the 2006, Conference of the Cognitive Science Society, 2658. Mahwah,
NJ: Erlbaum.
Gluck, K. A., Gunzelmann, G., Gratch, J., Hudlicka, E., & Ritter, F. E. (2006, July). The
Impact of Cognitive Moderators on Human Cognition & Performance. Symposium
abstract in R.Sun & N Miyake (Eds.), Proceedings of the Twenty-Eighth Annual
Meeting of the Cognitive Science Society (p. 2658). Mahwah, NJ: Lawrence Erlbaum
Associates.
Gluckman, J. P., Warm, J. S. Dember, W.N., Thielmann, J.A., and Hackcock, P.A.
(1988, November) Subjective Workload response to simultaneous and successive
vigilance tasks. Paper presented at the Annual Meeting of the Psychonomic Society,
Chicago, IL.
Godin, G., & Kok, G. (1996). The Theory of Planned Behavior: a Review of Its
Applications to Health-Related Behaviors. American Journal of Health Promotion, 11,
pp. 87-98.
Goldstein, J., Cajko, L., Oosterbroek, M, Michielsen, M., van Houten, O., & Salverda, F.
(1997). Videospelletjes en ouderen. Gevolgen voor reactietjd, cognitief functioneren en
welbevinden. [Video games and the elderly: Effects on reaction time, cognitive
functioning and well-being. Gedrag & Gezondheid: Tijdschrift voor Psychologie &
Gezondheid, 25(6), 296-302.
Gmez, O. (2000). Relacin entre estrs e inmunidad: Una visin crtica a la

investigacin. Revista Latinoamericana de Psicologa, 32,1, 31-45.

Grandjean, D., Sander, D., & Scherer, K. R. (2008). Conscious emotional experience
emerges as a function of multilevel, appraisal-driven response synchronization.
Consciousness and Cognition, 17(2), 484495. doi:10.1016/j.concog.2008.03.019
Gray, J. A. (1985). The whole and its parts: Behaviour, the brain, cognition and
emotion. Bulletin of the British Psychological Society. 38, 99-112.
Greenberg, H.& Lyman, Hann, D. M., Jacobsen, P. B., Azzarello, L. M., Martin, S. C.,
Curran, S. L., Fields, K. G, 1998, Measurement of fatigue in cancer patients:
Development and validation of the Fatigue Symptom Inventory, Quality of Life
Research, 7, 301-310
Greene, C.M., Bahri, P., Soto, D. (2010). Interplay between affect and arousal in
recognition memory. PLoS One. 5:e11739
Greenleaf, W., & Tovar, M. (1994). Augmenting reality in rehabilitation medicine.

Artificial Intelligence Medicine, 6, 289-299.
Greenwalt, M. K., Cook, E.W. and Lang, P.J. (1989), Affective Judgement and
Psychophysiological Response: Dimensional Covariation in the Evaluation of Pictorial
Stimuli, Journal of Psychophysiology, 3
Groeppel-Klein, Andrea (2005), Arousal and Consumer in-store behavior, Brain

Research Bulletin, 67, p. 428-437
Guimares H.C., Levy R., Teixeira A.L., Beato R.G., Caramelli P. Neurobiology of
Apaty in Alzheimers disease. Arq Neuropsiquiatr 2008;66(2-B):436-443.
Gulian, E., Matthews, G., Glendon, A. I., Davies, D. R., & Debney, L. M. (1989).
Dimensions of driver stress. Ergonomics, 32, 585-602.
Gump, A. et. al. (2002) Application of Fitts Law to individuals with cerebral palsy,
Perceptual and Motor Skills, vol. 94, pp. 883-895.
Gundel, A.; Marsalek, K.; ten Thoren, C. (2005) Support of Mission and Work
Scheduling by a Biomedical Fatigue Model. In Strategies to Maintain Combat
Readiness during Extended Deployments A Human Systems Approach (pp. 28-1
28-12). Meeting Proceedings RTO-MP-HFM-124, Paper 28. Neuilly-sur-Seine, France:
RTO. Available from: http://www.rto.nato.int/abstracts.asp.
Gunzelmann, G., & Gluck, K. A. (2008). Approaches to modeling the effects of fatigue
on cognitive performance. In J. Hansberger (Ed.), Proceedings of the Seventeenth
Conference on Behavior Representation in Modeling and Simulation, pp. 136-145.
Gunzelmann, G., Moore, R. L., Salvucci, D. D., & Gluck, K. A. (2009). Fluctuations in
alertness and sustained attention: Prediction driver performance. In Proceedings of the
9th International Conference of Cognitive Modeling (paper 239), Manchester, United
Kingdom.

Guzman,J. (2008) Methodology to obtain the users human values scale from smart
user models. Department of electronics, Computer Science and Automatic Control.
University of Girona.Spain.
H. J. M. Steeneken and J. H. L. Hansen, Speech under stress conditions: Overview of

the effect on speech production and on system performance,in IEEE Inter. Conf.
Acoustics, Speech, Signal Processing, vol. 4, Phoenix, AZ, Mar. 1999, pp. 20792082.
Hagger, M. S., Chatzisarantis, N. L. D., & Biddle, S. J. H. (2002). A Meta-Analytic

Review of the Theories of Reasoned Action and Planned Behavior in Physical Activity:
Predictive Validity and the Contribution of Additional Variables. Journal of Sport &
Exercise Psychology, 24, 3-32.
Hmlinen, M. et al. (1993), Magnetoencephalography theory, instrumentations

and applications to non-invasive studies of the working human brain, Review of
Modern Physics, vol. 65, no. 2
Hancock, P. A. (1986c). Sustained attention under thermal stress. Psychological

Bulletin, 99, 263-281.
Handbook of Psychology, by Donald K. Freedheim et al., pp 268-269, John Wiley and

Sons, 2003, ISBN 0471392634
Hans Selye, 1936, A Syndrome Produced by Diverse Nocuous Agents, (Department of

Biochemistry, McGill University, Montreal, Canada) Nature (vol. 138, July 4, 1936, p.
32);
Hansen, Flemming, Percy, Larry and Lundsten, Steen (2006), Outstanding Brands
Emotionally Speaking, European Advances in Consumer Research, vol. 7
Harris (1991). T.P. Harrigan and W.H. Harris , A three-dimensional non-linear finite
element study of the effect of cement-prosthesis debonding in cemented femoral total
hip components. Journal of Biomechanics 24 (1991), pp. 10471058.
Harris, J.A., Saltstone, R., & Fraboni, M. (1999) An Evaluation of the Job Stress
Questionnaire with a Sample of Entrepreneurs. Journal of Business and Psychology,
13 (3): 447-455.
Hazif-Thomas C, Thomas P, 2004. Loss of motivation in dependent elderly subjects.

Service de psychogriatrie, CHU Esquirol, 87025 Limoges Cedex, France
Hazlett, R.L. and Hazlett, S.Y. (1999), Emoitonal Resonse to Television Commercials:
Facial EMG vs. Self-report, Journal of Advertising Research, 39
Healey, J.A. & Picard, R.W. (2005). Detecting Stress During Real-World Driving Tasks
Using Physiological Sensors. IEEE Trans. Intelligent Transportation Systems, 6: 156-
166.
Health Place, December 2010:

http://www.healthyplace.com/depression/elderly/depression-in-elderly/menu-id-68/

Helen S. Mayberg et al. Reciprocal Limbic-Cortical Function and Negative Mood:

Converging PET Findings in Depression and Normal Sadness
Helton et al. (2007). The abbreviated vigilance task and cerebral hemodynamics.
Journal of Clinical and Experimental Neuropsychology, 29, 549-552.
Helton, W.S., Matthews, G., Warm, J.S. (2009) Stress state mediation between
environmental variables and performance: The case of noise and vigilance. Acta
Psychologica, 130: 204-213.
Hendy, C. K. Farrell, S.E. (1997).Implementing A Model of Human Information

Processing in A Task Network Simulation Enviroment: Defence and Civil Institute of
Enviromental Medicine, Canada.
Hennessy, 2000. D.A. Hennessy, The interaction of person and situation within the
driving environment: Daily hassles, traffic congestion, driver stress, aggression,
vengeance and past performance. Dissertation Abstracts International: Section B: The
Sciences and it Engineering 60 8-B (2000), p. 4301.
Het, S., Ramlow, G., and Wolf, O. T. (2005). A meta-analytic review of the effects of
acute cortisol administration on human memory. Psychoneuroendocrinology 30, 771
784.
Hobbis, I. C. A., & Sutton, S. (2005). Are techniques used in cognitive behaviour
therapy applicable to behaviour change interventions based on the theory of planned
behaviour? Journal of Health Psychology, 10, 7-18.
Holbrook, M. B. and Batra, R. (1987), Assessing the Role of Emotions as Mediators of

Consumer Responses to Advertising, Journal of Consumer Research, Vol. 14
Hopkins, R. and Fletcher, J.E. (1994), Electrodermal Measurement: Particularly

Effective for Forecasting Message Influence on Sales Appeal, In: Lang, A. (Ed),
Measuring Psychological Responses to Media, New Jersey: Lawrence Erlbaum
http://ibenessere.info/Depressione/depressione-test.html
http://www.ercim.org/publication/Ercim_News/enw53/cunningham.html
Huang, Ming-Hui (2001), The Theory of Emotions in Marketing, Journal of Business

and Psychology, vol. 16, no. 2
HUMAINE Emotion Annotation and Representation Language. http://emotion-

research.net/projects/humaine/earl. Retrieved June 30, 2006.
Humber, G. (1986). Stress y conflicto. Madrid: Parainfo.
Humphrey, J., & Everly, G. S., Jr. (1980). Factor dimensions of stress responsiveness
in male and female students. Health Education, 11, 3839.

Ide, B. A. (1996). Psychometric review of family disruption from Illness Scale. In B. H.

Stamm (Ed.), Measurement of stress trauma, and adaptation. Lutherville, MD: Sidran
Press.
Irving, L. J., Mann, L.(1977) Decision Making A Psychological Analysis of Conflict,

Choice and Commitment. The Free Press New York.
Izard, Carol D. (1977), Human Emotions, New York, Plenum Press
Jack, D., Boian, R., Merians A.S., Tremaine, M., Burdea, G.C., Adamovich, S.V.,
Recce, M., Poizner, H. (2001). Virtual reality-enhanced stroke rehabilitation. IEEE
Transactions on Neural systems and Rehabilitation Engineering, vol. 9, no. 3, pp. 308-
318.
James, W. (1884). What is an emotion? Mind, 9, 188-205.
Jastrzembski, T.S. & Charness, N. (2007). The Model Human Processor and the Older
Adult: Parameter Estimation and Validation Within a Mobile Phone Task. Journal of
Experimental Psychology: Applied, Vol. 13, No. 4, 224248.
Jevas, S. and J. H. Yan. 2001. The effect of aging on cognitive function: a preliminary
quantitative review. Research Quarterly for Exercise and Sport 72: A-49.
Jo-Anne Aubut BA, Shawn Marshall MD MSc FRCPC, Corbin Lippert BScN, Robert
Teasell, Fatigue and Sleep Disorders Post ABI, MD FRCPC
John, B. E. & Kieras, D.E. (1996). The GOMS family of user interface analysis
techniques: Comparison and Contrast. ACM Transactions on Computer Human
Interaction, Vol. 3, pp. 320-351.
Johnston, M. (1994). Models of disability. The Psychologist, 9, 205-210.
Johnston, M., Morrison, V., MacWalter, R., & Partridge, C. (1999). Perceived control,
coping and recovery from disability following stroke. Psychology & Health, 14, 181-192.
Jones, C.M. & Jonsson, I.-M. (2007). Performance Analysis of Acoustic Emotion
Recognition for In-Car Conversational Interfaces. In: C. Stephanidis (Ed.): Universal
Access in HCI, Part II, HCII 2007, LNCS 4555, Berlin: Springer, pp. 411420.
Jones, G., Ritter, F. E., & Wood, D. J. (2000). Using a cognitive architecture to examine
what develops. Psychological Science, 11(2), 93-100.
Jones, R. M., Lebiere, C., & Crossman, J. A. (2007). Comparing modeling idioms in
ACT-R and Soar. Proceedings of the Eighth International Conference on Cognitive
Modeling. Ann Arbor, MI.
Jonides, J. and Nee, D.E. (2006). Brain mechanisms of proactive interference in

working memory. Neuroscience, 139, 181-193.

Kahn, Itamar et al. (2002), The Role of the Amygdala in Signaling Prospective
Outcome of Choice, Neuron, vol. 33, march
Kaplan, S. (1991). Beyond rationality: Clarity based decision-making. In Garling, T. and

Evans, G.W., (Eds.) Environment, Cognition and Action: An Integrated Approach,
pages 171-190. Oxford University Press, New York
Karunaratne, S. & Yan, H. (2001). 3D Animated Movie Actor Training Using Fuzzy
Logic. Proceedings of Computer Graphics International (CGI 01).
Keates, S. & Trewin, S. (2005). Effect of age and Parkinson's disease on cursor
positioning using a mouse. In Proceedings of 7th International ACM/SIGACCESS
Conference on Computers and Accessibility (ASSETS 2005).
Keates, S., Clarkson, J., & Robinson, P. (2000). Investigating the Applicability of User
Models for Motion Impaired Users, In Proceedings ASSETS 2000, ACM/SIGACCESS
Conference on Computers and Accessibility, November 13-15, 2000.
Kemp, B. J. 1973. Reaction time of young and elderly subjects in relation to perceptual
deprivation and signal-on versus signal-off condition. Developmental Psychology 8:
268-272.
Kennard C. (2005):
http://alzheimers.about.com/od/professionalresources/a/emotions_aging.htm.
Kensinger, E. & Corkin, S. (2003). Memory enhancement for emotional words: Are
emotional words more vividly remembered than neutral words? Memory & Cognition.
31, 1169-1180.
Kieras, D. E. (2005). Fidelity Issues In Cognitive Architectures For HCI Modelling: Be

Careful What You Wish For. In Proceedings of 11th International Conference On
Human Computer Interaction (HCII 2005). Las Vegas, July, 2005.
Kieras, D. E., & Meyer, D. E. (1997). A computational theory of executive cognitive

processes and multiple-task performance: Part 1. Basic mechanisms. Psychological
Review, 104, 3-65.
Kirschbaum, C., Pirke, K-M., Hellhammer, D.H. (1993). The Trier Social Stress Test
A tool for investigating psychobiological stress responses in a laboratory setting.
Neuropsychobiology, 28, 76-81.
Kirschbaum, C., Wolf, O. T., Wippich, W., and Hellhammer, D. H. (1996). Stressand
treatment induced cortisol levels associated with impaired declarative memory in
healthy adults. Life Sci. 58, 14751483.
Klein, H. (1989). An integrated control theory model of work motivation. Academy of

Management Review 14(2): 150-172.
Kleinsmith, L. J. & Kaplan, S. (1963). Paired-associate learning as a function of arousal

and interpolated interval. Journal of Experimental Psychology, 65, 190-193.

Knutson, Brian and Greer, Stephanie, M. (2008), Anticipatory affect: neuroal correlates
and consequences for choice, Philosophical Transactions of the Royal Society B
(published online)
Kohn, P.M., & Macdonald, J.E. (1992). The Survey of Recent Life Experiences: A
decontaminated hassles scale for adults. Journal ofBehaviora1 Medicine, 15,221-236.
Kolodyazhniy, V., Kreibig, S.D., Gross, J.J., Roth, W.T., & Wilhelm, F.H. (2010, in
revision). Psychophysiological sensing in affective computing: towards subject-
independent and stimulus-independent classification of film-induced emotions.
Psychophysiology.
Kolodyazhniy, V., Spti, J., Frey, S., Kruchi, K., Gtz, T., Wirz-Justice, A., Cajochen,
C., Wilhelm, F.H. (2011). Estimation of human circadian phase in real life conditions via
a multi-channel ambulatory monitoring system and a multiple regression model. J. Biol.
Rhythms 26 (1) (in press).
Kolodyazhniy, V., Spti, J., Frey, S., Wirz-Justice, A., Kruchi, K., Cajochen, C., &
Wilhelm, F.H. (2009). Estimation of circadian phase using multi-channel ambulatory
recording. Poster presented at the 1st Conference of the Society for Ambulatory
Assessment, Greifswald, Germany.
Kostoulas, T., Ganchev, T., & Fakotakis, N. (2008). Study on Speaker-Independent

Emotion Recognition from Speech on Real-World Data. In: A. Esposito et al. (Eds.): HH
and HM Interaction 2007, LNAI 5042, Berlin: Springer, pp. 235242.
Kowalski-KM, Vaught-C (2001) :TIEMS, June 18-22, 2001, Oslo, Norway. Oslo,
Norway: The International Emergency Management Society
Kowalski-Trakofler-KM, Vaught-C, Scharf-T, Int J Emerg Manag 2003; 1(3):278-289
Kreibig, S.D. (2010a). Autonomic nervous system activity in emotion: A review.

Biological Psychology, 84:394421.
Kreibig, S.D. (2010b). Emotion, motivation, and cardiovascular response. In R. A.

Wright, & G. H. E. Gendolla, Motivation perspectives on cardiovascular response.
Washington, D. C.: American Psychological Association (In press).
Kreibig, S.D., Gendolla, G.H.E., Scherer, K.R. (2010). Psychophysiological effects of

emotional responding to goal attainment. Biological Psychology, 84: 474487.
Kreibig, S.D., Wilhelm, F. H., Roth, W. T., & Gross, J. J. (2007). Cardiovascular,
electrodermal, and respiratory response patterns to fear- and sadness-inducing films.
Psychophysiology, 44, 787806.
Krugman, D.M. et al. (1994), Do adolescents attend to warnings in cigarette

advertising? An eye-tracking approach, Journal of Advertising Research, vol. 34 (6)
Kryter, K. D. (1970). The effects of noise on man. NewYork: Academic

Kuhns R., 2004 Fighting Memory Loss and Reducing Fatigue With Elderly Parents,
The Columbia Encyclopedia, Sixth Edition Copyright 2004, Columbia University Press
Kumral E, Bayulkem G, Evyapan D, Yunten N. Spectrum of anterior cerebral artery

territory infarction: clinical and MRI findings. Eur J Neurol 2002;9:615-624.
http://www.scielo.br/pdf/anp/v66n2b/v66n2ba35.pdf
L. Pour Mohammad Bagher, Intelligent Agent System Simulation Using Fear Emotion,
Conference on Machine Intelligence (ICMI 2008), 2008
Labrador, F.J. (1992) El estrs: nuevas tcnicas para su control. Madrid: Temas de
Hoy. Isbn 84-7880-170-7
Laird, J. E., Newell, A., & Rosenbloom, P. S. (1987). SOAR: An architecture for general
intelligence. Artificial Intelligence, 33, 1-64.
Lang, P. J. (1980). Behavioral treatment and bio-behavioral assessment: computer

applications. In J. B. Sidowski, J. H. Johnson, & T. A. Williams (Eds.), Technology in
mental health care delivery systems (pp. 119-l37). Norwood, NJ: Ablex.
Lang, P. J. (1995). The Emotion Probe: Studies of motivation and attention, A study in
the Neuroscience of Love and Hate. Hillsdale, NJ: Lawrence Erlbaum Associates,
Publishers.
Lang, P.J. (1990), Involuntary attention and psychophysiological arousal evoked by

structural features and mild emotion in TV commercials, Communication Research, 17
(3).
Lang, P.J., Greenwald, M.K., Bradley, M.M., & Hamm, A.O. (1993). Looking at pictures:
Affective, facial, visceral and behavioral reactions. Psychophysiology, 30: 32613273.
Lange, C. (1887). Ueber Gemuthsbewgungen, 3, 8
Lawhorn, Tommie; Ennis, Demetria; Lawhorn, David C. Senior adults and computers in
the 1990s. Educational Gerontology, 22(2), Mar 1996, 193-201.
Lawton, C., Cook, S., May, A., Clemo, K., & Brown (2008). Postural support strategies
of disabled drivers and the effectiveness of postural support aids. Applied Ergonomics
39: 4755.
Lazarus, R. S. (1966). Psychological stress and the coping process. New York:
McGraw-Hill.
Lazarus, R.S., & Folkman, S. (1984). Stress, Appraisal and Coping. New York:
Springer.
Le Doux, Joseph E. (1996), The Emotional Brain, Phoenix
LeDoux, 2000; hman & Mineka, 2001

Levenson R (1994). Human emotion: A functional view. In Ekman P, Davidson RJ

(eds.), The Nature of Emotion: Fundamental Questions: 123-30. New York: Oxford
University Press.
Lewis, P.A., Critchley, H.D., Rotshtein, P., & Dolan, R.J. (2006). Neural Correlates of
Processing Valence and Arousal in Affective Words. Cerebral Cortex, 17(3):742-748.
Libin, A. (2001) Virtual Reality as a Complex Interactive System: A multidimensional

model of person - artificial partner co-relations. Proceedings of the 7th International
Conference on Virtual Systems and Multimedia (VSMM01).
Libow J. "Child and Adolescent Illness Falsification." Pediatrics 105, no. 2 (February
2000), 336.
Lisetti, C.L. & Nasoz, F. (2005). Affective Intelligent Car Interfaces with Emotion
Recognition. In Proceedings of 11th International Conference on Human Computer
Interaction, Las Vegas, NV, USA, July 2005.
Liston C, Miller MM, Goldwater DS, Radley JJ, Rocher AB, Hof PR, Morrison JH,
McEwen BS. J Neurosci. 2006 Jul 26;26(30):7870-4.
Liston, B. S. McEwen, and B. J. Casey Psychosocial stress reversibly disrupts

prefrontal processing and attentional control PNAS 2009 : 0807041106v1-
pnas.0807041106
Liu, X., Fengliang, X., & Fujimura, K. (2002). Real time eye detection and tracking for
driver observation under various light conditions. Proc. IEEE Intelligent Vehicle
Symposium (IV '02), vol. 2, pp. 344-351, Versailles, France, June 2002.
Liying, L. & Haoxiang, Q. (2008). The Study of Driver Fatigue Monitor Algorithm
Combined PERCLOS and AECS. Proc. 2008 Int. Conf. Computer Science and
Software Engineering, pp. 349-352.
Loeb, M., and Jeantheau, G. (1958). The inuence of nox-ious environmental stimuli on
vigi1ance.Journa1ofApplied Psychology, 42, 47-49
Luce, R. D. 1986. Response Times: Their Role in Inferring Elementary Mental

Organization. Oxford University Press, New York.
Luchies, C. W., J. Schiffman, L. G. Richards, M. R. Thompson, D. Bazuin, and A. J.

DeYoung. 2002. Effects of age, step direction, and reaction condition on the ability to
step quickly. The Journals of Gerontology, Series A 57(4): M246.
Luethi M, Meier B and Sandi C (2009). Stress effects on working memory, explicit
memory, and implicit memory for neutral and emotional stimuli in healthy men. Front.
Behav. Neurosci. 2:5. doi: 10.3389/neuro.08.005.2008
Lugaresi A., Montagna P., Morreale A., Gallassi R. Psychic akinesia following carbon
monoxide poisoning. Eur Neurol 1990;30:167-169.

Lundberg, U. & Frankenhaeuser, M. (1979) Pituitary-adrenal and sympathetic adrenal

correlates of distress and effort (Report 548). Stockholm, Sweden: University of
Stockholm, Department of Psychology
Lupien, S. J., and Lepage, M. (2001). Stress, memory, and the hippocampus: cant live
with it, cant live without it. Behav.Brain Res. 127, 137158.
Lupien, S. J., Fiocco, A., Wan, N., Maheu, C. L., Schramek, T., and Tu, M. T. (2005).
Stress hormones and human memory function across the lifespan.
Psychoneuroendocrinology 30, 225242.
Lupien, S. J., Gillin, C. J., and Hauger, R. L. (1999). Working Memory is more sensitive
than declarative memory to the acute effects of corticosteroids: a doseresponse study
in humans. Behav. Neurosci. 113, 420430.
Lupien, S. J., Maheu, F., Tu, M., Fiocco, A., and Schramek, T. E. (2007). The effects of
stress and stress hormones on human cognition: implications for the fi eld of brain and
cognition. Brain Cogn. 65, 209237.
Lupien, S., Lecours, A.R., Lussier, I., Schwartz, G., Nair, N.P.V. and Meaney, M.J.,
1994. Basal cortisol levels and cognitive deficits in human aging. J. Neurosci. 14, pp.
28932903 begin_of_the_skype_highlighting 28932903
end_of_the_skype_highlighting. View Record in Scopus | Cited By in Scopus (298)
Mac Namee, B. & Cunningham, P. (2003a). Creating Socially Interactive Non Player
Characters: The -SIC System. International Journal of Intelligent Games and
Simulation, Vol 2 No1, pp. 28-35.
Mac Namee, B., & Cunningham, P. (2003b). Enhancing Non Player Characters in
Computer Games using Psychological Models European Research Consortium for
Informatics and Mathematics News , pp.37-38.
http://www.ercim.org/publication/Ercim_News/enw53/cunningham.html
MacDonald, S. W. S., L. Nyberg, J. Sandblom, H. Fischer, and L. Backman. 2008.

Increased response-time variability is associated with reduced inferior parietal
activation during episodic recognition in aging. Journal of Cognitive Neuroscience
20(5): 779-787.
Macedo et al 2005 C.E. Macedo, G. Cuadra, V. Molina and M.L. Brando, Aversive
stimulation of the inferior colliculus changes dopamine and serotonin extracellular
levels in the frontal cortex: modulation by the basolateral nucleus of amygdala,
Synapse 55 (2005), pp. 5866.
Mackay, C. J., Cox, T., Burrows, C. G., & Lazzerini, A. J. (1978). An inventory for the
measurement of self reported stress and arousal. British Journal of Social and Clinical
Psychology, 17,283-284.
MacKenzie, I. S. (2003). Motor behaviour models for human-computer interaction. In J.

M. Carroll (Ed.) HCI models, theories, and frameworks: Toward a multidisciplinary
science. 27-54. San Francisco: Morgan Kaufmann.

Mackie, R.R. (1984). Research relevance and the information glut. Human Factors
Review, (pp 111), Santa Monica, CA: Human Factors Society
Maclean, N., Pound, P., Wolfe, C., & Rudd, A. (2002). The concept of patient
motivation: a qualitative analysis of stroke professionals attitudes. Stroke, 33(2): 444-
448.
Mactavish J. , Iwasaki Y, 2005, The Journal of Rehabilitation, Vol. 71
Makara, G., Palkovits, M., & Szentagothal, J. (1980). The endocrine hypothalamus and
the hormonal response to stress. In H. Selye (Ed.), Selyes guide to stress research
(pp. 280337). New York: Van Nostrand Reinhold.
Mano, Haim and Oliver, Richard L. (1993), Assessing the Dimensionality and
Structure of the Consumption Experience: Evaluation, Feelings, and Satisfaction,
Journal of Consumer Research, 20
Maren et al 2001 S. Maren, K. Yap and A. Goosens, The amygdala is essential for the
development of neuronal plasticity in the medial geniculate nucleus during auditory fear
conditioning in rats, J Neurosci 21 (2001), p. RC135.
Margaret M. Bradley and Peter J. Lang, Measuring emotion: The self-assessment

manikin and the semantic differentia University of Florida, U.S.A, 28 May 2002.
Mariano Chliz (2005): Psicologa de la emocin: el proceso emocional.

www.uv.es/=choliz
Marshall, W. H., S. A. Talbot, and H. W. Ades. 1943. Cortical response of the

anaesthesized cat to gross photic and electrical afferent stimulation. Journal of
Nerophysiology 6: 1-15.
Mart E.G. Munchausen's Syndrome by Proxy Reconsidered. Manchester, NH: Bally

Vaughn, 2002.
Mason, J. B. (1971). A re-evaluation of the concept of non-specificity in stress theory.

Journal of Psychiatric Research, 8, 323333.
Mason, J. W. (1968b). Organization of psychoendocrine mechanisms. Psychosomatic

Medicine, 30 (entire P. 2).
Mason, J. W. (1972). Organization of psychoendocrine mechanisms: A review and

reconsideration of research. In N. Greenfield & R. Sternbach (Eds.), Handbook of
psychophysiology (pp. 376). New York: Holt, Rinehart & Winston.
Mason, J. W., Maher, J., Hartley, L., Mougey, E., Perlow, M., & Jones, L. (1976).
Selectivity of corticosteroid and catecholamine responses to various natural stimuli. In
G. Servan (Ed.), Psychopathology of human adaptation (pp. 147-171). New York:
Plenum Press.

Massa, P., Watkins, C., Partridge, B (2002). Decision Making Under Stress. Available
from: http://web.umr.edu/~bpart/eman313/DMUS.htm [accessed 16/05/05]
Matthews, G. & Desmond, P.A. (1998) Personality and multiple dimensions of task-
induced fatigue: a study of simulated driving. Personality and Individual Differences,
25: 443-458.
Matthews, G., & Desmond, P.A. (2002) Task-induced fatigue states and simulated
driving performance. Quarterly Journal of Experimental Psychology, 55A, 659-686.
Matthews, G., Campbell, S. E., Falconer, S., Joyner, L. A., Huggins, J., Gilliland, K., et
al. (2002). Fundamental dimensions of subjective state in performance settings: Task
engagement, distress, and worry. Emotion, 2, 315340.
Matthews, G., Emo, A. K., Funke, G., Zeidner, M., Roberts, R. D., Costa, P. T., et al.
(2006). Emotional intelligence, personality, and task-induced stress. Journal of
Experimental Psychology: Applied, 12, 96107.
Matthews, G., Joyner, L., Campbell, S. E., Gilliland, K., Huggins, J., & Falconer, S.
(1999). Validation of a comprehensive stress state questionnaire: Towards a state big
three? In I. Merville, I. J. Deary, F. DeFruyt, & F. Ostendorf (Eds.). Personality
psychology in Europe (Vol. 7, pp. 335350). Tilburg: Tilburg University Press.
Matthews,G.,&Falconer,S.(2002).Personality,coping,andtask induced stress in

customer service personnel.In Proceedings of the Human Factors and Ergonomics
Society 46th Annual Meeting (pp.963
967).SantaMonica,CA:HumanFactorsandErgonomicsSociety.
Mayo Clinic Staff. (2010). Depression (major depression). Retrieved September 20,
2010, from http://www.mayoclinic.com/health/depression/DS00175
McClure et al. 2007, Emotion Regulation and Intelectual disability. Department of

Psychology,La Salle University,Philadelphia, PA.
Mcconatha, D., McConatha, J., & Dermigny, R. (1994). The use of interactive computer
services to enhance the quality of life for long-term care residents. Gerontologist, 34(4),
553-556.
McDougall, W. (1926). An introduction to social psychology. Boston: Luce.
McGaugh, J. L. 2006. Make mild moments memorable: add a little arousal. Trends in
Cognitive Sciences 10(8): 345-347.
McGaugh, J. L. Memory: A Century of Consolidation. Science, 2000, 287, 248-251.
McNAIR, D. M., LORR, M., & DROPPLEMAN, L. F. (1971) Manual for the Profile of
Mood States. San Diego, CA: Educational and Industrial Testing Services.
McNAIR, D. M., LORR, M., & DROPPLEMAN, L. F. (1971) Manual for the Profile of
Mood States. San Diego, CA: Educational and Industrial Testing Services.

Meharabien, A. and Russell, J. A. (1974), An Approach to Environmental Psychology,

Cambridge, MIT Press
Meichenbaum, D. (1985). Stress innoculation training. New York: Plenum Press.
Meichenbaum, D., & Jaremko, M. (1983). Stress reduction and prevention. New York:
Plenum Press.
Mendoza T, Wang XS, Cleeland CS, Morrissey M, Johnson BA, Wendt JK, Huber SL.
1999;85:1186-96, The Rapid Assessment of Fatigue Severity in Cancer Patients: Use
of the Brief Fatigue Inventory. Cancer
Merriam-webster [accessed on November 2010]: http://www.merriam-

webster.com/medlineplus/psychology
Meyer, D. E., Glass, J. M., Mueller, S. T., Seymour, T. L., & Kieras, D. E. (2001).
Executive-process interactive control: A unified computational theory for answering
twenty questions (and more) about cognitive ageing. European Journal of Cognitive
Psychology, 13, 123-164.
Mihailidis, A., Carmichael, B., Boger, J., & Fernie, G. (2003). An Intelligent Environment
to Support Aging-in-Place, Safety, and Independence of Older Adults with Dementia.
Proceedings of UbiHealth 2003: The 2nd Int. Workshop on Ubiquitous. Computing for
Pervasive Healthcare Applications, Seattle, WA, October 12. CD-ROM.
Miller, L., & Smith, A. (1982). The Stress Audit Questionnaire. Boston: Neuromedical
Consultants.
Mizoguchi K, Yuzurihara M, Ishige A, Sasaki H, Chui DH, Tabira T (2000) Chronic

stress induces impairment of spatial working memory because of prefrontal
dopaminergic dysfunction. J Neurosci 20:15681574.
Mogg, K., A. Holmes, M. Garner, and B. P. Bradley. 2008. Effects of threat cues on
attentional shifting, disengagement and response slowing in anxious individuals.
Behavior Research and Therapy 46(5): 656-558.
Moline, J. (1997). Virtual reality in health care: A survey. In G. Riva (Ed.), Virtual reality
in neuro-psychophysiology (pp. 3-34). Amsterdam: IOS Press.
Mooradian, T.A. and Oliver, J.M. (1997), I cant get no satisfaction: The impact of
personality and emotion on post-purchase processes, Psychology and Marketing, 14
Moore, D.G. (2001). Reassessing emotion recognition performance in people with

mental retardation: A review. American Journal on Mental Retardation, 106, 481502.
Morris, J.D. et al. (2002), The Power of Affect: Predicting Intention, Journal of
Advertising Research, 42
Mowrer, O. H. (1960). Learning theory and behavior. New York: Wiley.

Myers, David G. (2004) "Theories of Emotion." Psychology: Seventh Edition, New

York, NY: Worth Publishers.
Mylonakis, V., Soldatos, J., Pnevmatikakis, A., Polymenakos, L., Sorin, A., &
Aronowitz, H. (2008). Using Robust Audio and Video Processing Technologies to
Alleviate the Elderly Cognitive Decline. Proc. 1st Int. Conf. on PErvasive Technologies
Related to Assistive Environments (PETRA 2008).
National Institute of Mental Health: U.S. Department of Healthand Human Services -

National Institutes of Health. Attention Deficit Hiperactive Disorder (ADHD).
http://www.nimh.nih.gov/health/publications/attention-deficit-hyperactivity
disorder/adhd_booklet.pdf
Natter M., Ockerman J., Baumgart L., Review of Cognitive Metrics for C2 13th ICCRTS.
Nauta, W. (1979). Expanding borders of the limbic system concept. In T. Rasmussen &
R. Marino (Eds.), Functional neurosurgery (pp. 723). New York: Raven Press.
Neuberger, G.B. (2003) Arthritis & Rheumatism (Arthritis Care & Research). 49 (5S):
S175S183.
Newcomer, J. W., Craft, S., Hershey, T., Askins, K., and Bargett, M. E. (1994).
Glucocorticoid-induced impairment in declarative memory performance in adult
humans. J. Neurosci. 14, 20472053.
Newell, A. (1990). Unified theories of cognition. Cambridge, MA: Harvard University

Press.
Nikola N. Trajanovic and Colin M. Shapiro, Fatigue and sleepiness in the elderly: risk
factors and management strategies, pp. 289-298
OHare, D. (1999). Human performance in general aviation. Aldershot, England:

Ashgate.
Oatley, K., & Johnson-Laird, P. N. (1987). Towards a cognitive theory of emotions.

Cognition & Emotion, 1, 29-50.
Obrist, P.A. (1981). Cardiovascular Psychophysiology. Plenum, New York.
Obrist, P.A., Gaebelein, C.J., Teller, E.S., Langer, A.W., Grignolo, A., Light, K.C., &
McCubbin, J.A. (1978). The relationship among heart rate, carotid dP/dt, and blood
pressure in humans as a function of type of stress. Psychophysiology, 15 (2): 102115.
Oei, N. Y. L., Everaerd, W. T. A. M., Elzinga, B. M., van Well, S., and Bermond, B.
(2006). Psychological stress impairs working memory at high loads: an association with
cortisol levels and memory retrieval. Stress 9, 133141.
Orbell, S., Johnston, M., Rowley, D., Davey, P., & Espley, A. (2001). Self-efficacy and
goal importance in the prediction of physical disability in people following
hospitalization: A prospective study. British Journal of Health Psychology, 6, 25-40.

Ortony, A., & Turner, T. J. (1990). Whats Basic About Basic Emotions. Psychological
Review, 97(3), 315-331. (Myers and Deepa)
Osgood, C., Suci, G., & Tannenbaum, P. (1957). The measurement of meaning.
Chicago: University of Illinois Press.
Ott-Chevert, C., Rueegger-Frey, B., Klaghofer, R., & Six, P. (1998). Evaluation of a
computer-assisted cognitive training with elderly patients in a geriatric hospital.
[German]. Zeitschrift fuer Gerontopsychologie & Psychiatrie, 11(1), 13-33.
Oxford English Dictionary. http://dictionary.oed.com/
Pacak K, Palkovits M. (2001) Stressor specificity of central neuroendocrine responses:

implications for stress-related disorders. Endocr Rev 2001;22: 502 48.
Palmer, G.K. (Ed.). (1990). Proceedings from the Intermountain Leisure Symposium.
Provo, UT: Brigham Young University.
Palmer, S., Cooper, C., & Thomas, K. (2004). A model of work stress to underpin the
Health and Safety Executive advice for tackling work-related stress and stress risk
assessment. Counselling at Work Winter 2004, pp. 2-5. URL:
www.bacpworkplace.org.uk/journal_pdf/acw_winter04_a.pdf
Panayiotou, G. and S. R. Vrana. 2004. The role of self-focus, task difficulty, task self-
relevance, and evaluation anxiety in reaction time performance. Motivation and
Emotion 28(2): 171-196.
Panksepp, J. (1982). Toward a general psychobiological theory of emotions. The

Behavioral and Brain Sciences, 5, 407-467.
Parasuraman, R. & Giambra, L. (1991) Skill Development in Vigilance: Effects of Event

Rate and Age. Psychology and Aging, 6(2), 155-169.
Parasuraman, R. (1976) Consistency of Individual Differences of Human Vigilance

Performance: An Abilities Classification Analysis. Journal of Applied Psychology Vol.
61 No. 4 486-492
Parrott, W. (2001), Emotions in Social Psychology, Psychology Press, Philadelphia
Pashler, H., & Johnston, J. C. (1998). Attentional limitations in dual-task performance.

In H. Pashler (Ed.), Attention. Psychology Press/Erlbaum (Uk) Taylor & Francis, Hove,
England UK. pp 155-189.
Patrick Gebhard (2005). ALMA A Layered Model of Affect Proceedings of the Fourth
International Joint Conference on Autonomous Agents and Multiagent Systems
(AAMAS05), 29-36, Utrecht. P. Gebhard, K.H. Kipp (2006)
Paul Ekman (1999). Basic Emotions. In T. Dalgleish and M. Power (Eds.). Handbook of
Cognition and Emotion. Sussex, U.K.: John Wiley & Sons, Ltd.

Payne, J. D., Jackson, E. D., Hoscheidt, S., Ryan, L., Jabos, W. J., and Nadel, L.
(2007). Stress administered prior to encoding impairs neutral but enhances emotional
long-term episodic memories. Learn. Mem. 14, 861868.
Payne, Jessica Danielle (2005). Stress and episodic memory: The fate of neutral
versus emotional information. Ph.D. dissertation, The University of Arizona, United
States -- Arizona. Retrieved January 14, 2011, from Dissertations & Theses: The
Sciences and Engineering Collection.(Publication No. AAT 3167250).
Pecchinenda, A., & Smith, C.A. (1996). The affective significance of skin conductance
activity during a difficult problem-solving task. Cognition and Emotion, 10: 481503.
Peniston, L. (1991). The effects of a microcomputer training program on short-term

memory in elderly individuals. Dissertation Abstracts International, 52(2-B), 79.
Perry, L.S. (2010). Designing the Workplace for the Aging Workforce: How to Use
Ergonomics to Improve the Workplace Design,
http://www.zurichna.com/NR/rdonlyres/AEFC0FF5-EE0B-4765-B5D4-
F640D99E1412/0/Designingtheworkplacefortheagingworkforce.pdf. Website accessed
on 7 May September 2010.
Peters, B. (2001). Driving performance and workload assessment of drivers with

tetraplegia: an adaptation evaluation framework. J. Rehabil. Res. Dev. 38 (2).
Petrac DC, Bedwell JS, Renk K, Orem DM, Sims V. Stress. 2009;12(4):313-9.
Petty, R. E., & Cacioppo, J. T. (1986). Communication and Persuasion: Central and
Peripheral Routes to Attitude Change. New York: Springer-Verlag.
Picard, R. (1997). Affective Computing. The MIT Press, Cambridge, MA.
Picard, R.W. & Du, C. (2002). Monitoring Stress and Heart Health with a Phone and
Wearable Computer. Motorola Offspring Journal, Volume 1, November 2002.
Picard, R.W. (1995). Affective Computing. Technical Report TR 321.

http://affect.media.mit.edu/publications.php
Picard, R.W., Vyzas, E., & Healey, J. (2001). Toward Machine Emotional Intelligence:
Analysis of Affective Physiological State. IEEE Trans. Pattern Analysis and Machine
Intelligence, 23(10): 1175-1191.
Piron, H. 1920. Nouvelles recherches sur l'analyse du temps de latence sensorielle et

sur la loi qui relie ce temps a l'intensit de l'excitation. Anne Psychologique 22: 58-
142.
Pieters, R.; Rosbergen, E. and Wedel, M. (1999), Visual attention to Repeated Print
Advertising: A test of scanpath Theory, Journal of Marketing Research, vol. 16, no. 4
Plutchik, Ax Albert F. A critique of determinants of emotional state by schachter and

singer (1962)

Plutchik, R. (1980). A general psychoevolutionary theory of emotion. In R. Plutchik & H.

Kellerman (Eds.), Emotion: Theory, research, and experience: Vol. 1. Theories of
emotion (pp. 3-33). New York: Academic.
Plutchik, R. (1980). Emotion: A psychoevolutimary synthesis. New Tibrk: Harper &

Row.
Plutchik, R. and Kellerman, H (1974), Emotion Profile Index Manual, Western

Psychological Services, LA
Poels, Karolien and DeWitte, Siegfried (2006), How to capture the heart? Reviewing
20 years of emotion measurement in advertising, Working paper Department of
Marketing and Organisation Studies, KU Leuven
Portegys, T. (1999). A Connectionist Model of Motivation. Proc. Int. Joint. Conf. Neural
Networks (IJCNN'99), pp. 80-85.
Porter R. S., Kaplan J. L., Homeier B. P. The Merck Manuals - On Line medical Library.
http://www.merckmanuals.com/home/sec06/ch091/ch091j.html
http://www.merckmanuals.com/home/sec07/ch107/ch107b.html?qt=motivation&alt=sh
Prothero, J., Hoffman, H.G., Parker, D.E., et al. (1995). Foreground/background

manipulations affect presence [On-line]. Available:
www.hitl.washington.edu/publications/p-953//p-953.rtf.
Putman, P., van Honk, J., Kessels, R. P. C.,Mulder, M., and Koppeschaar, H. P. F.
(2004). Salivary cortisol and short and long-term memory for emotional faces in healthy
young women. Psychoneuroendocrinology 29, 953960.
Quervain, D. J., Henke, K., Aerni, A., Treyer, V., McGaugh, J. L., Berthold, T., Nitsch,
R. M., Buck, A., Roozendaal, B., and Hock, C. (2003). Glucocorticoidinduced
impairment of declarative memory is associated with reduced blood flow in the medial
temporal lobe. Eur. J. Neurosci. 17, 12961302.
Quervain, D. J.-F., Roozendaal, B., Nitsch, R. M., McGaugh, J. L., and Hock, C.
(2000). Acute cortisone administration impairs retrieval of long-term declarative
memory in humans. Nat. Neurosci. 3, 313314.
Ravaja N, Turpeinen M, Saari T, Puttonen S, Keltikangas-Jrvinen L (February 2008).

"The psychophysiology of James Bond: phasic emotional responses to violent video
game events". Emotion 8 (1): 11420. doi:10.1037/1528-3542.8.1.114. PMID
18266521. http://content.apa.org/journals/emo/8/1/114.
Ravaja, N. (2004), Contributions of psychophysiology to media research: Review and

Recommendations, Media Psychology, 6 (2)
Reed et al 1996 P. Reed, C. Mitchell and T. Nokes, Intrinsic reinforcing properties of

putatively neutral stimuli in an instrumental two-lever discrimination task, Anim Learn
Behav 24 (1996), pp. 3845.

Reed, M. A., & Derryberry, D. 1995. Temperament and attention to positive and
negative trait information. Personality and Individual Differences, 18, 135147.
Reifman, J. (2004) Alternative methods for modeling fatigue and performance. Aviat
Space Environ Med 2004; 75(3, Suppl.):A17380.
Reiman, E., Raichle, M. E., Robins, E., Butler, F. K., Herscovitch, P., Fox, P., &
Perlmutter, J. (1986). The application of positron emission tomography to the study of
panic disorder. American Journal of Psychiatry, 143, 469477.
Rejeski, W. J., Miller, M. E., Foy, C., Messier, S., & Rapp, S. (2001). Self-efficacy and
the progression of functional limitations and self-reported disability in older adults with
knee pain. Journals of Gerontology Series B-Psychological Sciences and Social
Sciences, 56, S261-S265.
Reynolds, R.M. et al. Diabetes Care 2010, 33(4): 714).
Richins, Marsha L. (1997),Measuring Emotions in Consumption Experience, Journal

of Consumer Research, vol. 24 (sep)
Richter, M., Friedrich, A., Gendolla, G.H.E. (2008). Task difficulty effects on cardiac
activity. Psychophysiology 45, 869875.
Richter, M., Gendolla, G.H.E. (2009). The heart contracts to reward: monetary
incentives and preejection period. Psychophysiology, 46: 451457.
Rickel, J.W. and Johnson, W.L. (1998). STEVE: A Pedagogical Agent for Virtual
Reality (video), Proceedings of the 2nd International Conference on Autonomous
Agents, May 1998.
Rigas, G., Katsis, C.D., Bougia, P., & Fotiadis, D.I. (2008). A Reasoning-Based
Framework for Car Drivers Stress Prediction. 16th Mediterranean Conference on
Control and Automation, pp. 627-632
Ritter, F. E., Belavkin, R. V., & Elliman, D. G. (1999a). Affective Computing: The role of
emotion in human-computer interaction. In A. Monk, A. Sasse, & A.Crerar (Eds.)
Proceedings of the British HCI Group one-day meeting in conjunction with University
College London
Ritter, F. E., Reifers, A. L., Klein, A. C., & Schoelles, M. J. (2006). Lessons from
defining theories of stress. In W. Gray (Ed.) Integrated Models of Cognitive Systems.
New York, NY: Oxford University Press.
Riva, G. Wiederhold, B., & Molinari, E. (Eds.). (1998). Virtual Environments in Clinical
Psychology and Neuroscience: Methods and Techniques in Advanced Patient-
Therapist Interaction. Amsterdam: IOS Press.
Riva,G. (Ed.) (1997). Virtual reality in Neuro-psycho-physiology: Cognitive, Clinical and

Methodological Issues in Assessment and rehabilitation. Amsterdam: IOS Press.

Rivera-Illingworth, F., Callaghan, V. & Hagras, H. (2007). Detection Of Normal and

Novel Behaviours In Ubiquitous Domestic Environments. The Computer Journal.
Volume 53, Issue 2, pp. 142-151.
Robinson, E. S. 1934. Work of the integrated organism. In C. Murchison (Ed.),

Handbook of General Experimental Psychology, Clark University Press, Worcester,
MA.
Roozendaal, B. (2002). Stress and memory: opposing effects of glucocorticoids on

memory consolidation and memory retrieval. Neurobiol. Learn. Mem. 78, 578595.
Rose, S. A., J. F. Feldman, J. J. Jankowski, and D. M. Caro. 2002. A longitudinal study

of visual expectation and reaction time in the first year of life. Child Development 73(1):
47.
Roseman, Ira J. (1991), Appraisal Determinants of Discrete Emotions, Cognition and

Emotion, 5
Rossiter, J.R. and Thornton, J. (2004), Fear-pattern Analysis supports the Fear-drive
Model of Antispeeding Road-safety TV Ads, Psychology & Marketing, 21 (11)
Ruiz, R., Plantin de Hugues, P., & Legros, C. (2009). Voice analysis as a significant
parameter of car drivers fatigue. In Proc. ACOUSTICS High Tatras 2009 34th
International Acoustical Conference - EAA Symposium.
Russell, J. A., Bachorowski., J. A., & Fernandez-Dols, J. M. (2003). Facial and vocal
expressions of emotion. Annual Review of Psychology, 54, 329 349.
Salvucci, D.D. & Lee, F.J. (2003). Simple cognitive modelling in a complex cognitive
architecture, In Proceedings of the ACM/SIGCHI Conference on Human Factors in
Computing Systems, Fort Lauderdale, FL, 2003, 265272.
Salvucci, D.D. (2006). Modeling Driver Behavior in a Cognitive Architecture HUMAN

FACTORS, Vol. 48, No. 2, Summer 2006, pp. 362380.
Salvucci, D.D., Boer, E.R., & Liu, A. (2001). Toward an integrated model of driver
behavior in a cognitive architecture. Transportation Research Record, No. 1779.
Salvucci, D.D., Chavez, A. K., & Lee, F. J. (2004). Modeling effects of age in complex
tasks: A case study in driving. Proc. 26th Annual Conference of the Cognitive Science
Society. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 1197-1202.
Samuele M. Marcora, Walter Staiano, and Victoria Manning, 2009, Mental fatigue
impairs physical performance in humans, School of Sport, Health and Exercise
Sciences, Bangor University, Bangor, Wales, United Kingdom
Sanders, A. F. 1998. Elements of Human Performance: Reaction Processes and

Attention in Human Skill. Lawrence Erlbaum Associates, Publishers, Mahwah, New
Jersey. 575 pages.

Sandi, C., and Pinelo-Nava, M. T. (2007). Stress and memory: behavioral effects and
neurobiological mechanisms. Neural Plast. 2007, 78970.
Sano, M. (1988). Using computers to understand attention in the elderly. American

Behavioral Scientist, 31(5), 588- 594. Sauro, M. D., Jorgensen, R. S., and Pedlow, C.
T. (2003). Stress, glucocorticoids and memory: a meta-analyticreview. Stress 6, 235
245.
Schell, K.L. & Cox-Fuenzalida, L.E. (2005) Neuroticism and Quality Control in Health
Services: A Laboratory Simulation. Current Psychology, 24 (4): 231-241.
Scherer, K. R., Schorr, A., & Johnstone, T. (Eds.). (2001). Appraisal processes in
emotion: Theory, methods, research. New York: Oxford University Press.
Schmidt L., Forgeot dArc B., Lafargue G., Galanaud D., Czernecki V., Grabli D.,
Schupbach M., Hartmann A., Levy R., Dubois B., Pessiglione M. Disconnecting force
from money: effects of basal ganglia damage on incentive motivation. Brain (2008),
131, 1303 -1310. http://brain.oxfordjournals.org/content/131/5/1303.full.pdf+html
Schoefer, Klaus and Diamantopoulos, Adamantios (2008), Measuring Experienced

Emotions During Service Recovery Encounters: Construction and Assessment of the
ESRE Scale, Service Business, vol. 2, no. 1
Schreier H.. "Munchausen by Proxy Defined." Pediatrics 110, no. 5 (November 2002),
985.
Schrepp, M. (2006). On the efficiency of keyboard navigation in Web sites. Univ

Access Inf Soc 5: 180188.
Seeman, T.E., McEwen, B.S., Singer, B.H., Albert, M.S. and Rowe, J.W., 1997.
Increase in urinary cortisol excretion and memory declines: MacArthur studies of
successful aging. J. Clin. Endocrinol. Metab. 82, pp.
Selye, H. (1956). The stress of life. New York: McGraw-Hill Book Co.
Selye, H. (1976). Stress in health and disease. Boston: Butterworth.
Serna, A., Pigot, H., & Rialle, V. (2007). Modeling the progression of Alzheimer's
disease for cognitive assistance in smart homes. User Modeling and User-Adapted
Interaction, 17, 415-438
Shalev, A., Peri, T., Canetti, L., & Schreiber, S. (1996). Predictors of PTSD in injured
survivors of trauma: A prospective study. American Journal of Psychiatry, 153, 219
225.
Silverman, B.G. (2001) "More Realistic Human Behavior Models for Agents in Virtual
Worlds: Emotion, Stress, and Value Ontologies" Center for Human Modeling and
Simulation. Available at: http://works.bepress.com/barry_silverman/16

Simpson, Joseph R. (2000): Emotion-induced changes in human medial prefrontal

cortex: II. During anticipatory anxiety, PNAS, vol. 98, no. 2, jan.
Sloman, Aaron. 1981. Why Robots Will Have Emotions. University of Sussex. In press.
Smeets, T., Otgaar, H., Candel, I., and Wolf, O. T. (2008). True of false? Memory is
differentially affected by stress-induced cortisol elevations and sympathetic activity at
consolidation and retrieval. Psychoneuroendocrinology 33, 13781386.
Smith, C. A., & Scott, H. S. (1997). A componential approach to the meaning of facial
expressions. In J. A. Russell & J. M. Fernandez-Dols (Eds.), The psychology of facial
expression (pp. 229, 254). New York: Cambridge University Press.
Starkstein S.E., Berthier M.L., Leiguarda R. Psychic akinesia following bilateral pallidal
lesions. Int J Psychiatry Med 1989;19:155-164.
Stephanidis, C., et al. (1998). Adaptable and Adaptive User Interfaces for Disabled
Users in the AVANTI Project, Intelligence in Services and Networks, LNCS-1430,
Springer-Verlag 1998, 153-166
Steven R. Hursh, 2003, Fatigue and Alertness Management using the SAFTE model
and FAST
Stoudemire A. (1995) Psychological Factors Affecting Medical Conditions, American

Psychiatric Press inc.
Sullivan S, Ruffman T. Emotion recognition deficits in the elderly. Int J Neurosci. 2004
Mar; 114(3):403-32. University of Sussex, Brighton, United Kingdom.
Tambe, M., Johnson, W.L., Jones, R.M., Koss, F. Laird, J.E., Rosenbloom, P.S. &
Schwamb, K. (1995). Intelligent Agents for Interactive Simulation Environments, AI
Magazine, 15-37.
Teichner, W.H. (1974). The detection of a simple visual signal as a function of time on
watch. Human Factors 16, 339-353.
Temple, L. & Gavillet, M. (1990). The development of computer confidence in seniors:

An assessment of changes in computer anxiety and computer literacy. Activities,
Adaption & Aging, 14(3), 63-65
Thackray, R. I., Bailey, J. P., & Touchstone, R. M. (1977). Physiological, subjective,

and performance correlates of reported boredom and monotony while performing a
simulated radar control task. In R. R. Mackie (Ed.), Vigilance (pp. 203-215). New York:
Plenum
Thase & Howland (1995) Biologycal processes in depression: An updated review and
integration. In E. E. Bckham & W .E. Leber (Eds) Handbook of depression.
The American Institute of Stress, Job Stress Questionnaire. Web site. URL:
http://www.stress.org/topic-workplace.htm (accessed November 5, 2010).

Thimbleby, H., Cairns, P. & Jones, M. (2001). Usability Analysis with Markov Models.
ACM Transactions on Computer-Human Interaction, Vol. 8, No. 2, pp. 99132.
Thomas P, Hazif-Thomas C, Clment JP. 2008. Loss of motivation in the elderly.

Service de psychogriatrie, CHU Esquirol, 87025 Limoges Cedex, France
Thorne et al, 1985, The Water Reed Performance Assessment Battery,

Neurohehavioural Toxicology and Teratology
Tollinger I. et. al. (2005). Supporting Efficient Development of Cognitive Models At

Multiple Skill Levels: Exploring Recent Advances In Constraint-Based Modeling, In
Proceedings of the ACM/SIGCHI Conference on Human Factors in Computing
Systems, Portland, Oregon, USA, 2005, pp. 411 420.
Tomkins, S. S. (1962). Affect, imagery, consciousness: Vol. 1. The positive affects.

New York: Springer.
Tomkins, S. S. (1963). Affect, imagery, consciousness. Vol. 2: The negative affects.

New York:
Tomkins, S. S. (1984). Affect theory. In K. R. Scherer & P. Ekman (Eds.), Approaches

to emotion (pp. 163-195). Hillsdale, NJ: Erlbaum
Tonn-Eichstdt, H. (2006). Measuring Website Usability for Visually Impaired People -

A Modified GOMS Analysis. Proc. ASSETS06, pp. 55-62.
Treistman, J. and Gregg, J. P. (1979), Visual, Verbal, and Sales Response to Print
Ads, Journal of Advertising Research, 19
Trewin, S. & Pain, H. (1999). Keyboard and mouse errors due to motor disabilities.
International Journal of Human-Computer Studies 50(2), 109-144.
Vallerand, R. J. & O'Connor, B.P. (1989). Canadian Psychology/Psychologie

canadienne. 30(3): 538-550.
Vallerand, R.J., & O'Connor, B.P. (1991). Construction et validation de l'chelle de

Motivation pour les Personnes ges (EMPA). Journal International de Psychologie,
26, 219-240.
van Waveren, J.P. (2001). The Quake III arena bot. Master's thesis, Delft University of
Technology. http://www.kbs.twi.tudelft.nl/docs/MSc/2001/Waveren_Jean-
Paul_van/thesis.pdf
Velsquez, J. (1998a). Modelling emotion-based decision making. In 1998 AAAI Fall

Symposium Emotional and Intelligent: The Tangled Knot of Cognition.
Vincelli, F. (1999). From the imagination to virtual reality: The future of clinical
psychology. CyberPsychology & Behavior, 2(3), 241-248.

Virone, G. (2009). Assessing everyday life behavioral rhythms for the older generation.
Pervasive and Mobile Computing, 5, pp. 606-622.
Volkow N.D., Wang G.J., Newcorn J.H., Kollins S.H., Wigal T.L., Telang F., Fowler
J.S., Goldstein R.Z., Klein N., Logan J., Wong C., Swanson J.M. Motivation deficit in
ADHD is associated with dysfunction of the dopamine reward pathway Molecular
Psychiatry (2010), 18.
VonDras DD, Powless MR, Olson AK, Wheeler D, Snudden AL. (2005) Differential
effects of everyday stress on the episodic memory test performances of young, mid-
life, and older adults. Aging Ment Health. 2005 Jan;9(1):60-70.
Vroman, G., Kellar, L., & Cohen, I. (1989). Cognitive rehabilitation in the elderly: A
computer-based memory training program (pp. 395-415). In Perecman, et al. (Eds.),
Integrating theory and practice in clinical neuropsychology. Hillsdale, NJ: Lawrence
Erlbaum Associates, Inc.
W. S. Reilly, J. Bates, "Building emotional agents", Pittsburg, PA: Carnegie Mellon

University, Technical Report CMU-CS-92-143, 1992.
Wade L. Knez 1,2 and Daniel J. Ham, 2006, A comparison of the effects of fatigue on
subjective and objective assessment of situation awareness in cycling, Journal of
Sports Science and Medicine
Watson, D. and Clark, L.A. (1992), On Traits and Temperament: General and Specific
Factors of Emotional experience and Their Relation to the Five-Factor Model, Journal
of Personality, vol. 6, iss. 2
Watson, D.; Clark, L.A. and Tellegen, A. (1988), Development and validation of brief
measures of positive and negative affect. The PANAS scales, Journal of Personality
and Social Psychology, 76
Watson, David and Tellegen, Auke (1985), Toward a Consensual Structure of Mood
Psychological Bulletin, 98 (sep)
Watson, J. B. (1930). Behaviorism. Chicago: University of Chicago Press.
Wehrle, T.: The Facial Expression Analysis Tool (FEAT) 1992, 1996, University of
Geneva
Weiner, B., & Graham, S. (1984). An attributional approach to emotional development.

In C. E. Izard, J. Kagan, & R. B. Zajonc (Eds.), Emotions, cognition, and behavior (pp.
167-191). New York: Cambridge University Press.
Weisman,S. (1983). Computer games for the frail elderly.Gerontologist, 23(4), 361-363.
Welford, A. T. 1980. Choice reaction time: Basic concepts. In A. T. Welford (Ed.),

Reaction Times. Academic Press, New York, pp. 73-128.

Wickens, C.D., & Hollands, J.G. (2000). Engineering psychology and human
performance. Upper Saddle River, NJ: Prentice Hall.
Wickens, C.D., Gordon, S. E., & Liu, Y. (1998). An introduction to human factors
engineering. Addison-Wesley: New York, NY.
Wikipedia: Behavioral Modeling, http://en.wikipedia.org/wiki/Behavioral_modeling
Wikipedia: Cognitive Architecture http://en.wikipedia.org/wiki/Cognitive_architecture
Wikipedia: Fitts Law http://en.wikipedia.org/wiki/Fitts%27s_law
Wikipedia: GOMS http://en.wikipedia.org/wiki/GOMS
Wikipedia: Ontology http://en.wikipedia.org/wiki/Ontology_%28information_science%29
Wikipedia: Psychological model, http://en.wikipedia.org/wiki/Psychological_model
Wilhelm, F.H., Kolodyazhniy, V., Kreibig, S.D., Roth, W.T., Gross, J.J. (2007). Affective
computing: using computational intelligence techniques to classify the
psychophysiological signatures of fearful, sad, and calm affective states. 47th Annual
Meeting of the Society for Psychophysiological Research (SPR), October 17-21, 2007,
Savannah, Georgia. Psychophysiology. 44 (S1), S110.
Williamson, A. (2007). Fatigue and coping with driver distraction. In: I.J. Faulks, M.
Regan, M. Stevenson, J. Brown, A. Porter & J.D. Irwin (Eds.). Distracted driving.
Sydney, NSW: Australasian College of Road Safety, 611-622.
Wish, M., Deutsch, M. & Kaplan, S. (1976). Perceived Dimensions of Interpersonal

Relations. Journal of Personality and Social Psychology. Vol. 33, pp. 409-420.
Wolf, O. T. (2003). HPA axis and memory. Best Pract. Res. Clin. Endocrinol. Metab.
17, 287299.
Wolf, O. T. (2006). Effects of stress hormones on the structure and function of the
human brain. Expert Rev. Endocrinol. Metab. 1, 623632.
Wolff Harold G. Emotions and gastric function, 1943
Woodworth, R. S. and H. Schlosberg. 1954. Experimental Psychology. Henry Holt,

New York.
Wright, R.A. & Kirby, L.D. (2001). Effort determination of cardiovascular response: an
integrative analysis with applications in social psychology. In: Zanna, M. (Ed.),
Advances in Experimental Social Psychology, vol. 33. San Diego, CA: Academic
Press, pp. 255307.
Wukmir, V.J., (1967): Emocin y Sufrimiento. Barcelona: Labor.
Wyler, R. A., Masuda, M., & Holmes, T. H. (1968). Seriousness of illness rating scale.
Journal of Psychosomatic Research, 11, 363374.

Yang, G., Lin, Y., & Bhattacharya, P. (2010). A driver fatigue recognition model based
on information fusion and dynamic Bayesian network. Information Sciences 180: 1942
1954.
Yarnold, P., Stewart, M., Stille, F., & Martin, G. (1996). Assessing functional status of
elderly adults via microcomputer. Perceptual & Motor Skills, 82(2), 689-690.
Yoon, Carolyn; Gutchess, Angela H.; Feinberg, Fred and Polk Thad A. (2006), A
Functional Magnetic Resonance Imaging Study of Neural Dissociations between Brand
and Person Judgement, Journal of Consumer Research, Vol. 33, June.
Yorkston, KM., Johnson, K., Klasner, ER., Amtmann, D., Kuehn, CM., & Dudgeon, B,
(2003) Getting the work done: a qualititative study of individuals with multiple sclerosis.
Disability and rehabilitation, 25, 369-379.
Young, R.M., Green, T.R.G., & Simon, T. (1989). Programmable user models for
predictive evaluation of interface designs. ACM SIGCHI Bulletin, Volume 20 , Issue SI,
pp. 15 19.
Zadeh, L.A. (1965). Fuzzy sets. Information and Control 8 (3). pp. 338353

ANNEX A: ACT-R Parameters

List of available parameters in the ACT-R framework to customize a modules
behaviour.
Module Parameter Description
General Sgp sgp is used to set or get the value of the

parameters from the modules of the
current model.
General get-parameter-default- The get-parameter-default-value

value command is used to get the default value
that a parameter was given when it was
defined.
Random randomize-time This parameter can be set to t, nil, or an

integer. It is used by the randomize-time
command to determine the range in
which a specified time will be
randomized. The randomize-time
command is used mainly by the
perceptual and motor modules to add
noise to the action times
Central :er This is the Enable Randomness

Parameters parameter. It specifies how
deterministically modules should operate.
It can be set to t which means act non-
deterministically or nil which means act
deterministically.
Procedural :vpft The Variable Production Firing Time

parameter controls whether the time of a
productions firing is constant or variable.
If the parameter is nil (the default value)
then each production takes its specified
action time exactly each time it fires. If
:vpft is t, then the randomize-time
command is used to randomize the
productions action time.
Imaginal :imaginal-delay The imaginal-delay parameter controls

how long it takes a request or
modification request to the imaginal
buffer to complete. It can be set to a

non-negative time (in seconds) and

defaults to .2
Imaginal :vidt The variable imaginal delay time

parameter controls whether the actions of
the imaginal buffer take exactly the
amount of time specified by :imaginal-
delay or if they are randomized with the
randomize-time command. If it is set to t,
then randomize-time is used to variablize
the delay time, and if it is set to nil, which
is the default, then the delay times are
fixed at the :imaginal-delay time.
Model Model-output The default is to send the model output to

Output the Lisp stream *standard-output*, but
there is a parameter available which
allows one to send the output elsewhere
or disable it.
Declarative :ans The activation noise s parameter

specifies the s value used to generate the
instantaneous noise added to the
activation equation if it is set to a positive
number. If it is set to nil, which is the
default, then no instantaneous noise is
added to the activations
Declarative :le The latency exponent value, f, in the

equation for retrieval times. It can be set
to any non-negative value and defaults to
1.0.
Declarative :lf The latency factor value, F, in the

equation for retrieval times. It can be set
to any non-negative value and defaults to
1.0.
Device :needs-mouse This parameter controls whether or not

the model will take control of the
computers mouse cursor when
interacting with a real interface. If it is set
to t, then the system will attempt to keep
the mouse pointer located where the
model has placed it within the current
device.

Device :viewing-distance This is the assumed distance between

the models eyes and the display in
inches. It is used in determining the size
of items in terms of degrees of visual
angle. The default is 15 and it can be set
to any positive number.
Vision :visual-attention-latency This parameter specifies how long a

visual attention shift will take in seconds.
The default value is .085.
Vision :visual-movement- This parameter controls how far an object

tolerance can move and still being considered the
same object by the vision module. It is
measured in degrees of visual angle and
defaults to 0.5.
Vision :visual-onset-span This parameter specifies how long an

item recently added to the visicon will be
marked as new and how long a scene
change notice will be available. It is
measured in seconds, and the default
value is 0.5.
Audio :digit-detect-delay This parameter controls the content delay

time given to digit sounds created with
the new-digit-sound command. It is
measured in seconds. It can be set to
any non-negative value and the default is
0.3.
Audio :digit-duration This parameter controls the duration

given to digit sounds created with the
new-digit-sound command in seconds. It
can be set to any non-negative number
and the default is 0.6.
Audio :digit-recode-delay This parameter controls the recode time

given to digit sounds created with the
new-digit-sound command in seconds. It
is also the time that it takes for a failure to
encode to occur when an audio event is
no longer available. It can be set to any
non-negative number and the default is
0.5.

Audio :sound-decay-time This parameter controls how long sound

events will stay in the audicon measured
in seconds. It can be set to any non-
negative number and the default is 3.0.
Audio :tone-detect-delay This parameter controls the content delay

time given to tone sounds created with
the new-tone-sound command measured
in seconds. It can be set to any non-
Audio :tone-recode-delay This parameter controls the recode time

given to tone sounds created with the
new-tone-sound command measured in
seconds. It can be set to any non-
Motor :cursor-noise If this is set to nil (which is the default)

then the mouse movements by the model
will be made to the exact pixel location
requested. If it is set to t then there will
be noise added to the location to which
the cursor is moved.
Motor :motor-burst-time This is the minimum time required for the

execution of any motor module
movement in seconds. the default is .05.
Motor :motor-feature-prep-time The time in seconds that it takes to

prepare each movement feature. The
default is .05.
Motor :min-fitts-time The minimum movement time in seconds

required to perform an aimed movement
(one for which the Fittss law timing is
applied). The default value is 0.1.
Motor :motor-initiation-time This is the length of the initiation time for

motor actions in seconds. The default is
.05.
Speech :syllable-rate This parameter controls the time it takes

the model to articulate each syllable in a
text string and is measured in seconds
(see the get-articulation-time below for
how it applies). It can be set to any non-

negative number and defaults to 0.15.
Speech :char-per-syllable This parameter controls how the model

breaks a text string into syllables and is
measured in a number of characters (see
the get-articulation-time below for how it
applies). It can be set to any positive
number and defaults to 3.
Speech :subvocalize-detect-delay This parameter sets the detect delay

timing for the sound events generated by
a subvocalize action (see the audio
module for details on the detect delay). It
is measured in seconds and can be set to
any non-negative value. It defaults to .3
(which is the same value that normal text
has).
Temporal :time-master-start- This parameter controls the length of the

increment 0th tick in seconds. It can be set to any
positive number and defaults to 0.011.
Temporal :time-mult This parameter sets the multiplier for

increasing the tick length. It can be set to
any positive number and defaults to 1.1.
Temporal :time-noise This parameter scales the s value of the

noise added to the tick lengths. It can be
set to any positive number and defaults
to 0.015.
Table 26: List of available parameters in the ACT-R framework

Veritas D1.5.1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Veritas D1.5.1

Uploaded by

Copyright:

Available Formats

Accessible and Assistive ICT

Abstract Behavioural and Psychological

Deliverable No. D1.5.1

SubProject No. SP1 SubProject Title User modelling

Dissemination level Pu (Public)

Project start date and 01 January 2010, 48 Months

Version History table

2 15th December 2010. Chapter 3 final. (ITACA)

3 21st December 2010. Chapter 4 first contributions ITACA, I+

4 5th of January 2011. Chapter 5 contributions ITACA,I+. (Tables and ontology

8 15th January 2011. Review of Conclusions (Chapter 6) by all partners involved.

9 Final formatting of the deliverable.(ITACA)

December 2010 iii ITACA

December 2010 iv ITACA

4.2.5 Assessment of fatigue and human fatigue effect............................................... 103

December 2010 v ITACA

December 2010 vi ITACA

December 2010 vii ITACA

Motor (ARPTT) Azimuth Deviations. Adapted from Russo, Kendall, Johnson

December 2010 viii ITACA

Figure 66: Cause hierarchy ............................................................................ 265

December 2010 ix ITACA

December 2010 x ITACA

December 2010 xi ITACA

December 2010 xii ITACA

Chapter 3 describes in depth what cognitive and behavioral integrated architectures

Chapter 4 describes the psychological facets selected in VERITAS as cognitive and

December 2010 1 ITACA

December 2010 2 ITACA

An initial analysis of current theoretical and computational psychological models

December 2010 3 ITACA

generated. However, this response is never automatic; it is not directly conditioned by

In VERITAS, we should focus on how different psychological facets affect VERITAS

An Abstract User Model is a descriptive representation of the different dimensions if a

December 2010 4 ITACA

2 State of Art in Behavioural and psychological

According to Dinerstein et al. (2004), a behavioural model is an executable model

On the other hand, a psychological model is considered to be a synonym of a cognitive

2.1 B&P modeling current computational approaches

December 2010 5 ITACA

The GOMS family models (http://en.wikipedia.org/wiki/GOMS) are used for modelling

Galata et al. (1999) presented an approach for automatically acquiring stochastic

December 2010 6 ITACA

December 2010 7 ITACA

Barry G. Silverman proposes human behavior models for Agents in Simulators

December 2010 8 ITACA

Fitts' law (Fitts, 1954) is a model of human movement in human-computer interaction

December 2010 9 ITACA

The ANN-based approach (e.g. as in Dinerstein et al., 2004; Chan et al.,1995) is

2.1.2 Description of existing computational models for each targeted

December 2010 10 ITACA

Mihailidis et al. (2003) developed a system for an intelligent environment to support

December 2010 11 ITACA

to simulate characters personalities, moods and relationships. The outputs of these

The approaches used in modelling intelligent characters in 3D games can be also

Personal Healthcare & Well-being: Psychological and neurophysiological studies

December 2010 12 ITACA

2.2 Analysis of the P&B computational models

Strengths of a model approach are:

it is an explicit model (descriptive)

applies for VERITAS users and for VERITAS domains

focused on non-skilled behavior

cost and efforts-effective