Proceedings of the CHI 2013 Workshop

on Designing and Evaluating

Sociability in Online Video Games
Georgios Christou, Effie Lai-Chong Law, David Geerts, Lennart E. Nacke, Panayiotis Zaphiris






























Publisher: Self Published
Proceedings of the CHI 2013 Workshop on Designing and Evaluating Sociability in Online Video Games
Georgios Christou, Effie Lai-Chong Law, David Geerts, Lennart Nacke, Panayiotis Zaphiris
XXX Pages. 2013
ISBN: 978-9963-2904-0-6
2013 for the individual papers by the papers authors. Copying permitted for academic and private purposes.
Re-publication of material on this page requires permission by the copyright owners.
Acknowledgments
First of all, we are grateful to the CHI 2013 Workshop Organizers Ido Guy and Nadir Weibel for all
their help throughout the organization of this workshop. Thanks must also go to the authors of the
workshop papers, whose contributions serve as rich sources of stimulation and inspiration to explore
the issues of interest from multiple perspectives. We are also grateful to Tasos Anastasiades who has
graciously designed the covers drawing without compensation.
DESVIG 2013 Table of Contents
Table of Contents
Guiding Principles for Designing a Social, Game-based and Enculturated Reading
Application for Omani Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Fatma Al Aamri, Stefan Greuter and Steen Walz
Player proling and oender classication from player complaints in online social games . . 7
Koray Balci and Albert Ali Salah
Towards Operationalizing Sociability in Online Video Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Georgios Christou and David Geerts
Being social in online games: Five research areas to consider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Mia Consalvo
Revival Actions in a Shooter Game. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Sauvik Das, Thomas Zimmermann, Nachiappan Nagappan, Bruce Phillips and Chuck
Harrison
Limelight Fostering Sociability in a Co-located Game . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Wolfgang Hochleitner, Michael Lankes, Jeremiah Diephuis and Christina Hochleitner
Sociability in Virtual Citizen Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Charlene Jennett, Laure Kloetzer, Margaret Gold and Anna L. Cox
Tribal Metaphors in Social Game Design; Creating Conict and Camaraderie through
Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Ben Kirman
Social Player Analytics in a Facebook Health Game . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Matthias Klauser, Lennart Nacke and Paul Prescod
Towards a Tripartite Theoretical Framework for the Sociality of Digital Games:
Psycho-Social-Biological . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Ee Lai-Chong Law and Mark Springett
The Relationship Between Game Rules And Team Cohesion: An Empirical Study . . . . . . . . 52
Eleanor Martin and Judith Good
Sociability in Social Network Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Janne Paavilainen, Hannu Korhonen, Kati Alha, Jaakko Stenros and Frans Mayra
World of Warcraft Machinima: Para-Game Communities as Cultures of Participation . . . . 62
Tyler Pace and Shad Gross
All Work No Play: Contextualizing the Missing Element in Exergames . . . . . . . . . . . . . . . . . . . 68
Tamara Peyton
Democracy has arrived! Promoting a reexive Approach to Power in Virtual Worlds
through Interface Modications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Patrick Prax
Encouraging Co-located Players Social Interaction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Soa Reis and Nuno Correia
1
DESVIG 2013 Table of Contents
Designing Games That Foster Equity and Inclusion: Encouraging Equitable Social
Experiences Across Gender and Ethnicity in Online Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Gabriela Richard
Supporting social interaction for older users in game-like 3D virtual worlds . . . . . . . . . . . . . . . 89
Panote Siriaraya, Panayiotis Zaphiris and Chee Siang Ang
2
Guiding Principles for Designing a Social, Game-based and
Enculturated Reading Application for Omani Children

Fatma Al Aamri, Stefan Greuter, Steffen P Walz
GEElab, School of Media and Communication
RMIT University
Melbourne, Australia
fatma.alaamri@rmit.edu.au, stefan.greuter@rmit.edu.au,
steffen.walz@rmit.edu.au

ABSTRACT
Reading for pleasure is not only important for literacy
attainment, but it also leads to a greater insight into human
nature. Particularly children seem to benefit from reading
for pleasure. Omani and Arab children, however, read
remarkably less than those in western societies. This paper
reports early stage research that is used as a guide to
develop an interactive reading application called (Trees of
Tales). The application is intended to motivate Omani
children in reading stories for pleasure. It also aims to
engage children in further activities by providing a
mechanism to create traditional folk tales jointly with their
friends and families. This paper describes possible reasons
that have contributed to the decline of reading among
Omani children. It also explores the literature around the
dimensions of reading motivation that are used as
guidelines for the design of the reading application. The
paper concludes with a reflection of the challenges and
constraints of applying such an approach for the Omani
society.
Author Keywords
Omani children; reading for pleasure; interactive reading
application; reading motivation
INTRODUCTION
Reading for pleasure increases general knowledge,
decision-making skills, community participation,
understanding of other cultures, and provides greater insight
into human nature [7]. At the same time, children have to
be motivated to engage in a reading activity [19].
Research indicates a low rate of literacy and a huge decline
in reading and publishing in Arab countries compared to
western societies. In J anuary of 2012, the Arab Thought
Foundation, Fikr reported that the Arab child reads, on
average, six minutes annually, compared to the
approximately 12,000 minutes a western child is engaged in
reading activities per year [3]. Many factors contribute to
the decline of reading within Arab society. Reasons quoted
for this decline in the Kingdom of Morocco for example
include (a) the weakness of family support for and
encouragement of reading, (b) financial issues placing
constraints upon the purchase of books, and (c) the spread
of illiteracy within the community [2].
However, not all of the reasons that were identified for the
Kingdom of Morocco seem to apply to all other Arab
countries. The Sultanate of Oman for example is listed as a
high-income country and within the best performing 35
countries in the world [25]. The lack of financial ability is
therefore, not the reason for the decline in reading in Oman.
Illiteracy is also not a problem in Oman. Literacy rates
increased greatly in Oman in recent years. According to the
UNESCO statistics, 86% of adults and 97.6% of youth in
Oman were literate by 2008 [26].
The following section identifies reasons that are commonly
associated with the decline of reading for pleasure in Oman.
REASONS FOR THE DECLINE IN READING IN OMAN
Reason I: Famil y
Based on my own experience as a citizen of the Sultanate of
Oman, there is a lack family support and encouragement of
reading for pleasure activities. Without the support of the
family it can be difficult for children to engage in a reading
activity for pleasure if the parents or their peers have not
modeled this behavior. It has been shown that children who
are encouraged to read by their parents are more likely to
enjoy reading and read more frequently than those who do
not receive any encouragement from their parents [6]. This
in turn positively influences a childs interest in reading,
attitude towards reading, and level of concentration in the
classroom [23]. In addition, to the active encouragement by
a family member, children are also able to learn from
passive behaviour, such as seeing parents read at home
Reason II: Access
Access to reading material is essential to motivate people to
read. Access to age appropriate material improves the
reading performance of children and encourages them to
read more and for an increased duration [18]. In western
countries, children are able to access reading materials
through a variety of sources and spaces including public
libraries, school libraries, friends, websites, bookstore and
at home. While Omani family members have access to
religious books such as the Holy Quran at home, there are
usually not many books that are attractive to children. There
are also not many public libraries in Oman, and the few that
exist only carry a limited number of books that are suitable
Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.

DESVIG 2013
1
for children [1]. The books available in school libraries are
often not printed in color and contain proportionally more
text than images, making them less attractive for children.
The books are often not sturdy enough to withstand the
sometimes rough handling by children and are therefore
often torn and pages are missing. Swapping books between
friends is also not common in Oman, since children do not
own or have access to books themselves.
There are only a few bookstores in Oman. The popular
Family Book Store in Oman offers more English reading
materials for foreign customers and only a few very
expensive high quality Arabic books for children.
Although, Omani families can afford expensive books, the
parents are often not aware of their educational value for
their children and therefore spend the money on other
things.
Children in Oman love to play computer games like so
many children in western countries. However, the internet
does not provide many interesting Arabic reading materials,
in comparison to the wealth of English material that is
available online. For example the Disney online books
website and the ABC reading eggs website provide fun
reading experiences for children who can read in English
language whereas there is nothing similar for children who
can read only Arabic language.
Reason III: Reading Space
Oman lacks reading spaces that are quiet but also motivate
children to select a book and read for pleasure. School
libraries are the only reading spaces in Oman that are built
to encourage children to read. Each public school in Oman
educates between 1500 and 2000 students. Despite the size,
however, the school library is only the size of one
classroom. The interior design and furniture of the library is
unimaginative and consists only of basic furniture such as
white rectangular tables and wooden chairs. The
environment is therefore not conducive for children to
spend time with a book.
Recent research indicates that incorporating technology
with traditional books can encourage the reluctant readers
to read more [20]. According to the Dorat Alajial Primary
School librarian, pupils hardly read books in the library and
although they have introduced computers to the libraries,
the school is unable to encourage the children to use the
library space effectively [Mrs. Al Hatali, personal
communication, December 16, 2011]. The computers in
Omani school libraries seem to compete with the library
books. The students play games on the library computers
but the games they paly do not motivate the children to
read.
Reflection
In order to motivate Omani children to read for pleasure,
the children require access to age appropriate texts in
Arabic language, family encouragement, and a space that is
conducive to reading. Family encouragement and student
motivation requires a culture change that may be kick
started with a reading application that provides a fun
reading experience for the whole family.
In order to better understand the requirements for such a
reading application it is important to understand the
dimensions of reading motivation that engage children in
reading for pleasure. The following section explores these
dimensions and provides a guideline for an enjoyable,
motivating, and sociable reading experience for Omani
children.
THE DIMENSIONS OF READING MOTIVATION
Motivation is essential to engage even the brightest children
in reading or other learning activities [19]. Reading
motivation is a multi-layered construct that includes aspects
such as the persons reading goals, intrinsic and extrinsic
motivation, self-efficiency, and other social motivations for
reading [15]. Interestingly, there is evidence that intrinsic
and extrinsic motivation relate differently to literacy
outcomes. It has been found that intrinsic motivation
predicts an increase in amount of reading for pleasure more
than extrinsic motivation [8, 27]. The intrinsic motivation
of reading has been described as the curiosity about reading
and a preference for challenge in reading [15]. From seven
different dimensions of reading motivation only two
dimensions curiosity and involvement have been
identified as components of intrinsic reading motivation
[24]. Curiosity usually motivates people to learn more
about personally interesting topics. Involvement means to
get immersed in a story, feel the actions, and empathize
with the characters of a story [24].
In the following, both curiosity and involvement will be
addressed as important dimensions to motivate Omani
children to read for pleasure.
Dimension I: Curiosity
To motivate a child to read, the provided material has to be
of genuine interest. Interest in the reading materials even
difficult material does not only motivate reading
frequency but it also improves comprehension of the
reading material [22]. Interest in reading material can be
affected by the childs culture and background [30]. When
students are provided with culturally relevant text, they
often become more engaged with the reading [12].
Providing children with a variety of culturally relevant
stories might increase the chances that children find reading
more interesting and they become curious about the subject
matter, and extend their engagement with the material,
which motivates them to read more. It is worth to mention
here that Oman like many other Arabic countries is rich of
its oral storytelling tradition that has been transformed
through generations. This tradition still exists in many
Omani houses as the grandparents still live with their
children and grandchildren. It is common to see children
surround their grandmothers after dinner to listen to a story
of their favorite characters such as J oha or Sindbad.
DESVIG 2013
2
Dimension II: Involvement
Involvement that causes the child to get immersed in a story
is an intrinsic dimension that motivates children to read for
pleasure. Flow and engagement are often used terms that
describe the state of deep involvement in an experience.
Csikszentmihalyi explains that when a person is in the state
of Flow, the concentration is so intense that there is no
attention left for other things and the self-consciousness
disappears. Engagement has been identified as a subset of
flow. In other words, engagement can be described as flow
in a more passive state, and flow without user control
[28]. Csikszentmihalyi has given the term optimal
experience or Flow to describe intense engagement or
complete absorption in a task [9]. This makes the concepts
of Flow and Engagement difficult to distinguish. However,
OBrien and Toms [21] identified attributes that are shared
between Flow and Engagement. These attributes include:
focused attention, feedback, control, activity orientation,
and intrinsic motivation. Moreover, OBrien and Toms
defined engagement as a quality of user experiences with
technology that is characterized by challenge, aesthetic and
sensory appeal, feedback, novelty, interactivity, perceived
control and time, awareness, motivation, interest, and
affect [21].
A SOCIABLE READING APPLICATION FOR OMANI
CHILDREN
Computer Games
Computer games have been designed, developed and
evaluated to support learning, politics, and social activities.
Gee [14] has done extensive research on the use of gaming
technology in schools and have noted many benefits such as
increased student motivation, rising test scores, increased
critical thinking, pattern identification and problem solving
skills, semantic and syntactic clue use, and negotiation
techniques. Bogost [4] suggested that games are strong in
motivating people through their rule-based representations
and interactions when other media are limited to the use of
words, text, images or animation.
Some computer games have social aspects that allow
players to play, interact, and compete with other players.
These games are typically online games part of social
networks such as Facebook. Games such as CityVille,
FarmVille, and SecondLife are played by millions around
the world and have been successful mainly because of their
social features. A study conducted by PopCap, the creator
of popular social games such as Bejeweled and
Insaniquarium, on social game players, found that fun and
excitement was the most popular motivation for playing
social games [13]. Therefore, elements that make family
members and friends work with each other or against each
other may be a useful mechanic for the design of the
reading application Trees of Tales.
An Omani Game-based Reading Application
Several aspects need to be considered in the creation of an
enjoyable reading experience for children in Oman. The
experience should include a careful selection of interesting
topics that are related to the Omani culture. That includes
choosing characters from the Omani folk tales such as J oha
and Sindbad to motivate children to read. Also, challenge,
feedback, and control are elements of game design and
components that strengthen the engagement in this
experience. Additionally, building interactivity that
involves the readers with the story and cause them to
experience flow could enhance enjoyment. All of that is
significant to insure a quality of user experience that
motivates children to come back and read more.
The investigation on the issues that led to the decline of
reading for pleasure activities in Oman and the research on
the intrinsic dimensions of reading motivation indicates that
an fun reading experience for the whole family provided in
the form of an interactive, social reading application may
help to kick start a culture change to motivate Omani
children to increase the amount of time spent on reading for
pleasure activities. Game mechanics will give the
application a playful experience to engage the children with
the reading. It will include stories based on famous Arabic
characters that children are already familiar with. Finally,
including features such as involving family members,
sharing stories with friends, and liking each others stories
will contribute to a more positive attitude towards reading
for pleasure.
The following section discusses the guiding principles for
the design of the game based reading.
Social Media Mechanics in the Application
Social media provides information, influences behavior and
improves the reading skills of its users. Hsu and Lu [16]
found from their investigation that social influences such as
social norms affected the attitudes and intentions of
participants to play on-line games directly. There are two
types of social norms identified in literature [11]. The first
type is the informational influence, which occurs when
users accept information from other users as evidence of
true and valuable information. This is why family
participation in the application is important. Children value
their familys stories and would be interested to share them
with their friends.
The second type is the normative influence, which occurs
when a person obtains rewards or avoids punishment by
corresponding to the expectations of others. When children
see how many likes their friends get, they will try to
improve their stories to get more likes and better comments.
This approach is common in many childrens online
communities such as scratch.mit.edu and
disney.go.com/create. Involving families and friends will
create a social influence on players in addition to having
new reading materials continuously posted. It is a loop of
continuous reading; creating, posting, and then liking each
others stories which is part of the social reward system.
DESVIG 2013
3
Although the application does not teach children reading
skills, it improves their reading abilities by practice. It also
adds to the children knowledge through its activities and the
abilities to share the stories with others. Vygotskys
sociocultural theory argues that childrens development
depends on interaction with people and the tools that the
culture provides to help form their own views of the world.
Based on the theory, social interaction leads to continuous
changes in childrens thought and behaviour that can vary
greatly from culture to culture [29]. Vygotskys theory
provides evidence on the importance of implementing ways
for social interaction of readings in the application.
Childrens interaction with family members to create new
stories can inspire and teach them new ways of storytelling.
The application will encourage children in the same class to
discuss their readings and stories with others and motivate
those who do not use the application. All of these reasons
make it necessary to implement features that support the
sociability of the reading experience through the Trees of
Tales application.
Guiding Principles
Table 1 illustrates the guidelines that were developed based
on the presented information. The guidelines provide a
starting point for the project and will be further refined as
the project continues.
Feature Main Purpose
Online-Game-
based
Allows access and flexibility for the
reading space and platform
Interactivity Provides engaging experience that
allows the child to modify the story and
adjust the scenery to create new stories
Family
participation
Provides a family support for a reading
activity which influences the childrens
attitude towards reading
Sharing stories Provide access to more reading material
Liking stories Increases children curiosity and motivate
them to read what most friends like
Commenting
on stories
Allows children to know what other
friends think of their stories and motivate
them to create better stories
Rewards Increase motivation to continue reading
Table 1: Guidelines of Features to include in the Reading
Application for Omani children
Discussion: Benefits of Adding Social Media Mechanics
The benefits of adding sociability to the planned application
are as follows:
Utilises motivational drivers such as interest,
social relationships, and personal advancement,
which are intended to change peoples attitude
towards reading.
Allows a child to see how friends/classmates are
progressing, which can inspire them to read more.
Encourages children to read more stories, which
are posted by their friends.
Children will develop an identity as participants in
a community of readers.
The display of the leaderboard will encourage the
competitive element in children so that they aim to
achieve more than others in reading.
Induces children to explore and search their school
library for more inspiring stories that they can add
to their collections.
Application Design
The following outline shows a conceptualisation of the key
elements of the interactive reading application:


Figure 1: The Framework of the Reading Application
The player starts by filling out some basic information to
register in the application. Upon starting, the child needs to
choose a character that he/she would like to read about.
Each character will have several original stories and an
open story that the child can manipulate to create his or her
unique story. The child will be provided with tools for
DESVIG 2013
4
drawing and dragging elements to build the scene of each
page according to the customised text. The open stories
motivate children to work with family members to create
new stories about the selected character. Children can take
images of parents or siblings faces and replace them with
the characters faces in the stories. This will personalise the
stories and make them more interesting as they related to
the childs own family. The created stories can be published
and shared with other members in the application. The
social features of the reading application will involve
sharing, reading, and commenting on each others created
stories. The player can see what other players have
published and mark them as liked. He or she can also share
the link of what they read through Facebook or Twitter.
This will increase the value of the activity and will raise the
public awareness of the application. Figure 1 shows the
choices available in the reading application.
CONCLUSION
We believe that by designing a rewarding, social, and
enjoyable reading application we will be able to improve
the Omani societys attitude towards reading for pleasure.
The applications social elements could be key motivational
drivers, which may serve to encourage reading and the
sharing and discussing of reading material all of which
are currently not part of the Omani culture. The finalised
application will be user tested with Omani school children
and the results will be analysed in order to assess the
effectiveness of the reading application and will allow us to
make recommendations for the development of future
reading applications.
The main challenge of this project is to motivate children
and their families to start using the reading application.
Once the children start reading, sharing, and commenting to
each others stories, we assume that they will become
engaged and will continue to read for pleasure. To achieve
this, we will discuss the importance of the reading
application with Omani parents and ask for their support
and participation in the research project prior to distributing
the application.
The interactive reading application will be designed to
facilitate the integration of additional characters and stories
to ensure renewed interest and more visits to the
application. The reading application will initially start with
three famous Arabic characters but we envision adding
more characters and stories in the future.
Finally, the roll out of the Internet as well as the increased
availability of technology such as the tablet devices has
always been a challenge in the low-income societies such as
the third world countries. However, Oman is not a low-
income country and according to the Information
Technology Authority in Oman, new policies are
implemented to improve Internet speed and provide better
digital services in the country [17]. Additionally, childrens
skills and abilities to use and navigate the device of the
Application can be overcome by demonstrating to children
how to use the application with the testing device prior to
handing them the devices.
Long-term solutions are needed to make long-term changes
in this field, especially when they concern the behaviour of
a whole society. As mentioned earlier, Arab people and
their children are not reading for pleasure. This project is
one step towards a more knowledgeable society in Oman by
engaging children to read for pleasure. It is a small step that
will hopefully make a difference to many Arabic children in
the world.
REFERENCES
1. Al Abri, Husna. (1995). An Investigation into the need
for Childrens Libraries in Oman (Unpublished masters
thesis). Manchester Metropolitan University.
2. Al-Ashraf, H. (2010). The Decline of Reading in
Morocco. Retrieved from
http://www.alarabiya.net/views/2010/10/12/243241.html
website: accessed 15 September 2012
3. Al-Yacoub, I. (2012). Sum of All Fears: Arabs read an
average of 6 pages a year, study reveals. Retrieved from
english.alarabiya.net website: accessed 21 August 2012
4. Bogost, I. (2007). Persuasive Games: the expressive
power of video games. Cambridge, Massachusetts: MIT
Press
5. Clark, C., & De Zoysa, S. (2011). Mapping the
interrelationships of Reading enjoyment, attitudes,
behaviour and attainment. London. National Literacy
Trust, 30.
6. Clark, C., & Hawkins, L. (2010). Young People's
Reading: The Importance of the home environment and
family support. London. National Literacy Trust, 32.
7. Clark, C., & Rumbold, K. (2006). Reading for Pleasure:
A research overview. London. National Literacy Trust,
30.
8. Cox, K.E., & Guthrie, J .T. (2001). Motivational and
Cognitive Contributions to Students' amount of Reading.
Contemporary Educational Psychology, vol. 26, 116-
131.
9. Csikszentmihalyi, M. (1975). Beyond Boredomand
Anxiety: Experiencing Flow in Work and Play. San
Francisco: J ossey-Bass
10. Csikszentmihalyi, M. (1990). Flow: The Psychology of
Optimal Experience. Harper Perennial, London.
11. Deutsch, M., & Gerard, H. B. (1955). A study of
Normative and Informational Social Influences upon
Individual J udgment. The J ournal of Abnormal and
Social Psychology, 51(3), 629-636.
12. Feger, M. V. (2006). I Want To Read: How Culturally
Relevant Texts Increase Student Engagement in
Reading. Multicultural Education, 13(3), 2.
DESVIG 2013
5
13. Fields, T., & Cotton, B. (2011). Social Game Design:
Monetization Methods and Mechanics. New York,
London: Morgan Kaufmann.
14. Gee, J . P. (2008). Learning and Games. [The J ohn D.
and Catherine T. MacArthur Foundation Series on
Digital Media and Learning]. The Ecology of Games:
Connecting Youth, Games, and Learning, 21-40.
doi:10.1162/dmal.9780262693646.021
15. Guthrie, J . T. & Wigfield, A. (2000). Engagement and
motivation in reading. In M.L. Kamil, P.B. Mosenthal,
P. D. Pearson, and R. Barr (Eds.), Handbook of Reading
Research (3rd Ed.). New York: Longman.
16. Hsu, C.-L., & Lu, H.-P. (2004). Why do people play
on-line games? An extended TAM with social
influences and flow experience. Information &
Management, 41(7), 853-868. doi:
http://dx.doi.org/10.1016/j.im.2003.08.014
17. Information Technology Authority (ITA). (2011). The
Fifth Report of the Information Technology Authority.
Retrieved from
http://www.ita.gov.om/ITAPortal_AR/MediaCenter/Doc
ument_detail.aspx?NID=72
18. Lindsay, J . (2010). Children's Access to Print Material
and Education-Related Outcomes: Findings From a
Meta-Analytic Review. Reading is Fundamental
Organisation, 212.
19. Maw, W. H., & Maw, E. W. (1961). Information
Recognition by Children with High and Low Curiosity.
Educational Research Bulletin, 40(8), 197-224. doi:
10.2307/1474761
20. Metsala, J . L., Wigfield, A., & McCann, A. D. (1997).
Children's Motivations for Reading. The Reading
Teacher, 50(4), 3.
21. Moss, B. (2012). New Literacies Books: Technology-
Based Formats for Motivating Tech-Savvy Middle
Graders. Voices Fromthe Middle, 19(4), 76-78.
22. O'Brien, H. L., & Toms, E. G. (2008). What is user
engagement? A conceptual framework for defining user
engagement with technology. J ournal of the American
Society for Information Science and Technology, 59(6),
938-955. doi: 10.1002/asi.20801
23. Renninger, K. A. (1992). Individual interest and
development: Implications for theory and practice. In K.
A. Renninger, S. Hidi, & A. Krapp (Eds.), The role of
interest in learning and development (pp. 361-376).
Hillsdale, NJ : Lawrence Erlbaum Associates
24. Rowe, K. J . (1991). The Influence of Reading Activity
at Home on Students' Attitudes Towards Reading,
Classroom Attentiveness and Reading Achievement: An
application of structural equation modelling. British
J ournal of Educational Psychology, 61(1).
25. Schiefele, U., Schaffner, E., Moller, J ., & Wigfield, A.
(2012). Dimensions of Reading Motivation and Their
Relation to Reading Behavior and Competence. Reading
Research Quarterly, 47(4), 427-463.
26. The World Bank. (2011). World Development
Indicators (WDI). Data by country. Retrieved from
http://data.worldbank.org/country/oman.
27. UNESCO, United Nations Educational, Scientific and
Cultural Organization. (2010). UNESCO Institute for
Statistics. Education profile Oman. Retrieved from
http://stats.uis.unesco.org/unesco/TableViewer/documen
t.aspx?ReportId=121&IF_Language=eng&BR_Country
=5800&BR_Region=40525.
28. Wang, J .H.Y. & Guthrie, J .T. (2004). Modeling the
Effects of Intrinsic Motivation, Extrinsic Motivation,
Amount of Reading, and Past Reading Achievement on
Text Comprehension between U.S. and Chinese
Students. Reading Research Quarterly, 39, 162-186.
29. Webster, J ., & Ahuja, J .S. (2006). Enhancing the design
of Web navigation systems: The influence of user
disorientation on engagement and performance. MIS
Quarterly, 30(3), 661678
30. Woolfolk, Anita E. (1998). Educational Psychology
Seventh Edition. Boston: Allyn and Bacon.
31. Worthy, J ., Moorman, M., & Turner, M. (1999). What
J ohnny Likes to Read Is Hard to Find in School.
Reading Research Quarterly, 34(1), 12-27. doi: 10.1598/
rrq.34.1.2

DESVIG 2013
6
Player Profiling and Offender Classification from Player
Complaints in Online Social Games
Koray Balc
Boazii University, Dept. of
Computer Engineering
stanbul, Turkey
balci@cmpe.boun.edu.tr
Albert Ali Salah
Boazii University, Dept. of
Computer Engineering
stanbul, Turkey
salah@boun.edu.tr

ABSTRACT
Online multiplayer games create new social platforms, with
their own etiquette, social rules of conduct and ways of
expression. What counts as aggressive and abusing
behavior may change depending on the platform, but most
online gaming companies need to deal with aggressive and
abusive players explicitly. This usually is tied to a reporting
mechanism where the offended player reports an offense. In
this paper, we discuss preliminary work on a tool for
analyzing player behavior to automatically classify
offenders. We rely on analysis of chat records and other
social activities in the game as well as factor in player
history in our system. This tool should be used to prioritize
more severe cases of abuse, and to lighten the load of
administrative staff that takes decisions on how to act on
the offense. We report our results on data collected from a
real online game with 360.000 users over a six months
period.
Author Keywords
Social games; gaming concepts; abusive behavior analysis;
cyberbullying; player profiling; social conduct; Bayes Point
Machines
ACM Classification Keywords
K.4.1 Computers and Society: Abuse and crime involving
computers
General Terms
Human Factors; Measurement
INTRODUCTION
Online social games provide great opportunities to analyze
user behaviour. Using analytics, one can record user
actions, assign target interactions, measure in which
configurations these targets are met. Next, with the help of
data, one can modify the flow of the game, fine tune
internal parameters. Using analytics and enhancing the
game mechanics, one can expect to prolong game play,
increase in-game sales, generate a higher number of
returning players or establish a more social and friendly
environment. A recent survey on human behavior analysis
for computer games illustrates that while game designers
analyze player behavior intensively when designing the
game, real time behavior analysis rarely is incorporated [8].
There are companies that adapt their game content to user
preferences by means of A-B testing, where a group of
users receive one version of the game, while a second group
receives a slightly modified version, and the preferences are
recorded to select one of the versions over the other.
Gaming companies that govern online games with many
subscribers also use data analysis tools in monitoring player
activity, for instance to detect cheating behaviors [10], or
for the analysis of player performance in different
dimensions like demographics, archetypes, classes, and
sub-classes [9].
A games socializing mechanisms designed to initiate
interaction between players may enable some players to
abuse, harass or offend others. Establishing a friendly and
welcoming environment and maintaining it is important for
social game companies to retain their players. Therefore
abusive or offending behaviors should be dealt seriously by
moderators of the game in order to keep all players content.
In this paper, we propose using machine learning for player
profiling in an online game of Okey to detect offending
behaviors. The Okey game is played with four players,
seated around a table. In order to start a game, the leading
player sets up a table, and other players join in. Once the
game is played out, the players may disband and join other
tables, or if the table is sufficiently entertaining, they may
continue playing together. Every once in a while, a player
will abuse or offend one or more players, these need to be
dealt with quickly and decisively in order to keep the
ambiance of the game intact. To achieve this, usually
players have means to submit complaints about other
players. However, analyzing these cases and deciding on a
verdict requires thorough investigation of both parties
profiles, game histories and logs. The lack of robust player
profiling tools to help human moderators dealing with the
complaints is a general problem for all social gaming
industry.
In this work, we aim to profile the accused players in
reported complaints to help human moderators. In order to
achieve this, we extract several attributes of players to form
a feature vector that constitutes a player profile. This vector
includes demographic data, game play statistics and non-
verbal cues for social interactions in the game. Next, we Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.
DESVIG 2013
7
perform binary clustering on a labeled set of players and tag
players with computed probabilities of giving offense.
These probabilities can later be used by moderators as
supplementary information when a new complaint is
submitted. Moderator time is a scarce resource, and our
system helps to prioritize the complaint submissions to take
faster action. Our profiling methodology performs with a
small number of false positives, and is now being
incorporated into the actual game environment. Reynolds et
al. have previously proposed to use text mining techniques
for automatically detecting cyberbullying from posts [7].
This work resembles our approach, but relies exclusively on
textual content, whereas we put the stress on historical
factors to determine prior probabilities of exhibiting
abusive behavior.
This paper is organized as follows. First, we give a brief
overview of the game of Okey used in our experiments,
which has a significant importance in Turkish culture. Next,
we explain our methodology to label the accused players.
We present a recently collected gaming database
(containing one month game usage of 360:000 players) and
present the first experimental results in the literature on this
database. We conclude with a short discussion and future
work.
AN ONLINE SOCIAL GAME: OKEY
Like in many countries, traditional Turkish games also have
seen their online counterparts hitting the markets. A very
well-known example of such a Turkish game is Okey,
probably of Chinese origin, but adopted (and adapted) in
Turkey and played socially since a very long time. Okey is
a part of the Kraathane (i.e. the coffeehouse, but the
word comes from reading house) culture in Turkey; the
coffehouses are social gathering places for men, who spend
long hours there to drink tea and coffee, smoke cigarettes
and hookah, read and discuss newspapers, and play card
games, backgammon, and Okey (See Figure 1). These
coffeehouses have existed in Turkey since the sixteenth
century [5].

Figure 1. Okey players in a coffeehouse.


Okey is one of the first social games in Turkey to be carried
on to the digital platform. Since the social aspect is the
primary aspect of the game, online Okey sites have also
offered means of player interaction, typically in form of a
chat window next to the main game window. The precise
mechanism of the game is not relevant for this paper,
suffice it to say that the stones of the game resemble cards,
and players draw and exchange stones to finish the game
first with a matching set. While a simple game, digital
versions of Okey attracts a very large audience as well.
According to independent statistics service AppData, the
top three most popular Okey games on Facebook platform
have a total of 6.500.000 monthly active unique users as of
J anuary 2013 [1].
METHODOLOGY
Our method starts with the processing of player complaints.
First, a training and benchmarking set is generated, where
player complaints are labeled as abusive or offending
manually by human moderators. The information of players
involved in these complaints is extracted from a central
game database. We are currently interested only in non-
verbal communicative cues, such as number of chat
attempts and abusive language usage. The precise contents
of chat messages will be studied as an extension. Each
player is defined with a feature vector.
To be able to decide whether a player falls into the offender
category, we train a supervised binary classifier where the
two classes stand for genuine offender and not an offender.
When a new complaint arrives, our system evaluates the
accused players profile, giving a likelihood of the player to
be a genuine offender. Figure 2 overviews our basic flow of
the methodology.
There are several supervised clustering methods in
literature that can be used for player classification [2]. Our
initial tests have shown that Bayes point machines can deal
with this problem successfully, and this approach is adopted
as our primary classifier. Presently, other classifier schemes
such as decision trees, support vector machines and k-
means clustering are tested for benchmarking purposes.
Bayes Point Machines (BPMs)
BPMs, first proposed by Herbrich et al. [4], are non-linear
kernel classifiers used for binary classification. We omit the
mathematical details of the scheme, and refer the reader to
the original paper. In summary, the BPM essentially
approximates the Bayesian inference for linear classifiers in
the kernel space. The classification is achieved by
determining the loss incurred by a class assignment,
weighting it according to the posterior probability, and
selecting the class label that achieves the minimum
expected loss. This approach improves the accuracy of the
center of mass of the version space when compared to other
kernel classifiers such as Support Vector Machines (SVMs)
[3]. The details of BPMs and their comparison with SVMs
can be found in [4].
DESVIG 2013
8
Evaluation
The BPM classifier outputs the likelihood of a player to be
in the offender group. Since it is a binary classifier, for each
sample, a likelihood value greater than a certain confidence
threshold results in the classification of an offender. In
order to evaluate our results, we count the number of true
positives (TP), false positives (FP), true negatives (TN) and
false negatives (FN) to calculate sensitivity, precision and
specificity. In information retrieval, these terms are defined
as follows:
=

+


=

+


=

+


Finally, we experiment with different values of thresholds
and try to maximize precision and specificity to aid human
moderators better. In general, one will try to set a threshold
to acquire a high precision and specificity, so that human
moderators can prioritize complaints for these players.
EMPIRICAL VALIDATION
CCSoft Okey Player Abuse (COPA) Database
In order to assess our methodology, we used the proprietary
CCSoft Okey Player Abuse (COPA) Database, consisting
of player demographics, statistics, game records, and player
complaints. The database is acquired from a commercial
Okey game over a six months period, and incorporates
roughly 360.000 unique players. This is the first study that
investigates this database. All the player identification
information is deleted prior to our analysis to anonymize
the data and to protect players privacy. In the mentioned
period, a total of 800.000 Okey games are recorded along
with the player interactions in the chat area. Within this
time frame, 1.066 player complaints were manually labeled
as player abuse or bad language by game moderators. These
complaints were reported by 726 unique players and a total
of 689 players were accused. The number of game table
logs generated by the players exceed 520.000 items. For
each player in the database, we process this set of table
logs, filter the ones that involve the related players, and
extract the number of chat and bad language attempts to
form the feature vector for player profiling.
The game incorporates a basic bad language detection
system that blocks a set of black-listed words and their
variants. When a word in the black list is intercepted, the
system blocks it and informs the offending player. Very
often, warned players will rework the offending sentence to
bypass the filter and they will usually manage to submit an
abusive sentence by tweaking the letters and words. In
addition, the filter has been reported to be only moderately
effective, as it generates a large number of false positives
when used with a large black list vocabulary. Therefore,
this filter is used with a small vocabulary and bad language
interception is not counted as direct evidence of actual
abuse.
In addition to chat, players have several other means to
socialize during the game. They can send virtual friendship
invitations to each other and establish virtual friendships,
which enables them to send offline messages to one another
and to locate a particular game they play when both parties
are online. The game also takes advantage of Facebook
social integration tools, in which players can invite their
Facebook friends and establish in-game virtual friendships,
and share their game achievements in their Facebook
profile. In exchange of Facebook interaction, players are
rewarded bonus virtual currency, which can be used for in
game betting. Using a standard double currency system,
players can also pay real money to buy virtual currency.
With virtual currency, players can purchase virtual gifts in
game tables and send them to other players in the game.
These means help people to enjoy their turn waiting time,
increase the engagement and decrease the virtual currency
inflation. One hypothesis we entertain is that players with
high social activities may tend to offend others more, as
Figure 2. The basic flow of the proposed methodology.
DESVIG 2013
9
well as being offended easily. We also argue that paying
players would take the game environment more seriously
and therefore will not offend others easily, but report abuses
and other offensive behaviour. Therefore, the amount of
social interaction and real life payment statistics are also
included in our feature vector.
The Experimental Protocol
In order to cluster players and analyze the complaints
better, we try to incorporate all social signals used in the
game to profile a players characteristics. To achieve this,
our feature vector for the player consists of the gender and
number of several profile aspects, namely, the numbers of
daily logins to the game, games played, game tables
entered, wins, incomplete games, social rewards obtained,
real life payments, virtual gifts sent, number of virtual
friends, number of private messages sent, number of chat
sentences and bad language attempts. We also use ratios of
some of these properties listed above. For instance, in our
feature vector we use the division of total chat sentences,
wins, incomplete matches with players total match count.
As a result, we form a feature vector with 12 dimensions.
For BPM clustering, we use the Infer.NET library [6]. We
use 1.060 players for training the classifier and test with
100 cases. In both training and testing groups, half of the
players are complaint reporters and the other half are
accused players. During the test phase, we systematically
change the confidence threshold to observe the effect of the
threshold on sensitivity, precision and specificity. This
parameter will ideally be optimized on a validation set, and
during the operation of the system, periodically updated to
draw upon ever larger sets of training data.
Experimental Results
In Figures 3 and 4, the results for our tests with various
confidence thresholds are presented. In Figure 3, the
number of false and true positives for different thresholds
are depicted. In Figure 4, the precision, sensitivity and
specificity scores are given.
Figure 3. Players marked as offenders vs. different confidence
thresholds.
The results establish that differentiating a player accurately
as an offender is a difficult task, and sensitivity values are
low in general. However, if we increase the threshold for
confidence, precision and sensitivity increases drastically.
High precision means that if our method labels a player as
offender, he or she is very likely to be a genuine offender.
On the other side, high specificity says that if an innocent
player has been reported as offender, we correctly conclude
that he or she is innocent in most of the cases. A threshold
value of 60% is sufficient for attaining 87.5% precision and
94% specificity, which are good hints for human
moderators when evaluating a complaint.

Figure 4. Precision, sensitivity, specificity vs. different
confidence thresholds.
CONCLUSIONS AND FUTURE WORK
In this study, we presented our analysis of player
complaints data acquired from a real online social game.
Our primary aim was to spot offenders by profiling players
according to their in-game behaviour and performance. We
propose a feature vector for player profile and a binary
clustering methodology, followed by an adaptive
thresholding for confidence to classify players reported in
complaints as genuine offenders. Our approach performs
well enough to aid human moderators in terms of precision
and sensitivity so that they can prioritize and schedule
which complaints to focus. However, by looking at the
sensitivity scores, we observe that our methodology fails to
classify all players in a fully automatized manner.
Therefore, human moderators are still needed for precise
judgments.
We argue player profiling and classification can be used as
a common methodology to evaluate player behaviour in
online games. In addition to general player information, in
this work, we propose features that are game independent
and quantify a players social interaction within the virtual
game community. The method is now extended 1) to test
other clustering algorithms and schemes to validate the
relative merits of the BPM approach, 2) to test a larger set
of non-verbal features for player profiling, 3) to dissect the
contribution of each feature separately and to gain insight
about the players, which can inform incremental game
design and improvement.
DESVIG 2013
10
ACKNOWLEDGMENTS
This research is supported by Bogazici University BAP-
6531project.

REFERENCES
1. AppData - Application Analytics for Facebook, iOS and
Android. http://appdata.com, 2013. [Online; accessed
11-J anuary-2013].
2. Bishop, C. M. Pattern Recognition and Machine
Learning (Information Science and Statistics). Springer-
Verlag New York, Inc., Secaucus, NJ , USA, 2006.
3. Hearst, M., Dumais, S., Osman, E., Platt, J ., and
Scholkopf, B. Support vector machines. Intelligent
Systems and their Applications, IEEE 13, 4 (1998), 18
28.
4. Herbrich, R., Graepel, T., and Campbell, C. Bayes point
machines. The J ournal of Machine Learning Research 1
(2001), 245279.
5. Lewis, B. Istanbul and the Civilization of the Ottoman
Empire. University of Oklahoma Press, 1963.
6. Minka, T., Winn, J ., Guiver, J ., and Knowles, D.
Infer.NET 2.5, 2012. Microsoft Research Cambridge.
http://research.microsoft.com/infernet.
7. Reynolds, K., Kontostathis, A., and Edwards, L. Using
machine learning to detect cyberbullying. In Machine
Learning and Applications and Workshops (ICMLA),
2011 10th International Conference on, vol. 2, IEEE
(2011), 241244.
8. Schouten, B., Tieben, R., van de Ven, A., and Schouten,
D. Human behavior analysis in ambient gaming and
playful interaction. In Computer Analysis of Human
Behavior, A. Salah and T. Gevers, Eds., Springer
(2011), 387403.
9. Shim, K., Sharan, R., and Srivastava, J . Player
performance prediction in massively multiplayer online
role-playing games (mmorpgs). Advances in knowledge
Discovery and Data Mining (2010), 7180.
10. Yan, J ., and Choi, H. Security issues in online games.
The Electronic Library, 2 (2002), 125133.

DESVIG 2013
11
Towards Operationalizing Sociability in Online Video
Games
Georgios Christou
European University Cyprus
P.O.Box 22006, 1516, Nicosia, Cyprus
christoug@acm.org
David Geerts
CUO | Social Spaces, iMinds / KU Leuven
Parkstraat 45 Bus 3605, 3000 Leuven, Belgium
david.geerts@soc.kuleuven.be

ABSTRACT
The question of what sociability is and how it is manifested
in social video games is of prime importance now, when
most players and most games turn towards online multi-
player mechanics. Here we present a preliminary
operationalization of the term, through the examination of 9
online video games that conform to different social
mechanics. We then examine the few methods that include
an evaluation of social mechanics in a game, and conclude
that the existing methods of evaluation of sociability are not
enough to handle most types of sociability as it is expressed
in online video games.
Author Keywords
Video Game; Online Games; Social Games; Sociability;
Evaluation; MMORPG; Facebook Games

ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g., HCI):
Miscellaneous; K.8.0 [General]: Games
General Terms
Design; Measurement.
INTRODUCTION
Sociability is one of those primary differences between
online and offline games. But what is meant exactly when
we talk about sociability in online video games? Is it the
opportunity to play together with other people? Is it the
opportunity to chat with other people inside the game? Or is
it something else entirely?
The question of what sociability is, and how it is manifested
in online video games should be of central concern to
anyone who wants to design for, and evaluate the impact of,
sociability in a video game. In this article, we try to provide
a general overview of how sociability is defined both in the
literature and in the trade press. We ask specific questions
about the nature of sociability and how online games
provide mechanics towards a social gaming experience, and
provide our preliminary answers to these.
SOCIABILITY
Williams states that as humans are social creatures, they
want to have social ties with other humans. These ties may
be instrumental, as these are defined by Habermas [1],
where one human requires the help of others towards
his/her gain, and merely views others as resources, or they
could be social [1], where the actor takes into account the
actions and the reactions of others. The real world offers
certain places that are specifically built to provide the stage
for social interactions. Oldenburg calls them third-places
[2]; places that sustain social gatherings of individuals that
are not connected to the home or the workplace. However,
as Putnam argues [3], a decline in civic and shared spaces,
and a decline in real-world places to meet has stymied the
outlets of real-world sociability. Thus, socializing has
moved to virtual worlds, including the game worlds [4].
Simmel defines sociability as follows: If one stands by the
sociability impulse as the source or also as the substance of
sociability, the following is the principle according to
which it is constituted: everyone should have as much
satisfaction of this impulse as is consonant with the
satisfaction of the impulse for all others. If one expresses
this not in terms of the impulse but rather in terms of
success, the principle of sociability may be formulated thus:
everyone should guarantee to the other that maximum of
sociable values (joy, relief, vivacity) which is consonant
with the maximum of values he himself receives [5].
The concept of socializing in a game world is as old as the
concept of the online world itself. Curtis [6] examined the
social interactions that occurred in a Multi-User Dungeon
(MUD), widely considered the precursor of current
Massively Multi-Player Online Role Playing Games
(MMORPGs). He observed that players were attracted to
more populated areas, a phenomenon he called social
gravity [6]. This parallels behavior seen today in
MMORPGs, where in World of Warcraft (WoW) [7],
perhaps the most widely played MMORPG at the time of
this writing, most players are usually found in the games
cities, chatting with other players. It seems that the cities in
MMORPGs act as an in-game third-place where people
go to socialize while playing the game.
J akobson and Taylor [8] suggest another way of socializing
in MMORPGs that of taking your existing social network
from real life, and transferring it into the game, or vice
versa: your social network takes you into the game and
helps you. They describe their experiences in playing
another MMORPG called Everquest, and how their existing
Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.

DESVIG 2013
12
social network helped them during their start in the game,
explicitly providing examples of how this social network
made the initial stages of the game easier than had they
began playing solo. Thus, MMORPGs bring a new mode of
social interaction, that of actually helping your friends in
starting the game in an easier fashion than it would have
been had you began playing solo.
However, it becomes evident that not all MMORPG players
are eager to socialize. In fact, Bartle [9] defined four such
types of players, only one of which plays the game to
actively socialize. The types of players according to Bartle
[9] are: Achievers those interested in achieving the
ultimate the game has to offer, Explorers those interested
in exploring each nook and cranny of the virtual world and
exploiting that knowledge for some kind of in-game
benefit, Socializers those that play the game to actively
socialize with other players, and Killers those players that
actively want to provide discomfort to other players. This
last category could be seen as a special type of social
player, as without other human players this categorys
gameplay would hold no meaning.
The modern MMORPG however, seems to draw players to
solo play, in contrast to the MUD. The requirement for the
player to gain levels, reaching the coveted end-level, pushes
players towards playing alone together [10], instead of
providing structures that require meaningful social play. In
fact, as Duchenaut et al. propose from a large study of how
players interact in WoW [10], ad-hoc groups are rarer in the
beginning stages of the game, and become more common as
one examines higher level player behaviors. This perhaps
provides additional evidence towards J akobson and
Taylors [8] position, that the real life social network is the
one that helps players when they first enter a MMORPG.
Our own research corroborates the same. From chat logs
and observations performed over a 5 year period in two
MMORPGs, WoW and Star Wars: The Old Republic
(SWTOR), players tend to play solo to level up, and once
they reach the end-level, they start to actively look for
groups, so that they can experience the games group-only
content. This has been recognized by MMORPG
development companies, at least in the two games referred
to here. Thus, in-game tools have been developed to allow
players to randomly join other players specifically to play
each games group-only content.
Nevertheless, more and more players expect full multi-
player capabilities from any game they play. For example,
looking at a very recently (at the time of this writing)
Kickstarter funded game, headed by Chris Roberts
(designer of the Wing Commander series [11]) entitled Star
Citizen [12], one of the driving concepts is to provide both
a multi-player experience, as well as a single-player
experience.
Also, Facebook games are turning towards social
mechanics which even involve real-time Player Vs. Player
(PvP) battles [13]. It was generally accepted that the chaotic
structure of social network sites, such as Facebook, does
not favor the matching of players towards PvP [14]. But, as
has been shown by several games that aim towards the
hard-core gamer [15], players that belong to this category
are starting to embrace this change.
Thus, with such a turn by seemingly all types of online
games towards social content, it is opportune to ask the
following questions:
1. What exactly is sociability in online games, and
how does it manifest?
2. How can we design for a social experience, and
what social mechanics can be leveraged to provide
for a social gaming experience?
3. How do we evaluate social gaming experiences?
4. How do we evaluate the effect of social mechanics
that supposedly offer social gaming experiences?
It is beyond the scope of this article to answer all of these
questions, or to even attempt to answer them completely.
However, we will try to operationalize the definition of
sociability in online video games, and hopefully this will
aid in tackling answers towards the abovementioned
questions.
OPERATIONALIZING SOCIABILITY
Operationalizing any term means to define the term so that
it eventually becomes a measurable thing. Towards this,
here we examine definitions of sociability, and try to extract
features that are associated with game mechanics and game
features. As the concept of sociability is intimately
connected to the concept of online community a
definition for this concept is also relevant. Several
definitions have been proposed, but in this article we will
use the term as defined by Preece [16]: [Online
community is a] virtual social space where people come
together to get and give information or support, to learn, or
to find company. The community can be local, national,
international, small or large. Preece [16] also provides a
definition of sociability for online communities. Preece
argues that three components contribute to good sociability:
purpose a shared focus by the people that belong to the
community, people and how these people may take
different roles in the community, and policies methods
and protocols by which people should behave while they
are in the online community.
We can thus ground our operationalization in the fact that in
order for sociability to manifest we need a community; in
the case of online games, an online community. Through
interactions between the members of the community,
sociability manifests itself.
Player Interactions
To understand what player interactions lead to the
manifestation of sociability, we first need to examine
DESVIG 2013
13
typical interactions between players in online games. Table
1 aims to provide a first attempt to summarizing social
interactions that can occur in online games, categorized
according to the type of games they typically occur. The
social features of 3 MMORPGs (WoW, SWTOR, DC
Universe), 3 core facebook games (Avengers Alliance,
Ghost Recon Commander, and Red Crucible 2), and 3
casual facebook games (ChefVille, FarmVille, and Treasure
Isle). While a complete list of game mechanics that lead to
socializing has been proposed by Raph Koster [17],
Kosters list includes mechanics that have never been used
in online games, so we decided to go with our own list
using the limited sample.
We found that there are three types of social interactions
possible in a game, beyond the obvious one: chatting. These
are Economic Transactions, PvP, and collaborative PvE.
The full list of how we categorized the in-game interactions
is shown it Table 1. For each category we examined the
interactions that can occur in each game. Our preliminary
categorization included the following Economic
Transactions: Gifting, which is a way of sending items from
the game to friends who could be playing the game or not.
Trading, where players can exchange items between them,
or buy and sell items without any mediator, and Auctioning,
where players may place game items for auction at in-game
auction houses. For Player vs. Environment interactions we
found that there could be combinations of cooperative
group play and alone together group play [10], where
players may belong in a group but play in different areas in
the game. Finally, for Player vs. Player interactions, we
found that there could be various combinations, as shown in
Table 1.
It is evident that casual online games provide a very limited
amount of social interactions. Casual games involve the
recruitment of friends towards the exchange of in-game
currency and goods. Thus, in these games, the only social
interaction between players is gifting and/or trading. We
did not find any evidence that Auctioning is part of such
games. Economic transactions at their basic level (that is
excluding Auctioning) allow basic social interaction
between people, but these people must first have a social tie
between them. Economic transactions in casual games can
only occur between friends, where the player sends a gift,
and expects that their social network will reciprocate. The
player can usually send gifts to friends who dont play the
game as well, so that these friends may be intrigued and
begin playing the same game as well. Thus, casual games
use a form of recruiting and marketing through the gifting
mechanism.
As explained however, economic transactions are only
allowed between the players social network. Therefore,
these transactions are based on the fact that the player will
recruit a social network to play the game with, that already
exists, or that will be created through some out-of-game
structure. Hence, economic transactions can be considered
as a social mechanism that utilizes the social network, but
does not allow the building of new social capital [3].
PvP, PvE, and chat are all offered through some, or all of
the two game categories: MMORPGs and Core online
games. Using the interactions that belong to these
interaction categories caters to players enlarging their
social network, not just retaining it. Through public chat for
example, players can meet other players whom they do not
know in real life. Through the different PvP interactions,
players may play together or group against other live
players. These are social features not offered in casual
online games, that make the social experience broader, and
perhaps more satisfying.
Thus, a first operationalization of the term socialization
for online games may begin with the types of features that
online games offer, and according to how well they support
social actions. Casual online games only support actions
that can be performed in an existing social network. Core
Table 1 Social interactions by game and interaction type
Game Type I nteraction Type I nteraction
Casual Online
Games, Core
Online Games,
MMORPGs
Economic
Transactions
Gifting
Casual Online
Games, Core
Online Games,
MMORPGs
Economic
Transactions
Trading
MMORPGs Economic
Transactions
Auctioning
Core online
games,
MMORPGs
Player vs.
Environment
(PvE)
Cooperative play
in group
MMORPGs PvE Alone Together
Core online
games,
MMORPGs
Player vs. Player
(PvP)
One vs. One
MMORPGs Player vs. Player
(PvP)
One vs. Group
Core online
games,
MMORPGs
Player vs. Player
(PvP)
Group vs. Group
Core Online
Games,
MMORPGs
Chat In-game private
chat
Core Online
Games,
MMORPGs
Chat In-game public
chat

DESVIG 2013
14
online games and MMORPGs though, offer actions that
allow players to grow their social networks. Furthermore,
MMORPGs allow two social actions that are not available
in either core games or casual games. The first is
auctioning, which is a mechanism that allows players to
provide in-game goods to every other player in the game.
The second is 1 vs. Group interactions. Even in core online
games which are based on live-PvP (vs. PvP against teams
of AI-controlled opponents, whose members have been
created by other human players) the mechanic of 1 vs.
Group is not catered for directly. Because of the open world
nature of the MMORPG, groups of players may seek other
players who play solo, and attack them. Sometimes this
falls under griefing, where a player or a group of players
repeatedly kill another player, effectively not allowing
him/her to play the game [18].
Using the above, we can place the three groups of online
games in a series, by which game allows the most
sociability:
MMORPG >Core online game >Casual online game
The abovementioned operationalization, together with the
features identified in Table 1, allows us to devise a scheme
for manipulating the social features of a game, making it
more or less social. However, Table 1 features, as well as
the operationalization of the previous section do not take
into account out-of-game social features, nor content
developed to support and require grouping. Christou et al.
[19] have developed a framework for thinking about
designing the social features of online games that takes into
account out-of-game structures, as well as in-game
structures that promote sociability.
EVALUATION OF SOCIAL GAME MECHANICS
Now we have preliminarily defined sociability in online
games, and identified several types of social interaction for
operationalizing it, the question remains on how to evaluate
these social game mechanics and sociability in general.
As in usability evaluation, there are two common
approaches to evaluating sociability: expert based and user
based. An expert based method that is quite popular, due to
its low cost and the relative ease of applying, is heuristic
evaluation. Paavilainen [20] reviews four existing sets of
heuristics for video games, and analyses how many refer to
social aspects of the game. Not surprisingly, most heuristics
contain little reference to social aspects, except for one set
of heuristics specifically aimed at multi-player games.
These Network Game Heuristics [21] cover only some
types of interaction aimed at supporting online sociability
as we have defined it, e.g. chatting (Support social
interaction: provide support for planned and opportunistic
social interactions or Appropriate communication tools:
provide communication features that accommodate the
demands of game play) or PvP (Support coordination:
provide features that allow players to coordinate their
actions during cooperative game play). Our earlier
operationalization highlights aspects that are missing in this
set of heuristics. Paavilainen presents his own sets of
heuristics aimed at social games on Facebook, which we
referred to as casual online games. Some interaction types
in our operationalization are covered in these heuristics as
well, such as Sharing. Provide means for players to share
information and in-game resources. The strength of
heuristics is also their weakness: as they should be
applicable to a wide range of games, they are rather coarse
and do not give very specific guidance on how well a
certain feature is implemented, and how sociability can be
improved e.g. by different means of enabling players to
share information. Especially in our more detailed
operationalization of sociability, the heuristics fail to
address several aspects of each interaction type.
To overcome the limitations of heuristics, a more elaborate
and costly method is often used: user testing, or playtesting
as it is usually called in the context of games. While it
seems obvious that it is easier to evaluate sociability when
you have groups of people playing against each other while
observing them, it is far from straightforward to effectively
measure sociability directly. It is e.g. possible to count the
amount of social interactions or economic transactions, but
a higher count doesnt necessarily imply a higher degree of
sociability in the sense of sociable values (joy, relief,
vivacity) as defined by Simmel. As only observing or
logging certain interactions is not sufficient, questionnaires
could be used in conjunction with playtesting, e.g. for
measuring connectedness as one aspect of sociability.
While this is seldom used in research on social gaming, it is
more common in an area such as awareness systems [22].
However, connectedness alone is not sufficient as a
measure for sociability, as the range of interaction types is
quite broad. As with heuristics, our operationalization
highlights the need for more specific measures for each of
these types.
CONCLUSION
In this article we have tried to provide a preliminary
operationalization of sociability, through features and social
mechanics included in various types of online games. We
examined 9 video games from three different categories:
MMORPGs, Core online games, and Casual online games.
We found that the amount of social mechanics included in
the games can be a valid way of operationalizing
sociability.
We then examined existing methods of evaluating online
games, placing our attention to how these methods propose
sociability evaluation. We found that all the methods
examined lack mechanisms that are powerful enough to
evaluate whether sociability structures work in the way they
are designed in a game.
Because of our findings we suggest that more research is
required, first in establishing an operationalized definition
for sociability, and then for the use of that definition in
methods and frameworks to evaluate sociability structures
DESVIG 2013
15
and mechanics. Once this is done, then game designers and
developers will have a way to structure their games around
how they want their players to socialize, and will have a
way to measure whether their sociability goals are being
met by their designs.
REFERENCES
[1] Habermas, J . (1981) The theory of communicative
action (Volume 1): Reason and the realization of society.
Beacon Press, Boston, MA.
[2] Oldenburg, R. (1989) The great good place. Marlowe &
Company, New York, NY.
[3] Putnam, R. D. (2000) Bowling alone: The collapse and
revival of american community. Simon & Schuster, New
York.
[4] Williams, D. (2006) Why game studies now? Gamers
don't bowl alone. Games and Culture, 1(1), 13-16.
[5] Simmel, G. and Hughes, E. C. (1949) The sociology of
sociability. The American J ournal of Sociology, 55(3), 254-
261.
[6] Curtis, P. (1996) Mudding: Social phenomena in text-
based virtual communities. MIT, City.
[7] Blizzard Entertainment (2010) Blizzard Entertainment:
World of Warcraft: Wrath of the Lich King. City.
[8] J akobson, M. and Taylor, T. L. (2003) The Sopranos
meets Everquest: Social Networking in MMOGs.
Proceedings of DAC 2003, Melbourne, Australia.
[9] Bartle, R. A. (1996) Hearts, Clubs, Diamonds, Spades:
Players who suit MUDs. J ournal of MUD Research, 1(1).
[10] Ducheneaut, N., Yee, N., Nickell, E. and Moore, R. J .
(2006) "Alone together?": exploring the social dynamics of
massively multiplayer online games. In Proceedings of the
CHI 2006 Conference of Human Factors in Computing
Systems, Montreal, Quebec, Canada, ACM Press, 407-416.
[11] Roberts Space Industries Announcing Wing
Commander, Retrieved J anuary 16th, 2013 from
http://robertsspaceindustries.com/announcing-wing-
commander/.

[12] Roberts Space Industries (2012) Star Citizen,
Retrieved 16th J anuary 2013, from
http://www.robertsspaceindustries.com/star-citizen/.
[13] Ricchetti, M. (2012) What makes social games social?
Gamasutra.com, Retrieved 16th J anuary 2013, from
http://www.gamasutra.com/view/feature/135102/what_mak
es_social_games_social.php.
[14] Curtis, T. (2012) Real-time multiplayer games: Could
they help save Facebook? Gamasutra, City, Retrieved 16th
J anuary 2013, from
http://www.gamasutra.com/view/news/175581/Realtime_m
ultiplayer_games_Could_they_help_save_Facebook.php#.
UPZgUCc4sgU..
[15] Hindman, B. (2013) Core Corner: In 2013, how will
core games continue the social invasion? , Retrieved 16th
J anuary 2013, from
http://blog.games.com/2013/01/04/hardcore-social-games-
2013-predictions/.
[16] Preece, J . (2001). Sociability and usability in online
communities: determining and measuring success.
Behaviour & Information Technology, 20(5), 347-356.
[17] Koster, R. (2011). Social mechanics: The engines
behind everything multiplayer. Presented at the Game
Developers Conference 2011 (Austin, TX, 2011).
[18] Chesney, T., Coyne, I., Logan, B. and Madden, N.
(2009). Griefing in virtual worlds: causes, casualties and
coping strategies. Information Systems J ournal, 19(6), 525-
548.
[19] Christou, G., Law, E. L.-C., Zaphiris, P. and Ang, C. S.
(Accepted). Challenges of Designing for Sociability to
Enhance Player Experience in Massively Multi-player
Online Role Playing Games. Behaviour & Information
Technology.
[20] Paavilainen, J. (2010). Critical review on video game
evaluation heuristics: social games perspective. In
Proceedings of the International Academic Conference on
the Future of Game Design and Technology, ACM Press,
Vancouver, BC, Canada, 58-65.
[21] Pinelle, D., Wong, N., Stach, T. and Gutwin, C.
Usability heuristics for networked multiplayer games. ACM
Press, Sanibel Island, FL, USA, 169-178.
[22] Romero, N., Markopoulos, P., Baren, J ., Ruyter, B.,
IJ sselsteijn, W. and Farshchian, B. (2007). Connecting the
family with awareness systems. Personal Ubiquitous
Computing, 11(4), 299-312.




DESVIG 2013
16
Being social in online games:
Five research areas to consider
Mia Consalvo
Concordia University
Montreal, Canada
mia.consalvo@concordia.ca



ABSTRACT
This position paper articulates and expands upon five areas
for continued exploration in studying sociality in online
games. Types of players, player networks, a games design
and platform, as well as player motivations all matter to
how sociality is performed in online videogames. Further
study of that complexity is necessary to lead to better
understandings of what happens when people gather
together to play online.
Author Keywords
Online videogames, sociality, player motivations, family,
friends, strangers, sociability, platform, design
ACM Classification Keywords
Human Factors; Design; Measurement
General Terms
Human Factors; Design; Measurement.
INTRODUCTION
For as long as the pop culture stereotype of the gamer nerd
sitting alone in a basement playing games and eating junk
food has existed, game studies scholars have been pushing
the opposite idea that games are, in fact, a social activity.
Games bring people together, deepen relationships, expand
our knowledge of the world, and let us engage in activities
with friends and family. Yet while many scholars continue
to promote such views, we know surprisingly little about
who is playing together, how they are doing so, and why.
Likewise, we need better knowledge of how game
platforms and game design shape sociality in play, as well
as how pre-existing social networks and structures
influence play. What follows is an admittedly incomplete
list, drawn from my own work in player culture, of
elements that need to be consideredas well as studied
more extensivelyif we wish to gain a more thorough
grasp on how the social is enacted via online gameplay.
1. Who is being social, and in what configuration?
When researchers consider the composition of online player
groups, it seems that most of the attention goes to
problematic play and the challenges that griefers and
cheaters offer to those wishing for more balanced play, and
how structural issues related to anonymity and lack of
sanctions for offenders continue to keep griefers and
cheaters a common occurrence. For example, my own work
on cheating in online games demonstrated how players
wanted the ability to punish those who cheated, or at the
least to identify them and exclude the cheaters from future
encounters [1]. Those findings presupposed that the
cheaters were strangers, and there was little way to track
such people between encounters, across games, or over
time. While most of those players were active in MMOGs
and competitive multiplayer games such as Counter Strike,
players of social network games have also been found to
worry about cheating in games, and wish to ensure that
those they play with do not have any unfair advantages [2].
Thus strangers continue to pose challenges to successful
sociality in online games, and players respond accordingly.
Yet in addition to the problems that strangers bring to
online gameplay, there are additional layers of complexity
when considering how friends interact with one another in
online games, as well as how family members do the same.
The relationships that friends have with one another, and
the many different roles and responsibilities of family
members can bring amazingly complex systems to bear on
games that have been given little to no consideration. When
considering multiplayer there is usually little to no
distinction drawn between who is playing with whomis it
college friends who are co-located at a LAN party, or
childhood friends reuniting at a distance to engage in an old
pastime? Families that play together can pose even greater
challenges, given the hierarchies involved and the evolving
nature of family relations. There have been a few early
studies of families who played videogames together, yet
they have mainly concentrated on nuclear families with
young children [3]. Increasingly we see adult family
members play cooperatively and competitively in various
game genres and across multiple platforms. So





Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.
DESVIG 2013
17
understanding how an extended family might use a social
network game in comparison with how friends use the same
game is a key project to undertake. Likewise, investigating
how friends use games to both build and maintain
relationships as well as engage in a common activity- and
how different games, platforms, genres and play styles
encourage or discourage those activities- is another
important piece of the puzzle to study.
2. Platform matters
Player communities and groups are dynamic entities that
can be both remarkably durable as well as extremely
fragile. For example, when I first began studying the player
community for the casual MMOG Faunasphere, the game
was available via web browser for the general public.
During the beta period and early public launch, players
created norms and expectations for how to behave within
the game space [4]. When the games developers
subsequently enabled the game within the Facebook
platform, new players with different play histories as well
as expectations for play styles were added to the mix,
resulting in dissatisfaction for a subset of players. The
earlier and established player culture of the game had been
disrupted by players who expected others to simply give
them needed items a norm developed and institutionalized
by Facebook game developers such as Zynga. The resulting
clash over expectations caused rifts in the player
community, and led at least some players to quit the game
in frustration.
Furthermore, platforms such as Xbox Live have become
known as being supportive of toxic gamer culture, where
players must come to expect certain levels of trash-talking
that draws on sexist, racist, and homophobic speech [5].
Web sites such as Fat, Ugly or Slutty have emerged to
document that abuse and make an intervention in the space,
yet with little larger effect on player culture. But we have
little systematic knowledge of how sociality differs by
platform, or even how one games sociality might change as
the game is encountered across various platforms.
3. Designed sociality/Game objectives and sociality
The 2002 MMOG Final Fantasy XI was originally designed
to force players to come together to play from a very early
point in the gameupon reaching approximately level 10,
players were implicitly required to begin grouping if they
wished to advance at all into higher levels and endgame
activities. In contrast, World of Warcraft and other similar
titles have taken a different approach, encouraging players
to come together yet largely not requiring it. WoW makes
more activities open to soloing, and has over time reduced
the number of players required to participate in certain
activities. FFXI has made similar changes, attempting to be
more flexible in response to player interests. This is another
way to understand how and why players are social in
gameswhether the game requires sociality via grouping
(or other social means) to advance, or if it makes it
optional. Going beyond MMOGs, we can ask how other
types of games (social network games, mobile games)
require, encourage or ignore opportunities for players to
play together or be social in novel ways. Likewise it is key
to see what kinds of sociality are required or encouraged
it is competitive PvP or cooperative PvE? How is
asynchronous versus synchronous sociality managed?
Finally, what the designed systems that games offer to
player to be social, and how do these systems work in
practice? Are they functional, effective, broken and/or
valuable? How do players actually use them?
4. Quality and type of social interaction
Its not enough to say that players simply wish to play with
others in online spacesor that some multiplayer instances
are disliked by some or many individuals. What types of
interactions matter to players? How extensive must or
should interaction be? Is communication either via text or
voice a necessary component of being social, or is doing
something together also valued? We dont have many
answers about how players feel in relation to these
questions, across various types of games or across different
player groups. While some research has indicated that
players enjoy being alone together in online spaces [6],
we need more information on how and why they feel this
way; and if better designed spaces might change their
views.
Likewise, although we are beginning to understand how
groups within games communicate with one another via
various game tools, players also use other programs and
ways to be socialboth synchronously with gameplay and
asynchronously outside of direct play. How are players
using social network sites, twitter, instant messaging, email,
private forums, and private servers to engage in both
instrumental sociality and sociability [7]?
5. Player motivations in and around sociality
Initial research on players [8, 9] has explored the varied
motivations for why individuals play games, including
possibilities to explore, achieve, and be social (among many
other reasons). Yet within the category of being social
in what ways do players want to be social? What is of prime
importance, and what is less important to them? For players
that cheat, how is it considered a social or anti-social
activity? How do social needs or interests change over
time- both for individual players, and over the life of a
game? For example, in Faunasphere players initially
enjoyed being social but it wasnt a top reason for playing
the game [10]. When the games closure was announced,
spending time with friends in the game suddenly became a
more important activity for many players in the game.
Clearly ending points for games might change how players
DESVIG 2013
18
enact sociality, but other temporal periods within games are
also important to explore. For example, we can also ask if
beta periods, for example, are more social or have different
types of social activity than more stable periods in a
games life span. The particular makeup of player groups in
beta periods makes them another key area to study.
CONCLUSIONS
This brief discussion paper offers no real conclusions, but
instead tries to articulate areas for continued exploration in
studying sociality in online games. Types of players, player
networks, a games design and platform, as well as player
motivations all matter to how sociality is performed in
online videogames. Acknowledging that complexity can
only lead to better understandings of what happens when
people gather together to play online.
REFERENCES
1. Consalvo, M. Cheating: Gaining advantage in
videogames. MIT Press, Cambridge, MA, 2007.
2. Consalvo, M. Cheating2.0 Paper presented at Gaming the
Game Conference, Davis, CA, 2012.
3. ONeill, M. Videogame playing and family
communication patterns. J ournal of the Northwest
Communication Association 31, 68-80, 2002.
http://www.acm.org/class/how_to_use.html.
4. Begy, J . & Consalvo, M. Achievements, motivations and
rewards in Faunasphere. Game Studies 11, 2011.
5. Kennedy, T. The voices in my head are idiots:
Rethinking barriers for female gamers. Paper presented at
the annual conference of the Association of Internet
Researchers, Milwaukee, WI, 2009.
6. Ducheneault,N.,Yee,N.,Nickell,E,&Moore,R. Alone
together? Exploring the social dynamics of massively
multiplayer online games. CHI 06 Proceedings of the
SIGCHI Conference on Human Factors in Computing
Systems, 407-416, 2006.
7. Eklund, L. The sociality of gaming. Doctoral
dissertation, Stockholm University Department of
Sociology, 2013.
8. Bartle, R. Hearts, clubs, diamonds, spades: Players who
suit MUDS. Available online at
http://www.mud.co.uk/richard/hcds.htm
9. Yee, N. A model of player motivations. Available online
at http://www.nickyee.com/daedalus/archives/001298.php?
page=4 2005.
10. Consalvo, M. & Begy. J . Virtual pets and the end of the
world: Studying an MMOGs closure. Paper presented at
the Philosophy of Computer Games conference, Madrid,
Spain, 2012.




DESVIG 2013
19
Revi val Actions in a Shooter Game
Sauvik Das
Carnegie Mellon University
sauvik@cmu.edu
Thomas Zimmermann
Microsoft Corporation
tzimmer@microsoft.com
Nachiappan Nagappan
Microsoft Corporation
nachin@microsoft.com

Bruce Phillips
Microsoft Corporation
Bruce.Phillips@microsoft.com
Chuck Harrison
Microsoft Corporation
charlesh@microsoft.com


ABSTRACT
In this paper we present an analysis of revival actions in a
third person shooter video game. We observe that players
who revived their teammates played significantly more ses-
sions. They appear also to be more skilled and successful in
the game in terms of kills, wins, and deaths. We then dis-
cuss how to extend our analysis methodology to other types
of social play as well as other games.
Author Keywords
video games; social play; online multiplayer
ACM Classification Keywords
K.8.0 Personal Computing: GeneralGames
General Terms
Human Factors; Measurement.

INTRODUCTION
With the advent of computer games being increasing dis-
tributed and played online it is important to understand the
social implications, relationships and behavior of players.
In 2006, Ducheneaut et al. [1] studied Massively Multiplay-
er Online Games (MMOGs) and found that there was little
empirical data to assess players social experiences. Based
on longitudinal gameplay data from the World of Warcraft
(WoW) game, the authors analyzed various social factors
such as player time and leveling, grouping patterns, social
networks and their densities. The authors also observed that
while MMOGs were clearly social environments, joint ac-
tivities are not very prevalent, especially in the early stages
of the game. The players instead of playing with others
were playing for them as an audience or more specifically
like being alone together surrounded by others, but not
necessarily actively interacting with them. More recently
Nacke et al. [3] analyzed game websites to learn more
about social behaviors. Using the website log files they
explored three predominant themes: permanence, whether
people formed a long-term association with the site; social
interaction, in terms of shared activity and verbal commu-
nication; and formation of ties, whether people made con-
tacts with others [3]. The authors discussed the nature of
social interaction in terms of game structures, actions, and
impersonal interaction. They also provided recommenda-
tions for site design and the benefits of impersonal and ac-
tivity-based interaction in these sites.
All this research points to growing interest in social play in
the games community. In this paper we present an analysis
of social behavior in a third-person shooter game, more
specifically focusing on in-game behavior like kills, assists,
helping teammates in danger and sharing weapons with
respect to the revive actions a player gives within the
game. These results serve as a starting point for us to look
at how people interact within a third-person shooter game.
We also discuss how to generalize our analysis to other
types of social behavior and other games.
More closely related to our research, Xu et al. [4] analyzed
social relationships in Halo, a popular first-person shooter
games. Using a mixed methods approach they observed
relationships in terms of configuring social relationships to
enjoy the game and configuring games for social bonding.
They also quantified various types of player networks, for
example star networks, where one or two players in the
center sharing the same playmates as the participant form-
ing a tightly knit player group.

THE GAME
For this paper, we analyze a popular third-person shooter
game on Xbox 360. Third-person shooter games are closely
related to first person shooter games [5] and players take on
the role of a story character (avatar), armed with weapons
to shoot down their foes. The game includes a multiplayer
online component that is key aspect of the gameplay and
provides players the ability to play with and against other
human players online. Players can band together with
friends to make a team composed of specific players or can
join a team of other players who may be unknown. All mul-
ti-player sessions consist of two opposing teams trying to
achieve some pre-specified goal based on the type of game
being played. Players who do especially well can win tro-
phies for specific accomplishments in the game.
Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.

DESVIG 2013
20
SOCIAL PLAY
We operationalize the term social as any behavior that
involves coordinated action across teammates or is done for
the benefit of another teammate. In the game, players have
several opportunities to exhibit social behavior in multi-
player game sessions.
When players are knocked down, but not yet killed,
their team members can revive them.
Players can assist team members in targeted kills.
Players can give teammates weapons or ammo.
Players can spot enemies, where they mark their target
for coordinated assault.
In this paper we will focus on revival actions by players.
ANALYSIS OF REVIVAL ACTIONS
We used gameplay data that was collected as part of public
beta of the game from players with their consent. For pur-
poses of scale, we randomly selected a sample of 26,000
players who played more than 928,000 sessions in total.
We categorized the players into two groups:
Group A: The player never revived any team mates.
Group B: The player revived team mates at least once.
Next we compared the groups across several dimensions
(see Table 1 for a summary of the increase from Group A to
Group B):
Engagement: We found that players who revived their
teammates played significantly more sessions than the
players who never revived: 122.96 vs. 30.94 sessions,
an increase of 297.44%.
Skill: Players who revived also turned out to be more
skilled. They had almost twice as many kills as the
group with no revives and thrice as many kills when an
enemy was executing a teammate. They were less like-
ly to be in need of reviving and died fewer times.
Success: We observed that players who revived team-
mates are also 18.88% more likely to win a game.
Social: Players who revived other were more likely to
exhibit other social activities such as giving weapons
(an increase of 486.14%).
These findings suggest that players who exhibit social be-
havior have a better experience in general in the game that
we analyzed for this paper.
FUTURE WORK
In the previous section, we presented a preliminary study of
social gameplay. In our future work, we plan to extend our
study by investigating more characteristics than the ones
listed in Table 1. In addition we plan to analyze the impact
of social behavior at three levels: (1) impact of a players
own social behavior on his experience, (2) impact of his
teams social behavior, and (3) impact of his enemys social
behavior within their own team.
While player experience is a broad term encompassing
many variables, we plan to specifically focus on retention,
which we define as total session count, similar to Weber et
al. [6] Retention is perhaps one of the most descriptive and
easily quantifiable measures of experienceplayers who
enjoy a game can be expected to play it for longer, whereas
players who do not enjoy a game can be expected to stop
playing it quickly.
Thus, we have a single, continuous response variable (total
session count) and several predictor variables (features de-
scribing social behavior). Though such a setup naturally
lends itself to multivariate regression analysis [7], the pres-
ence of other variables in the gameplay data is potentially
confounding (because of potential covariance). Further,
regression analyses generally tend to account for only a
certain percentage of the variance in the data. As such, we
plan to first focus on isolating the social features and con-
sider the presence of social variables in a broader context in
a later analysis.
Our main analysis strategy will be a combination of cluster
and contrasting techniques:
1. We first use k-means clustering to cluster players in
our sample along the social features.
2. We then analyze the cluster centroids to understand the
differences in social behavior across clusters.
3. Finally, we run a survival analysis to observe trends in
retention across clusters.
This strategy is a generalization of the analysis that we ex-
emplified with revival actions in the previous section. Ra-
ther than automated clustering we manually specified the
clusters using the presence of revivals (Step 1+2) and then
compared the clusters with respect to several characteristics
such as session counts, kills, matches won, etc. (Step 3).
We believe that this quantitative analysis approach can also
be broadly applied to other games.
Table 1: Characteristics of the group of players who revived
other players relative to the players who never revived. When
appropriate characteristics were normalized per session.
Dimension Characteristic Change
Engagement Session count +297.44%
Skill Kills +100.21%
Killing an enemy that is
executing a teammate
+194.85%
Was revived 54.55%
Deaths 12.44%
Success Likelihood to win match +18.88%
Social Gave weapon +486.14%

DESVIG 2013
21
ACKNOWLEDGEMENTS
Sauvik Das performed this work during a summer intern-
ship at Microsoft Research. Thanks to Coco Inoue for her
help and her insightful feedback.
REFERENCES
[1] N. Ducheneaut, N. Yee, E. Nickell and R. J . Moore,
""Alone together?": exploring the social dynamics of
massively multiplayer online games.," in Proceedings of
the 2006 Conference on Human Factors in Computing
Systems (CHI), 2006.
[2] L. Nacke, G. McEwan, C. Gutwin and R. L. Mandryk,
""I'm just here to play games": social dynamics and
sociality in an online game site," in Proceedings of
CSCW '12 Computer Supported Cooperative Work,
2012.
[3] Y. Xu, X. Cao, A. Sellen, R. Herbrich and T. Graepel,
"Sociable killers: understanding social relationships in
an online first-person shooter game," in Proceedings of
the 2011 ACM Conference on Computer Supported
Cooperative Work (CSCW), 2011.
[4] E. Adams, Fundamentals of Game Design, New Riders;
2nd edition, 2009.
[5] B. G. Weber, M. J ohn, M. Mateas and A. J hala,
"Modeling Player Retention in Madden NFL 11," in
Proceedings of the Twenty-Third Conference on
Innovative Applications of Artificial Intelligence (IAAI),
2011.
[6] L. Wasserman, All of Statistics: A Concise Course in
Statistical Inference, Springer , 2003.



DESVIG 2013
22
Limelight Fostering Sociability in a Co-located Game
Wolfgang Hochleitner, Michael Lankes,
J eremiah Diephuis
University of Applied Sciences Upper Austria
Softwarepark 11, 4232 Hagenberg, Austria
{wolfgang.hochleitner, michael.lankes,
jeremiah.diephuis}@fh-hagenberg.at
Christina Hochleitner
CURE Center for Usability Research and
Engineering
Businesspark MARXIMUM,
Modecenterstrae 17, 1110 Vienna, Austria
hochleitner@cure.at

ABSTRACT
We examine the requirements for sociability in games and
assess their relevance for co-located play in a public space
game. Particular focus is placed on the design of different
player roles and a smooth transition between these roles.
Observations from an initial public exhibition of the game
are discussed and ideas for future work are presented.
Author Keywords
games; public display; sociability; player roles; co-location
ACM Classification Keywords
K.8.0 Personal Computing: Games
General Terms
Design; Human Factors
INTRODUCTION
Games enable us to consciously control, structure and
delimit social interaction [12]. While playing alone is
certainly a rewarding experience, playing with others offers
a whole range of new possibilities. Collaboration or
competitive play may emerge, yet players often simply
enjoy the experience of being together (in a sense of
playing together) with others. This phenomenon, termed
sociability, was coined by the sociologist Georg Simmel in
[11] and has recently received increased interest in the
games research community. In particular, the field of
multiplayer online games (MOGs) has been the focus of
several research activities, e.g. [3, 5, 13].
Apart from online games, co-located play proves to be
another interesting area of research for sociability. Co-
located play manifests itself in various ways: playing a
game next to each other on one computer, competing in
LAN parties or interacting together in a public space setting
(c.f. [10]). For games taking place in public spaces and
online games, social interaction is of particular importance.
Involved players may work together to reach a game goal,
they may take action against each other, or they might
simply enjoy the presence of others. From this standpoint,
the aspect of sociability is also crucial for non-players.
Even if they do not need to collaborate, all participants,
including spectators, share a similar experience and may,
for example, get in each other's way [12]. Designing games
that address and make use of these considerations is a
challenging task.
We created Limelighta co-located public gameto gain
insight into sociability factors for co-located games. The
following section gives an overview of related work; we
then describe the design of the game and relate it to
requirements from the field of MOGs. We conclude with
the description and discussion of observation data gathered
throughout an initial run of the game.
RELATED WORK
Co-located games are a challenge in a public setting as they
have to take into account both individual preferences and
group dynamics. Thus, player roles and sociability aspects
need to be addressed.
Player roles
When several people play together, different types of user
behavior emerge. In this regard, Brignull and Rogers [1]
investigated large public display interaction and identified
three classes of users based on their activity patterns. The
first class is in direct contact with the installation, whereas
bystanders can either have a focal awareness of the display
or simply share a peripheral awareness.
Finke et al. [8] found support for this conceptual framework
by introducing the following player types: bystanders are
those individuals that do not have a strong interest in the
presented content. Conversely, spectators are consciously
aware of the displayed content and the environment, but do
not participate in connection with the content on the
display. The last type consists of actors who are in direct
contact with the installation and play an active role in the
manipulation of the presented content. The researchers also
stress the importance of motivating players to carry out a
contributor role. To encourage individuals to interact with a
system, they advocate designing applications to support
transitions between the class boundaries (especially those in
crossing the threshold from peripheral to focal awareness
activities).
While examining these roles, it becomes quite apparent that
participants have different goals and motivations in a game

Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.
DESVIG 2013
23
scenario (e.g. active playing, spectating or idle by-
standing). However, to design games in this context, it is
not sufficient to operate on an individual level; rather, it is
essential that social aspects be taken into account. We
accomplish this by incorporating sociability into our game
design process.
Sociability
A survey among researchers and game designers in [3]
revealed six distinct requirements to enable sociability in
the field of MOGs:
In-game communication: Refers to the communication
during the game (e.g. chat).
Off-game communication: Defines experiences situated
around the game (e.g. forums).
Sociability through grouping and rewards: The tendency
to join groups, possibly supported by in-game rewards.
Sociability through empathy: Defining a clear enemy and
therefore a common goal.
Sociability through world design: Making sure that the
design of the world puts together players of similar
experience.
Sociability through designed relationships: Defining
gameplay in a way that players have to rely on others to a
certain degree in order to succeed.
Even though these requirements were compiled with MOGs
in mind, they might as well apply to any multiplayer game.
In-game communication will occur at some point in a game,
whether it is via a built-in chat system or just face-to-face,
as in the case of a co-located game.
Off-game communication in MOGs might benefit from the
fact that these games usually have a large user base that
either creates additional information outside the games
themselves (e.g. through fan sites) or makes game
companies ship them with the game anyway (e.g. ArenaNet
hosts a wiki for Guild Wars 2
1
themselves). Still, even
games that take place in the same location enable this kind
of communication, e.g. in the form of a verbal post-mortem
between players.
The tendency to formgroups also happens frequently when
people come together. Especially in a public setting, it is
easier for people to play as a group in order to counteract
social embarrassment [2].
Also empathy, defined in [3] as identifying a common
enemy, is not limited to MOGs. Forming a collective due to
a common enemy is a known sociological principle.

1
http://wiki.guildwars2.com/
Creating sociability through world design, on the other
hand, is a requirement unique to games that implement a
level system (such as most online role-playing games). If
the term world design is seen in a more generalized way
(independent of player levels), it can certainly also be used
for other multiplayer games.
Finally, designed relationships will also work in co-located
multiplayer games. The necessity to rely on others can be a
vital factor for social interaction, althoughas mentioned
in [3]they can be counterproductive if overdone.
In order to further assess the potential for sociability in
games, we endeavor to investigate if these requirements can
be applied to a co-located game setting. For this purpose,
we designed and created a co-located game that facilitates
different player roles and addresses relevant sociability
requirements. The next sections will describe the game
design, location challenges and observations regarding
player behavior while the game was played. We will then
discuss these observations in terms of sociability.
LIMELIGHT GAME DESIGN
Limelight is a co-located public game that was created for
the 2012 Ars Electronica Festival
2
. It was designed for
Deep Space, a special room within the Ars Electronica
museum featuring a 169 meter wide projection surface on
the wall as well as an equally large floor projection area
that also includes an optical tracking system. Players of
Limelight can assume one of three distinct roles in the
game: Limus is the main character. He is a wizard's
apprentice who is sent to the cellar by his master to retrieve
three magical stones. The cellar is completely dark and only
lit by the flare emitted by his wand. In order to navigate
Limus through the cellar, represented by a classic 2D-style
platform maze on the projection wall, the player uses a
conventional gamepad. Limus can be supported in his task
by a varying number of Lumees, will-o'-wisp-like creatures
that can hover around freely in the cellar and create a pool
of light, illuminating parts of the level. Lumees are
controlled by people who step onto the tracked floor
projection area in Deep Space and move around within its
boundaries. Finally, there are the SMorcS. These menacing
creatures roam the cellar and can harm Limus, but cannot
be controlled directly. Instead, they can be created in one
out of three locations in the top-level regions of the maze
by spectators who choose to send an email or SMS. The
message contains the desired spawn point, a desired color
and a custom name. Once received by the system, a SMorc
is created and will work its way down through the platform
maze. The basic layout of the game, including Limus,
Lumees and a SMorc, is shown in Figure 1.

2
http://www.aec.at/festival/en/
DESVIG 2013
24

Figure 1. The basic layout of Limelight. Limus is shown on the
right; two Lumees are illuminating a SMorc and a magic stone.
Our game design addresses the previously introduced
sociability requirements formulated by Christou and
colleagues [3].
In-game communication: due to the co-located setting,
players are able to communicate directly via voice and/or
body language. The game's music is intentionally
unobtrusive and features only a low-volume, atmospheric
soundtrack to grant players an efficient and effective way
of interacting with each other.
Off-game communication: as the dimensions of the Deep
Space display are very large, spectating visitors have the
opportunity to both observe the current game status and
continually make comments about current or past
situations.
Sociability through Grouping and Rewards: The game
design features three different player groups (Limus,
SMorcS and Lumees) that can be joined at any time.
Rewards are given by reaching game goals, such as
collecting all the magical stones, or by helping Limus as
a Lumee to detect SMorcS as well as other obstacles.
Sociability through Empathy: Limelight is a game that
offers players various types of playing styles. As Limus,
the obstacles are the SMorcS, the level itself and, of
course, the darkness within a level. From the perspective
of the SMorcS, the little wizard is the enemy that has to
be stopped from collecting all the magical stones.
Lumees, on the other hand, do not have a common enemy
as they exist on a meta-level, which enables them to set
their own goals and define their enemies (i.e. helping
Limus or revealing SMorcS).
Sociability through World Design: the visual quality of
the game (characters and levels) and the setting (Deep
Space) complements the presence of players with a quite
similar game experience.
Sociability through Designed Relationships: in Limelight,
players are continually in close contact with each other
both embedded in the game design and the physical
setting. The design fosters competition, communication
and collaboration.
PLAYING IN DEEP SPACE
In addition to taking sociability requirements into account,
several location-related aspects had to be considered. The
specific features of the museum, as well as the possibilities
and requirements of the Deep Space premises led to our
formulation of three major goals during the design process
of Limelight, similar to the goals described in [2]:
Scalability: Being a public game, it needs to work with
any number of people (within the given technical
boundaries).
Smooth transition gameplay: Because of limited time and
the step-in-and-play concept, technical and social barriers
for the players need to be kept as low as possible.
Participant interaction: The boundaries between the
different player roles (actor, spectator, bystander) need to
be eliminated or blurred in order to foster interaction and
sociability.
Due to the public nature of Limelight, it seemed
problematic to limit or set the amount of people playing to a
definite number. The game should work with only one
person playing as Limus (increasing the difficulty due to a
greater degree of darkness) or with any given number of
Lumees and SMorcS. We tried to achieve this through
establishing a careful balance between each of the
characters features.
Limus tasks are to run and jump, avoid pitfalls and
SMorcS and find the three magical stones. In order to
survive in the dark cellar, the player controlling Limus
needs to see enough of the surrounding level layout to be
able to make controlled jumps between the platforms.
Lumees can fulfill the task of illuminating vital parts of the
level, but in order for the game to function with only a few
Lumees (or even none at all), Limus himself has to emit a
little bit of light. The quantity has to be enough to be able to
play to some degree, but still too little to easily master
everything alone.
This scenario allows the aforementioned Lumees to step in
andliterallyshed some light on the situation. Their light
radius also has to be carefully balanced in order to avoid
total illumination when a large number of Lumees is
present, yet still be helpful to Limus if only one or two
Lumees are in the scene. This kind of designed relationship
is meant to carefully amplify social interaction.
Finally, the number of SMorcS in the scene is also of
importance. Too many enemies make the game too hard to
beat, and having too few presents a very limited challenge
to the player controlling Limus. Therefore, when spawning
the SMorcS, the game tries to maintain a constant amount
of enemies in the scene by limiting their number to four and
reserving newly created enemies in a queue. This avoids an
accumulation of enemies but also ensures that the SMorc
DESVIG 2013
25
creators will see their monsters in action (although possibly
with some time delay). To ensure that some enemies are
present even if none of the spectators opt to spawn their
own SMorcS, the system randomly generates new foes in
the queue whenever it is empty.
Smooth Transition Gameplay
In order for people to start playing the game, potential
barriers need to be minimized as much as possible, creating
a compelling user experience and steep learning curve.
Such barriers can either be of a technical or social nature.
We define technical barriers as obstacles for the player that
occur due to unfamiliarity with computing systems or
hesitation due to unknown and complex interaction
modalities. Social barriers keep people from playing due to
deterrents such as social embarrassment [1].
To play as Limus, we opted for a classic 2D jump-and-run
style game played with a gamepad. This type of game as
well as the standard controller are usually familiar to most
people and require little or no explanation. To make the
control of the Lumees as easy as possible, we employed an
optical tracking system installed in Deep Space. When a
person enters the tracking area, a Lumee is automatically
created at the corresponding (but vertically inverted)
position in the game and given a unique color. When the
tracking area is left, the Lumee disappears again. There is
no need for in-game calibration, registration or any other
tedious processthe player can start right away. Figure 2
illustrates how Lumees are mapped to the positions of
individuals in the tracking area.
Spectators can create SMorcS simply by sending an SMS
text message or an email to a number or address visibly
positioned on additional displays. Both communication
technologies are well known and at least one of the two is
available on a standard mobile device. Finally, interacting
with the game using these methods mitigates the problem of
social embarrassment. Even people that would never play
Limus or enter the tracking area to control a Lumee are able
to anonymously contribute to the game. Furthermore, being
able to give their SMorc a personalized name, the players
receive clear feedback from the system without their
anonymity being compromised.
Another feature that reduces the potential for social
embarrassment is due to the nature of the game itself. As
the Deep Space room needs to be dark in order to view the
projections in the best possible quality, the reduced lighting
creates a kind of anonymity that further lowers barriers to
join the game.
Participant Interaction
Brignull and Rogers as well as Tang et al. describe clearly
drawn boundaries between the different user roles in a
public space [1, 8]. They also advocate a process that
supports bystanders to become spectators or even actors
more easily. Nevertheless, the potential of the different
roles interacting to achieve a common goal has not been
sufficiently explored. Through the aforementioned ease of
joining the game and the collaborative nature of its
gameplay, these roles are frequently blurred within
Limelight. This situation is clearly demonstrated by
spectators entering the secluded Deep Space in a passive
role. While watching players interact with the system and
then deciding to try out the interaction themselves,
spectators become actors, actively influencing the
gameplay. Once they decide to end their interaction, they
revert back to spectators, but are capable of taking on active
roles again at any time (e.g. as Lumees or SMorcS).
The encouragement to collaborate with other players and
help Limus to achieve his objective is another vital aspect
of the game. As stated before, Limus struggles when trying
to master the game only by himself. Therefore, a reliance
on the support of other players (the Lumees) is promoted.
Figure 2. A schematic view of the Deep Space room. The game is
shown on the wall projection surface whereas the floor is used
for tracking the individuals playing as Lumees (yellow). The
person playing Limus (blue) as well as bystanders (red) remain
outside of the tracking area.
RUNNING LIMELIGHT
Limelight was shown during the Ars Electronica Festival
for five consecutive hours (Figure 3 shows the game in
action). During this time, at least one of the developers was
always present to monitor the game technically and observe
people playing. Since this was the initial run (with the
exception of very short test runs), we therefore only present
observations made by the developers and offer possible
interpretations for certain behaviors.
Generally, we noticed a lot of interaction between the active
players (Limus and the various Lumees). This interaction
consisted mainly of cooperation in which Lumees would
help Limus without being advised by the player to do so.
This support mostly consisted of the Lumees hovering
around Limus current location to generate a large light
radius around the protagonist. In some cases, players would
also move away from Limus to explore the cellar by
themselves and look for the hidden stones. This, however,
DESVIG 2013
26
did not seem to result in the game being mastered more
quickly since the person playing Limus usually focused
only on her immediate surroundings, not noticing Lumees
that were further away. Only verbal statements such as
The red stone is in the upper right made the player aware
of the hidden stones. This however seemed to happen
mostly in groups that knew each other ([9] also confirms
this). Interestingly enough, only very few players
controlling Limus actually gave commands to fellow
Lumee players.
Another interesting discovery we made was how people
started to play as a Lumee. The first thing almost everyone
did was to enter the tracking area and simply play around
with their Lumee avatar. People would move in different
directions and watch how their avatar moved on the wall.
Othersespecially childrenwould come close to the wall
and actually try to touch their Lumee. While this happened,
the players did not focus on the actual gameplay, which
meant that they did not actively support Limus at all.
We also noticed that a number of visitors returned to Deep
Space after some time, either to simply re-play the game, or
choose a different role (e.g. Limus instead of a Lumee).
Some even came back in a group so they could experience
the game together. Sometimes users would also encourage
others to join the game (also mentioned in [2]), especially
when the number of players was rather low. On the other
hand, players did not limit their involvement to only one
round of the game, regardless of whether they played as
Limus or as a Lumee. Many players decided to just stay and
play again and, at times, even left during an active round of
the game.
The number of emails and text messages received to create
SMorcS was, unfortunately, rather low. A total of 61
messages from 11 different people were received. This
equals approximately one message (email or SMS) for
every five minutes of the game.

Figure 3. Limelight in action. Spectators watch the game from
outside the tracking area while actors engage in play as Lumees.
DISCUSSION
Our observations of people playing Limelight confirmed
statements made previously by other researchers. One of
the more obvious behaviors was grouping. Lumees, in
particular, would work together and try to help Limus
without being asked or instructed to. This might seem odd
at first, but it has previously been mentioned by [3, 6, 7]
that groups do not necessarily have to communicate in
order to form. The act of grouping is a social act in and of
itself. Strictly speaking, Limus and the Lumees might not
be seen as a group, however. While Limus has a clear goal
(finding the magic stones), the Lumees do not. They can
assist the wizard but do not have to. This raises the question
of exactly how sociability is created in this particular case.
Comparing this once again to the requirements stated in [3],
in-game and off-game communication can almost be ruled
out since both only rarely happened. The exact nature of
grouping may be unclear, as previously mentioned, yet it
most likely plays at least some role. Empathy is also only
minimally involved since there is no real common goal. Of
course, Lumees might adopt the goal of helping Limus for
themselves, but as we observed, a lot of players did not
probably because the game did not force them to. The
factor of world design can be disregarded, since there is no
level system included in the game. This leaves designed
relationships, which certainly are the crucial element here.
The game employs different roles that can function alone,
but work much better when put into context with others.
The previously mentioned observation of people coming
back to re-play the game adds another aspect to these
designed relationships. Since it was easily possible to
assume a different character (role), players could also
experience these social interactions from a different
perspective. This clearly distinguishes the concept of such a
public game with its aforementioned smooth transition
gameplay from standard MOGs. These games obviously
permit the changing of a player class, but this usually
means starting over from the beginning of the game.
Limelight offered players the possibility to instantly change
their role, even while a game was running. We feel that
being able to assume different roles and therefore
experience the whole range of possible social interactions is
a good way to increase or even create sociability in a game.
During the initial public exhibition of Limelight, a survey
was conducted using the Social Presence in Gaming
Questionnaire (SPGQ) [4]. However, it was not possible to
collect enough data for a reliable statistical analysis over
such a short period of time. Although we were able to learn
a great deal from observation during the exhibition, more
than observation data alone is required to adequately assess
our hypotheses. To obtain more accurate results, we plan on
running the installation over a longer period of time and
using a similar questionnaire with the addition of questions
related specifically to Limelight (e.g. choice of class, initial
impression, etc.) to gain insight into the players
motivations. We also intend to utilize video and audio
DESVIG 2013
27
material of the playing experience for further examination.
A detailed analysis of such data could then lead to a further
refinement of sociability requirements that more clearly
define the importance of different player roles in digital
games.
CONCLUSION
We developed Limelight to gain insight into sociability in a
co-located game context. We correlated recently published
guidelines for MOGs to the game design principles applied
to Limelight to determine if there was sufficient common
ground. Based on observations collected during an initial
test run, we concluded that these requirements do indeed
apply to co-located games, although not to their full extent.
Further research including the collection and analysis of
qualitative data is necessary to more closely examine our
approach.
ACKNOWLEDGMENTS
We would like to thank the students Gerald Hauzenberger,
Ren Ksuz, Thomas Peintner and Magdalena Soukup for
their efforts on Limelight as well as the Ars Electronica
Center for providing their facilities. We would also like to
thank the A1 Telekom Austria AG for providing us with
their SMS server solution.
REFERENCES
1. Brignull, H., and Rogers, Y. Enticing people to interact
with large public displays in public spaces. In
Proceedings of the IFIP International Conference on
Human-Computer Interaction, INTERACT 2003
(2003), 1724.
2. Cao, X., Massimi, M., and Balakrishnan, R. Flashlight
jigsaw: an exploratory study of an ad-hoc multi-player
game on public displays. In Proceedings of the 2008
ACM Conference on Computer Supported Cooperative
Work, CSCW 08, ACM (New York, NY, USA, 2008),
7786.
3. Christou, G., Lai-Chong Law, E., Zaphiris, P., and Ang,
C. S. Challenges of designing for sociability to enhance
player experience in massively multi-player online role
playing games. Behaviour & Information Technology
(2012), 134.
4. de Kort, Y. A. W., IJ sselsteijn, W. A., and Poels, K.
Digital games as social presence technology:
Development of the Social Presence in Gaming
Questionnaire (SPGQ). In Proceedings of PRESENCE
2007: The 10th International Workshop on Presence
(Barcelona, Spain, 2007), 195203.
5. Ducheneaut, N., Moore, R. J ., and Nickell, E. Virtual
third places: A case study of sociability in massively
multiplayer games. Computer Supported Cooperative
Work (CSCW) 16, 12 (April 2007), 129166.
6. Ducheneaut, N., Yee, N., Nickell, E., and Moore, R. J .
Alone together?: exploring the social dynamics of
massively multiplayer online games. In Proceedings of
the SIGCHI Conference on Human Factors in
Computing Systems, CHI 06, ACM (New York, NY,
USA, 2006), 407416.
7. Eklund, L., and J ohansson, M. Social play? A study of
social interaction in temporary group formation (PUG)
in World of Warcraft. In Experiencing games: Games,
Play, and Players: Proceedings of DiGRA Nordic 2010,
DiGRA (Stockholm, Sweden, 2010).
8. Finke, M., Tang, A., Leung, R., and Blackstock, M.
Lessons learned: game design for large public displays.
In Proceedings of the 3rd International Conference on
Digital Interactive Media in Entertainment and Arts,
DIMEA 08, ACM (New York, NY, USA, 2008),
2633.
9. Holopainen, J ., Lucero, A., Saarenp, H.,
Nummenmaa, T., El Ali, A., and J okela, T. Social and
privacy aspects of a system for collaborative public
expression. In Proceedings of the 8th International
Conference on Advances in Computer Entertainment
Technology, ACE 11, ACM (New York, NY, USA,
2011), 23:123:8.
10. Machaj, D., Andrews, C., and North, C. Co-located
many-player gaming on large high-resolution displays.
In Proceedings of the 2009 International Conference on
Computational Science and Engineering Volume 04,
CSE 09, IEEE Computer Society (Washington, DC,
USA, 2009), 697704.
11. Simmel, G. The sociology of sociability. American
J ournal of Sociology 55, 3 (November 1949), 254261.
12. Stenros, J ., Paavilainen, J ., and Myr, F. The many
faces of sociability and social play in games. In
Proceedings of the 13th International MindTrek
Conference: Everyday Life in the Ubiquitous Era,
MindTrek 09, ACM (New York, NY, USA, 2009),
8289.
13. Tyni, H., Sotamaa, O., and Toivonen, S. Howdy
pardner!: on free-to-play, sociability and rhythm design
in Frontierville. In Proceedings of the 15th International
Academic MindTrek Conference: Envisioning Future
Media Environments, MindTrek 11, ACM (New York,
NY, USA, 2011), 2229.

DESVIG 2013
28
Sociability in Virtual Citizen Science
Charlene J ennett
1
, Laure Kloetzer
2
, Margaret Gold
3
and Anna L. Cox
1

1
UCLIC, University College London, UK, WC1E 6BT
2
TECFA-FPSE, University of Geneva, CH-1211 Genve 4, Switzerland

3
The Mobile Collective, GMI Ltd, London, UK WiU 5J X
charlene.jennett, anna.cox @ucl.ac.uk
laure.kloezter@unige.ch; margaret@gold-mobileinnovation.co.uk

ABSTRACT
Past research suggests that sociability can enhance
volunteers experiences of virtual citizen science (VCS).
We define four types of sociability. We also describe how
outreach events - Thinkcamps can be used to support the
design of social tools for VCS platforms.
Author Keywords
Citizen Science; Sociability; Communities; Serious Games;
Design.
ACM Classification Keywords
H.1.2. User/Machine Systems: Human Factors.
General Terms
Human Factors.
INTRODUCTION
Virtual citizen science (VCS) is a unique form of computer-
mediated interaction, where members of the public
collaborate with professional scientists to conduct scientific
research [14]. Volunteers participate because they are
intrinsically motivated to contribute to a scientific project
by an interest in the topic e.g. astronomy [8], protein-
folding [7], brain-mapping [6], theoretical physics [17],
volunteer computing [1]. There are a wide variety of VCS
projects, all with their own research questions and
tasks/games for participation.
Similar to MMORPGs [3], sociability is thought to enhance
the experience of VCS volunteers. Many VCS projects
include social tools - such as forums, blogs and social
media to support collaboration between volunteers and
scientists [14]. Some VCS projects also involve team-play
and competition. For example, in Foldit [7], players
compete against one another in teams. The use of game
elements in non-game contexts (such as VCS) is referred to
as gamification [5] or games with a purpose [20].
However designing for sociability in VCS presents several
design challenges. As Crowston and Wiggins [4] explain:
[V]CS projects share characteristics with other kinds of
open communities, but with significant differences due to
their scientific goals. The increasing scale of [V]CS
projects suggests a need for additional research on
appropriate technology support for this mode of scientific
collaboration. Similarly, Newman [11] writes:
Additional features to support social interaction between
volunteers, project managers, and scientists are needed.
In our research we believe it is important to include
scientists and volunteers as part of the design process. First
we review previous research and define four types of
sociability in VCS. Then we describe our plans to run a
series of outreach events with scientists and volunteers
Thinkcamps to support the design of social tools in our
own VCS platforms.
PREVIOUS RESEARCH
Existing studies of VCS projects suggest that sociability is
one of several reasons why volunteers are motivated to take
part. Raddick et al. [13] surveyed volunteers of Galaxy Zoo
[8] and identified 12 motivation categories, including
Contribute (I am excited to contribute to original
scientific research).
Nov et al. [12] surveyed volunteers of Stardust@home [18],
and reported that collective motives (the importance
attributed to the projects goals) were most salient,
alongside intrinsic motives (enjoyment).
Rotman et al [16] surveyed and interviewed volunteers of
ecological Citizen Science projects. They found that
volunteers motivations were dynamic and changed over
time. Although initial motivations may be very diverse,
social factors played an important role in secondary
motivations and staying with a project.
DEFINING SOCIABILITY
It is evident that social factors are important for sustaining
volunteers engagement in VCS projects. However we
should define more precisely what we mean by sociability
in VCS. Based on our work so far (participant observations
and interviews with several VCS volunteers), we
hypothesize that there are four types of sociability in VCS.
Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.

DESVIG 2013
29
1. Sociability Among the Larger Scientific
Community
Most participants in our interviews report the feeling of
contributing to current research in the scientific community
as an important motivational factor for taking part in VCS
projects. Their initial participation is grounded in their
personal interest for science and further encouraged by the
presence of key actors in the scientific scene. For example,
one participant of Eyewire [6] said: "I have a lot of friends
who are neuroscientists so also the brain activity all that is
really interesting to me."
Open and quick feedback from the scientific team in the
forums is appreciated and may encourage increased
participation. For example, one participant in the
Test4Theory project [19] reported his joy of being able to
interact with one of the founding developers of the
platform: ()one of the creators of Boinc was here,
discussing in the forum, answering our questions and
taking into account our feedback, I thought wow this is nice
and very dynamic..."
Other tools and activities, like blog posts on scientific
progress and discoveries, virtual lab visits, co-publication
of scientific papers, may increase this feeling of
participation to the scientific adventure in a world-wide
community.
2. Sociability in the Gamers Community
Some participants also report the feeling of belonging to a
larger community of players. They have common problems
and objects to discuss.
Some tools strengthen this sense of belonging to a
community. For example, ranking systems induce
comparison to others and some awareness of who the best
players are. There are also tools to share work-in-progress -
showing your results to others in galleries, comparing your
work with the work of others, checking what others did, etc.
Forums are important tools for community expression and
management, as well as for seeking support and providing
advice. For example, one participant of Eyewire [6] said: "I
really like the forum, because they seemvery active. They
answer quickly(...) At the beginning I had some problems so
I went there and you can sense that there is a small
community, but active, so that's interesting."
3. Team Sociability
Some VCS projects - like FoldIt [7] - open the possibility to
play in teams. Team members collaborate to achieve their
goals (mainly in a competitive way with other teams) or
cumulate their points. They exchange ideas, strategies,
plans, programs, thoughts, scripts or recipes. One of our
hypotheses is that similar to previous research findings
for MMORPGs [3] - the feeling of belonging to a team is a
strong motivational factor for long-term participation in
VCS projects.
Teams may select players because they have already
demonstrated that they can achieve high scores. For
example, one FoldIt [7] participant said: "Most groups are
closed, most of the time we ask players to join when we see
that players have prestige." Teams also establish rules:
"()if someone is not behaving well, I can kick himoff the
group". In this case, team sociability is a kind of
collaboration to improve one's own scoring as well as team
scoring in the game.
In other VCS projects, team sociability may add a
competitive dimension to the core mechanisms of the game,
like on the Boinc projects [1], where volunteers chose to
create teams, based on language or country, to compete
with other teams. When teams are dedicated to progress or
competition, it is common for some volunteers to create
specific tools for evaluating and comparing performances;
advanced graphics or lists based on available scoring
systems, for example, are updated and studied to improve
future performance.
4. Sociability as Acti ve Community Management
Some volunteers take full responsibility for supporting and
animating the social life of those involved in the project.
These volunteers change roles to act as moderators or
experts in the online community. In these circumstances,
sustaining and developing the community becomes their
primary objective.
In some VCS communities, volunteers have created game
mechanisms in additional to those provided by the projects.
For example on the Boinc platform [1], some national teams
provide community-enhanced gamification for their
members, organising internal and external competitions:
"We organise raids, races, limited in time. We select a
number of projects to crunch, we contact the admin and
discuss with them, once we have their official agreement we
integrate these projects in our selection and launch the
raid". These communities, animated solely by players, are
by-products of the game.
There are also cases where volunteers themselves take
responsibility for maintaining the servers required to host
the activities of the community: "You transformyourself
into a small company when you do that!" explains one of
our participants.
Volunteers may also develop autonomous web sites or
wikis, and engage in active online and offline promotion of
the project: "()we prepared a tutorial to present volunteer
computing in universities, we answer people's most frequent
concerns and explain what it is about, how and why one
should do it". This sociability is an extreme form, as it
implies that a community of players does the support,
design and community management work usually done by
the professional scientific team.
DESVIG 2013
30
DESIGNING FOR SOCIABILITY
Overall it is evident that there is potential to encourage
many different kinds of sociability in VCS. Some
volunteers experience just one type of community feeling,
whereas others span the whole range of sociability. But
how can we design social tools to support these different
kinds of sociability?
In our research project, Citizen Cyberlab (CC), we will be
building and evaluating four new digital VCS projects [2].
The science domains include particle physics, disaster
mapping, synthetic biology, and community environmental
projects. We believe it is important to include scientists and
volunteers as part of the design process. Therefore a novel
way that we plan to support the design of sociability in our
CC platforms is by organizing a series of outreach events
called ThinkCamps. This is a new type of un-conference
format that aims to optimize creative problem solving,
where developers collaborate with users in order to design
tools to support their needs.
Traditional Un-conference Formats
Hackdays, Barcamps, and other Un-conference event
formats continue to grow in popularity as a creative outlet
for developers and a way for organisations to engage with a
wider community of participants than usually possible. The
format of an Un-conference typically starts with one or
more presentations about the subject matter of the event.
The rest of the content is suggested and delivered by the
participants. People then freely flow between the sessions
that are the most interesting and relevant to them. At a
Hackday (which focuses on creative innovation and
learning) or Hackathon (which focuses on moving a body
of code or a technology platform forward in a rapid burst of
collaborative development), participants form ad-hoc
collaborative teams around ideas that they would like to
work on together. These usually involve hands-on software
or website development, or hardware hacking, but can
also involve fleshing out a product, service or business idea.
At the end of Hackdays and Hackathons, there is usually a
series of demonstrations in which each group presents their
results. There is sometimes a contest element as well, in
which a panel of judges (usually peers and colleagues in the
field) select the winning teams, and prizes are given.
ThinkCamp Methodology
Citizen Cyberlab (CC) will progress beyond the state-of-
the-art of the varying Un-conference formats by applying
and further developing the hybrid ThinkCamp
methodology developed by The Mobile Collective.
ThinkCamps combine the improvisational creativity of the
Hackday, with the self-organising principles of Open Space
Technology (allowing participants to fully pursue their
passions and interests), and the more focused structure of
traditional idea-generation techniques [9]. This provides the
ideas that have been developed during the ThinkCamp with
a clear developmental path beyond the event itself, and
allows the adhoc teams which have formed a more formal
role in the delivery of those ideas after the event. The result
is a much more strongly engaged community that is
actively involved in delivering on a shared mission. The
ThinkCamp methodology also incorporates the inter-
disciplinary approaches to Open Innovation of the Fuzzy
Front End" of R&D [17], which optimises creative
problem-solving by taking the process outside the walls of a
singular organization [15].
In CC we will be organizing a series of real world
participatory ThinkCamp events for creative problem
solving and collaborative learning where scientists and
citizens can meet, share their experiences, devise new
projects for VCS, and further develop the community
toolkit. The experience of several of the partners in the CC
consortium shows that this kind of event directly benefits
community engagement and participation, provides an
opportunity for improvisational creativity and tacit learning,
and serves as a litmus test of software tools for VCS.
By involving scientists and citizens from the outset, we
hope that this will allow us to effectively design tools that
can best support their social needs. Potential design benefits
include:
Developers have the opportunity to engage with
real citizens to develop personas of typical users;
Developers will employ participatory design [10]
methods by conducting design activities with
citizens.
Citizens can be involved in building and
developing the tools.
Another possible outcome is that, due to their early
involvement in the projects, these citizens may be more
engaged and could potentially form part of the core project
community possibly even becoming the first group of
moderators/experts of the respective projects.
CONCLUSION
In this position paper we have presented four different types
of sociability in VCS projects. We have also described how
Thinkcamps can be used to support the design of social
tools for VCS platforms. This is a novel approach, which
we hope will allow us to design VCS platforms that are
successful in supporting sociability and volunteers feelings
of community membership.
ACKNOWLEDGMENTS
Charlene J ennett, Laure Kloetzer, Margaret Gold and Anna
L. Cox are all supported by the EU project Citizen Cyberlab
(Grant No 317705).
REFERENCES
1. BOINC. http://boinc.berkeley.edu/
2. Citizen Cyberlab. http://citizencyberlab.com/
DESVIG 2013
31
3. Christou, G., Law, E. L.-C., Zaphiris, P. and Ang, C. S.
Challenges of designing for sociability to enhance
player experience in massively multi-player online role
playing games. Behaviour and Info Technology (2012).
4. Crowston, K. and Wiggins, A. Supporting citizen
involvement in scientific research.
http://conway.isri.cmu.edu/hicss2011-sci-
workshop/crowston-hicss-supporting.pdf
5. Deterding, S., Dixon, D., Khaled, R. and Nacke, L.
From game design elements to gamefulness: Defining
gamification. Proc. MindTrek 11, ACM Press (2011),
9-15.
6. Eyewire. https://eyewire.org/
7. Foldit. http://fold.it/portal/
8. Galaxy Zoo. http://www.galaxyzoo.org/
9. Herring, S., J ones, B. and Bailey, B. Idea generation
techniques among creative professionals. Proc. HICSS
(2009).
10. Muller, M. J . Participatory design: The third space in
HCI. In J . A. J acko and A. Sears (Eds.), The Human-
Computer Interaction Handbook. L. Erlbaum Associates
Inc., Hillsdale, NJ , USA (2003), 1051-1068.
11. Newman, G., Wiggins, A., Crall, A., Graham, E.,
Newman, S. and Crowston, K. The future of citizen
science: Emerging technologies and shifting paradigms.
Front Ecol Environ 10, 6 (2012), 298-304.
12. Nov, O., Arazy, O. and Anderson, D. Dusting for
science: Motivation and participation of digital citizen
science volunteers. Proc. iConference 11, ACM Press
(2011), 68-74.
13. Raddick, M. J ., Bracey, G., Gay, P. L., Lintott, C. J .,
Murray, P., Schawinski, K., Szalay, A. S. and
Vandenberg, J . Galaxy Zoo: Exploring the motivations
of citizen science volunteers. Astronomy Education
Review 9 (2010).
14. Reed, J ., Rodriguez, W. and Rickhoff, A. A framework
for defining and describing key design features of
virtual citizen science projects. Proc. iConference 12,
ACM Press (2012), 623-625.
15. Rochford, L. Generating and screening new product
ideas. Indust Marketing Management 20, 4 (1991),
28702.
16. Rotman, D., Preece, J ., Hammock, J ., Procita, K.,
Hansen, D., Parr, C., Lewis, D. & J acobs, D. (2012).
Dynamic changes in motivation in collaborative citizen-
science projects. Proc. CSCW 12, ACM Press (2012),
217-226.
17. Rubinstein, A. At the front end of the R&D / innovation
process idea development and entrepreneurship. Int. J .
of Tech Management 9, 5-6 (1994), 652-677.
18. Stardust@home. http://stardustathome.ssl.berkeley.edu/
19. Test4Theory. http://lhcathome2.cern.ch/test4theory/
20. von Ahn, L. and Dabbish, L. Designing games with a
purpose. Comm. of the ACM 51, 8 (2008), 58-67.


DESVIG 2013
32
Tribal Metaphors in Social Game Design; Creating Conflict
and Camaraderie through Context
Ben Kirman
University of Lincoln
Lincoln, UK
bkirman@lincoln.ac.uk



ABSTRACT
This paper briefly explores the emergence of tribal patterns
in social play. The formation of community groups is not
just a result of game design, but a fundamental part of the
social nature of the species. Tribal effects, such as
favouritism towards fellow group members, have been
studied by social psychologists for decades (e.g. [22,15,7]).
We highlight some specific existing mechanics that enable
and encourage this kind of behaviour in social games, and
identify the importance of social feedback in generating
tribal feelings.
An experiment is described, that explores the minimal
conditions of tribal behaviour player groups in social
games. Specifically, it identifies the importance of feedback
on social context in supporting the emergence of in-group
favouring tribal play.
Author Keywords
Social games, sociability, game design, tribalism, tribal
play, social play
INTRODUCTION
The wider communities of online social games can involve
thousands, or even millions, of players, but these are not
always close-knit societies. Within the community of a
large social game, the players tend split into smaller groups.
These tribes can be defined formally by the game designers,
or informally by the players; they can be long-lasting
ventures involving hundreds of players and complicated
internal politics, or they can be associations of convenience
that exist for just a few hours. In any case, exploring the
behaviours of players as they join together in groups is
important in understanding social play.
Us and Them
A great deal of attention in social psychological research
has been given to investigating the behaviour of people in
groups in terms of theories of social identity. One of the
most robust, and replicated, findings in this field is that
when a person identifies themselves as a member of a social
group, this can lead to significant changes in their self-
perception, as well as changes in their behaviour both
towards other members of that social group as well as to
people who are not associated with that group [23].
Specifically, people show favouritism for members of the
same group (the in-group) ahead of non-members (the
out-group).
Making a distinction between ``us" and ``them" within a
larger community has a psychological effect that has
defined large-scale social movements since the dawn of
culture. Study of this group-favouring behaviour has a basis
in the anthropological study of ``Tribalism", where
members of social groups feel a strong sense of cultural
identity [21].
This form of social behaviour is not just historical -
academics in sociology note that ``Neo-Tribalism" is
observed in modern contexts, such as in neighbourhood
watch movements, youth subcultures and hobby clubs [15].
J oining such social groups has direct effects on self-
perception for the individuals who identify with these
groups.
In an online context, where identity is generally concealed,
it is intuitive to believe that more complex effects of group
identity aren't present - that somehow there is a blank slate
where people act more rationally. However this is not the
case. For example, new members joining online
communities actively change their behaviour to better
match the norms of their new tribe [13]. The SIDE model
(social identity model of deindividuation effects [18])
argues that social identity plays an important part in
determining behaviour, and it has been applied in
experimental studies of numerous online contexts [19, 20,
4] with varying degrees of anonymity. The development of
social identity based on groups appears to be one of the
most fundamental human social traits, and can be observed
in a large variety of contexts. It is not limited to developing
cultures or direct face-to-face contexts, but appears
wherever group members possess a strong feeling of
identity and loyalty to their group. It is therefore not
surprising that social identity and in-group bias is a
common feature of social play.

Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.

DESVIG 2013
33
TRIBALISM IN GAMES
Despite the history of tribalism being littered with dire
warnings from extremist abuses, it can also lead to positive
effects. It has been experimentally demonstrated (e.g. [2])
that competition between groups, even when randomly
selected, leads to better overall coordination and efficiency
in tasks when compared with individual efforts. With
restraint and a strong emphasis on perspective, tribalism
can be even be a source of fun!
Many large-scale social games already use tribal metaphors
to great success. The MMOGs Dark Age of Camelot,
Anarchy Online and World of Warcraft all formally split the
player-base into pre-determined competing factions as part
of the wider narrative - J ust like the athletes of different
nationalities compete as part of the FIFA World Cup. Eve
Online, Shadowbane and Age of Conan also have larger
wars as part of their narratives, but the tribes are more
informal - they are defined and organised by the players
rather than the developers.
Tribalism brings people together- it can give feelings of
identity and belonging, and makes `friends' out of strangers.
Within games, the psychological drives of tribalism create
opportunities for exciting forms of play, and within a safe
social environment.
Social Architectures for Motivating Tribalism
The mechanics of a game form a social architecture that
greatly affects the patterns of social behaviours within that
game [11]. Careful design of social mechanics can be used
to encourage, or discourage, specific social behaviours such
as tribalism. At its simplest form, players can be
encouraged to form groups by creating challenges that are
unachievable, or extremely difficult, for individuals to
manage alone. It is then natural for players to cooperate in
small, temporary groups in order to meet the challenge
(presuming the reward is suitable).
Although tweaks to the central mechanisms of the game
(e.g. murdering wildlife for experience points) can
encourage group behaviours in games, for players to feel a
part of a tribe, the architecture must support ways for
players to organise at a larger scale. There needs to be a
way for players to self-organise into us-es and thems.
In designing online communities, Amy J o Kim highlights
the potential benefits of allowing users to form groups [8],
but to flourish, social groups must be given spaces, tools
and support from the architecture of the system beneath.
The established standard in larger social games is in the
support of team structures usually known as guilds or
clans. Players don't need to have formal architectures to
support team play, for example FPS games have a long
history of self-organised teams, leagues and tournaments,
however, the architecture can be designed to reduce the
effort needed by individuals to maintain such systems
(creating websites, moderating IRC channels...).
Formal in-game tribal mechanisms mean the developers can
include useful social features guild chat allows players
to talk to fellow guild members at any time regardless of
distance. Shared property can give tribes a presence in the
game world, and guild inventories allow tribes to own and
share equipment. In their study of grouping behaviours in
World of Warcraft, Nicolas Ducheneaut and colleagues find
[5] that players using these social mechanics (i.e. being a
member of an in-game guild) increase the amount of groups
that they are involved with during play, the rate of
advancement in the game and also increase the amount of
time spent playing. Identification within groups such as
guilds is an important source of social value within games
[25].
Feelings of tribal identity are commonly further supported
by allowing players to publicly display their affiliation.
This might be by having a badge in the player profile, or
even appending a guild abbreviation to names. In order to
create a them, systems of inter-tribe competition are
created. These can be directly competitive such as in team
PvP competitions and guild ranking systems, or indirectly
competitive through social means - guilds acting as a social
group, working together on the more difficult group
challenges ahead of other tribes. More complicated tribal
systems can have more in common with governments than
guilds. In Eve Online, a notoriously cutthroat massively
multiplayer sci-fi space opera, the social structure for tribes
is through complex corporations. Eve supports player-
controlled craft up to the equivalent of naval capital ships.
However, these craft are far too expensive for individuals to
ever manufacture or purchase, so are essentially limited to
corporation level play - purchased through the taxation of
the players in that corporation. Corporations are engaged in
elaborate webs of espionage and counter-espionage, also
dabbling in intelligence and propaganda. The political
system in Eve is so complicated, the players have
essentially formed some of the first functioning virtual
nation-states, complete with the equivalent of dictators,
armed forces and a repressed proletariat.
Socio-contextual Feedback and Social Identity
In order for game design to support tribalism, it is important
that the players have a feeling of social identity. The social
architecture must therefore reinforce and enhance feelings
of community, by specifically demonstrating to players that
they are in a community. The design of online social
systems specifically has complications of social context.
For example, when studying face-to-face and online social
interactions, Adrianson and Hjelmquist [1] found that
typically face-to-face communication features more
complex negotiation and diplomacy. However, when
mediated by the Internet, it showed a greater reliance on
extended information sharing and greater amounts of data
exchange.
The difference of context means that non-verbal aspects are
often lacking in mediated communication. Efforts to
DESVIG 2013
34
increase the richness of communication online are generally
based around making the implicit social factors in face-to-
face communications more explicit in a virtual context.
Researchers in Computer-mediated communication (CMC)
have conducted a great deal of work on replicating typical
non-verbal aspects of face-to-face communication in
computer mediated social environments and also
introducing new non-verbal cues in a manner that is
respectful of the context in which these interactions take
place (e.g. [6]). One of the major strategies for increasing
the richness of social presence has been by exposing the
underlying social behaviours of group members, and
making implicit factors in computer-mediated
communication explicit. For instance, applications have
been created that expose social network information [17] as
Socio-Contextual information (data that makes hidden
information about the social context of individuals more
visible) to participants within group working scenarios [9,
3, 14].
By exposing these implicit relationships to the players, the
game designer can use these same principles to trigger
changes in their social behaviour. For example, in a
controlled study of cooperation within a treasure hunting
game, Francesco Martino and colleagues [16] found that
displaying socio-contextual feedback in the form of Social
Network Analysis (SNA) statistics to players had a positive
effect on increasing group ability to cooperate and engage
with the game objectives.
RANDOM TRIBES AND MINIMAL PLAYER GROUPS
Given that there is significant evidence showing the links
between socio-contextual feedback and generally social
behaviour in games, we were interested in the potential of
generating broader tribal behaviour with the same approach.
Specifically, we were interested to see the effect of putting
people into randomly determined tribes actually
generated social behaviour typical of these kind of social
groupings, and the role of socio-contextual information on
that behaviour. In this case, we use in-group favouring as
the metric for tribal behaviour. By collecting information
about each social interaction within the game, we can
determine if the players show any preference for interacting
with members of their own tribe, even where there is no
pre-existing relationship tying those players together.
PASION Fruit
The game PASION Fruit is a multiplayer fruit trading game
that was released openly on the web. When registering to
play the game, the system automatically joins the player to
one of ten random groups as they register for the game.
The game itself was based on trading several varieties of
fruit. Each player grows a fruit garden that natively holds a
few varieties of fruit. Players could send one another gifts
of fruit trees to grow and collect, and points were awarded
based on the variety of fruit collected and reduced by the
environmental impact of the transaction (based on real-
world distance between players). Importantly, in the game
design, although cooperation and negotiation were required
for personal achievements (i.e. higher scores), this was
explicitly separate from the mechanics of group
membership. There was no in-game benefit for choosing to
interact with a fellow group member ahead of any other
player. Similarly, there was no restriction on the choice of
possible recipient. Players were able to give gifts to any
member of the wider player-base, regardless of location or
group membership.
In the game, each group had a screen that showed members
of the group as well as context information about the social
behaviour of group members. This included social network
indices and a social network visualisation to encourage
social activity.
In order to see the effects of social context on the tribal
behaviour, a control version of the game was also
implemented that was identical in every way to the original
except for the name Fruit Loot) and the lack of social
context information for group members. In this condition,
players were assigned groups randomly in the same fashion
except the group information pages simply showed a list of
fellow members. The point of the control was to be able to
identify the difference that social feedback has on in-group
favouring behaviours.
Both games were released to the public simultaneously, and
a seed group of players were recruited from a pool of
volunteers sourced online and through prior contact with
the overarching project. The game was open to the public
and players were encouraged to invite friends into the
game, therefore creating a snowball sample of players
with interest in the game.

Figure 1 - PASI ON Fruit Garden

DESVIG 2013
35
Results
After a period of 11 weeks, data was gathered to evaluate
the effect of socio-contextual feedback on tribal behaviour
in games. In this experiment, the mean number of events
(fruit exchanged or received) by each participant was 13.4
in PASION Fruit, and 5.83 in the control condition. Means
showed a larger activity (and therefore higher engagement)
for participants involved in the condition with socio-
contextual feedback about group activity. This finding
supports existing findings that show the positive effect of
such feedback on engagement with social games.
Analysis of the interactions in both conditions showed that
the cumulative function of player activity between users in
both conditions followed similar heavy-tailed distributions,
which is normal for social activity in games [11, 12].
The similarity in distribution shows that the macroscopic
patterns of social interactions were close across both
conditions. Players at similar levels of activity interacted
with a similar number of co-players. Importantly, this
confirms that the similarity of the social architecture
between conditions. Despite having different players, the
mechanics of both games resulted in similar patterns of
social interactions. Any difference in the choice of co-
player is therefore due to the experimental variables and has
not been confounded by external factors.
In-group Bias
By comparing the volume of social interactions between
members of the same group, a picture of in-group bias
emerges. In this experiment, presuming players chose
recipients strictly randomly and given the groups are of
equivalent size; the expected in-group bias would tend
towards 10% based on chance (since there are 10 groups in
both conditions, of the same size). However, in the social
condition, the mean value for bias across the user-base was
23.3% compared to the control condition with 5.97%.
Figure 3 shows a comparison of the two conditions based
on the number of gifts a user sent to members of their own
group (in-group) and other players (out-group). The lines
show the expected split of in/out-group partnerships based
on random player choice (i.e. 10% in-group), so points
above that line represent players that favoured group
members disproportionately. As can be seen, many players
in the social feedback condition favoured the in-group more
than would be expected at random. Comparing the
proportion of in-group bias using an unpaired t-test
(presuming player bias, calculated as in-group interactions
over out-degree, follows a Gaussian distribution), the
condition with social feedback showed greater in-group
favouritism with p <0.01. Therefore, the data gathered
about actual player behaviour gives support to the
argument that showing players this level of socio-
contextual feedback regarding group membership results in
players showing disproportionate bias towards interacting
with group members.
Social Identity
In a post-trial questionnaire, users in the social condition
responded to questions about their membership of the
groups. Players reported mixed opinions about group
identity asked for their agreement with the statement ``I
felt a part of my group", 36.3% agreed or strongly agreed,
and 39.3% disagreed or strongly disagreed; the rest being
neutral. In response to the statement ``I was more likely to
give gifts to members of my group", 57.6% of users stated
they agreed or strongly agreed; compared to 27.3% who
disagreed or strongly disagreed. Players were asked to
select how important various factors were in choosing
whom to send a gift. In response to ``Membership of your
group", 54.5% of users thought it was somewhat, or very,
important compared to 24.1% who thought it was
somewhat or very unimportant. The responses to the
questions about group identity were mixed, however the
analysis of server logs shows that when choosing a player
to whom a gift will be sent, there was a higher probability
to find that players would choose fellow group members in
Figure 3 - I n-group favouring behaviour in PASI ON Fruit


Figure 2 - Social feedback within groups

DESVIG 2013
36
the social condition when compared to the control
condition. This reflects the non-intuitive aspect of the
minimal group paradigm. Individuals may assume, looking
back, that their choices of interaction partners were based
on rational and measured decisions (e.g. interacting with the
most suitable player for personal gain, regardless of group).
However the reality of the activities as exposed by the
quantitative data in the server logs highlights the cognitive
bias at play in social systems that use tribal metaphors.
DISCUSSION
Tools of Tribalism and social identity are frequently used
for negative purposes: as dictators have known for
centuries, creating divisions and artificial groups leads to
measurably greater engagement and fervour for a cause. In
other words, it is used because it works. This paper explores
the application of these techniques to discover if the same
principles apply in the context of the design of mechanics
that drive social games.
Many social games already use tribal metaphors such as
guilds, clans and factions to great success, and those
mechanisms appear to directly convert into increased
engagement of players with the game. In experimental
conditions, research has shown that socio-contextual
feedback that augment a game with explicit feedback about
implicit social factors, directly support the ability of groups
of players to collaborate on problem solving [17] and
increase general social engagement within games [10].
This paper introduced an experiment to understand the
social behaviour of players within a tribal metaphor. In the
social game PASION Fruit, players were randomly placed
in groups, which showed some social feedback in the form
of social network visualisations. A control version of the
game was also released in the same format, except without
the socio-contextual feedback.
The results of the experiment show that the broader social
behaviours in both games were comparable (similar
patterns of interaction were observed), however players in
the social condition exhibited a greater degree of in-group
favouring behaviour than in the control.
This means that the tribal mechanic of placing users in
groups only resulted in significant tribal effects where there
was sufficient contextual information displayed to those
players. This finding echoes that of Tajfel and Turner [24]
in non-game social groupings. They identified that
prevailing context was one of several factors that
contributed to the emergence of in-group favouritism. In the
context of online social games, the key factor appears to be
that group members require a minimum level of social
feedback in order to trigger these feelings of social identity
and to in-group bias in their actions. In other words, it is
important to make it clear to the users that there is a
meaningful distinction between the groups. Social gaming
applications must carefully design the social experience
with consideration to social feedback mechanisms. In order
to encourage group behaviour, a minimum level of
feedback is required - in this example, social network
visualisations were used but other feedback mechanisms
may also be effective. The emergence of in-group bias, in a
system with enough social feedback, is directly measurable
in the activities of the users, even if they do not strongly
recognise this behaviour in self-reflection of their actions.
Even when tribe membership is determined completely
randomly, with the addition of social feedback, players
show preferences for social play within their own tribes.
ACKNOWLEDGMENTS
This work is part of the PASION Project, which is funded
under the Presence II Initiative in the Future Emerging
Technologies within the European Framework VI
Programme
REFERENCES
1. Adrianson, L. & Hjelmquist, E. Group processes in
solving two problems: face-to-face and computer-
mediated communication. Behaviour & Information
Technology, 1999, 18:3
2. Bornstein, G., Gneezy, U. & Nagel, R. The effect of
intergroup competition on group coordination: an
experimental study. Games and Economic Behavior,
2002, pages 1-25
3. DiMicco, J . M., Pandolfo, A. & Bender, W. Influencing
group participation with a shared display. Proceedings
of the 2004 ACM conference on Computer supported
cooperative work (CSCW), ACM, 2004
4. Douglas, K. & McGarty, C. Identifiability and self-
presentation: Computer-mediated communication and
intergroup interaction. British J ournal of Social
Psychology, 2001, 40:18, pages 399-416
5. Ducheneaut, N., Yee, N., Nickell, E. & Moore, R. J .
"Alone together?": exploring the social dynamics of
massively multiplayer online games. In proc: SIGCHI
conference on Human Factors in computing systems,
ACM Press, 2006, pages 407-416
6. Erickson, T. & Kellogg, W. A. Social translucence: an
approach to designing systems that support social
processes ACM Transactions on Computer-Human
Interaction (TOCHI), ACM, 2000
7. Gladwell, M. The Tipping Point: How Little Things Can
Make a Big Difference. Abacus, 2002
8. Kim, A. J . Community building on the web: Secret
strategies for successful online communities. Peachpit
Press, 2000
9. Kim, T., Chang, A., Holland, L. & Pentland, A. Meeting
mediator: enhancing group collaboration using
sociometric feedback. in Proceedings of CSCW, 2008
10. Kirman, B., Lawson, S., Linehan, C., Martino, F.,
Gamberini, L. & Gaggioli, A. Improving social game
engagement on Facebook through enhanced socio-
DESVIG 2013
37
contextual information. In proc: international
conference on Human factors in computing systems
(CHI), ACM Press, 2010, pages 1753-1756
11. Kirman, B., Collov, F., Davide, F., Ferrari, E.,
Freeman, J ., Lawson, S., Linehan, C. and Ravaja, N.
Social Architecture and the Emergence of Power Laws
in Online Social Games, Proceedings of Think. Design.
Play., the 5th DiGRA conference on games and play,
2011, Utrecht
12. Kirman, B., Lawson, S. & Linehan, C. Gaming on and
off the Social Graph: The Social Structure of Facebook
Games International Conference on Social Computing
symposiumon social computing applications, IEEE
2009
13. Lampe, C. & J ohnston, E. Follow the (slash) dot: effects
of feedback on new members in an online community
GROUP '05: Proceedings of the 2005 international
ACM SIGGROUP conference on Supporting group
work, ACM, 2005, pages 11-20
14. Madan, A. & Pentland, A. VibeFones: Socially Aware
Mobile Phones. 10th IEEE International Symposiumon
Wearable Computers, 2006
15. Maffesoli, M. The Time of the Tribes: The Decline of
Individualismin Mass Society. Sage Publications,
London, 1996
16. Martino, F., Ba, R., Spagnolli, A. & Gamberini, L.
Presence in the age of social networks: augmenting
mediated environments with feedback on group activity
Virtual Reality, Springer, 2009
17. Martino, F., Miotto, A., Davide, F. & Gamberini, L.
Exploring Social Network Indices As Cues To Augment
Communication and to Improve Social Practices. In
Proc: Workshop on Maps Based Interaction in Social
Networks (MapISNet '07), 2007
18. Postmes, T., Spears, R. & Lea, M. Breaching or
Building Social Boundaries? Communication research,
Sage Publications, 1998
19. Spears, R. & Lea, M. Panacea or Panopticon?
Communication Research, Sage Publications, 1994
20. Spears, R., Lea, M. & Lee, S. De-individuation and
group polarization in computer-mediated
communication. British J ournal of Social Psychology,
British Psychological Society, 1990
21. Sumner, W. G. Folkways: A Study of the Sociological
Importance of Usages, Manners, Customs, Mores and
Morals. Ginn and Company, 1902
22. Tajfel, H. Social Psychology of Intergroup Relations.
Annual Review of Psychology, 1982, pages 1-39
23. Tajfel, H., Billig, M., Bundy, R. & Flament, C. Social
categorization and intergroup behaviour. European
J ournal of Social Psychology, Wiley, 1971, pp 149-178
24. Tajfel, H. & Turner, J . C. An integrative theory of
intergroup conflict. The social psychology of intergroup
relations, 1979
25. Williams, D., Ducheneaut, N., Xiong, L., Yee, N. &
Nickell, E. From Tree House to Barracks: The Social
Life of Guilds in World of Warcraft. Games and Culture
1:4, 2006


DESVIG 2013
38
Social Player Analytics in a Facebook Health Game
Matthias Klauser
University of Ontario Institute
of Technology, Oshawa, ON
matthias.klauser@uoit.ca
Lennart E. Nacke
University of Ontario Institute
of Technology, Oshawa, ON
lennart.nacke@acm.org
Paul Prescod
Ayogo Games
Vancouver, BC
paul@ayogo.com


ABSTRACT
Social health games can drive healthy real-life behaviour
through motivating game engagement. There is a lack of
game design research on which social interactions in a
game help players stay engaged and how to quantify
socially-engaging gameplay behaviour based on player
interactions. We address this gap by presenting a
quantitative study of player behaviour in a social health
game called HealthSeeker (developed by Ayogo). This
Facebook game targets people with diabetes to help them
manage health goals in real life by integrating health
missions into the social game. We studied which functions
led to more gameplay success and how the behaviour of
successful players differs from unsuccessful players. Our
results show that functions aiming at social interactions
motivate players to solve more missions, to fulfill more
healthy goals and to play the game much longer. We
conclude that having a well-connected social network can
improve player success in solving healthy game missions
and to live a healthier life.
Author Keywords
Game Design; Social Games; Game Metrics; Facebook
Games; Sociability; Online Games; Health Games; Game
Analytics; Behaviour Change; Games User Research
ACM Classification Keywords
K.8.0 [General]: Games.
INTRODUCTION
Researchers try to understand why games are fun and how
players behave in different situation. Especially companies
such as Zynga or BigPoint with freemium games based on
inner purchases have a huge interest in who their players
are and how and why they play the game. These types of
games are popular social games for a casual target group on
social network sites and are becoming more popular in
digital applications.
Facebook itself as a community has almost 1 Billion users
and is probably the biggest social Internet community.
Besides casual fun games, games with a purpose
(GWAP) are trying to use the Facebook community to
support users in achieving their goals. People are
connecting on these social network sites (SNS) for different
reasons: For example, for work, to find relationships or to
share interests [10].
In the past, social connections and (actual or virtual)
resources have been positively linked with different health
benefits [10]. Since browser and Internet technology are
becoming more powerful, those SNSs are including games
and sub communities, which we are called Social Network
Games (SNG). SNGs are often browser-based online games
such as Farmville or DiamondDash, which can be accessed
and played via social networks like Facebook. While most
SNGs are casual games, some of these games are trying to
establish smaller independent communities to increase the
intensity of bonding, discussions or health support (in the
case of GWAP).
HealthSeeker is a social health GWAP on Facebook
developed by the company Ayogo-Games. Within the
game, players or users create a social community and solve
healthy missions in daily life. They report these missions
back into the game when they are accomplished. For each
mission, a player earns experience and achievements. Each
mission has a different life goal and involves the player or
his friends and family. The game also provides a support
system called Kudos with which players can show support
to one anothers missions. Also a player wall called fridge
door offers the possibility to report problems, thoughts or
status updates. It is possible for a player to challenge other
players by sending health challenges. Since each mission
aims at actions done in real life, they are directly connected
to a users healthy behaviour.
In this paper, we will analyse HealthSeekers key gameplay
metrics and report how we used them to determine
sociability measures for effective health game design. We
analyzed the influence of social game features on the in-
game success of players and on improving healthier living.
RELATED WORK
Game Metrics and Game Analytics
The analysis of recorded data or events for analysing player
behaviour is a common games user research method [13,
14,15]. The use of automated systems to track player
behaviour in complex entertainment environments is still at
its beginning. Tychsen et al. define Game Metrics as
numerical data obtained from the user interaction with the
game software itself, and that data can be recorded at
different degrees of temporal and spatial resolution [14].
The limitation of this data is that it only contains
information provided by the game itself, not from the
player or their environment [14]. Researchers and
Copyright is held by the author/owner(s).
DESVIG 2013
39
companies have developed methods to collect and use
automated events for different purposes. For example, the
Microsoft Game Studios used automated recorded metrics
to evaluate game design and user experience for the game
Halo 3 by visualizing player behaviour [15]. Canossa and
Tychsen used tracked events and other metrics (e.g., player
death and position) to detect frustration in the game Kane
& Lynch 2 [17] or to develop play personas in Hitman
Blood Money [14]. Tychsen et al. defined play personas,
play styles and play modes as a part of an overall player
strategy description, which can be used for design decisions
in games.
Also, user behaviour metrics are used to analyse interfaces
through click behaviour [18] and testing game features
[19]. Many methods, which use game metrical analysis are
also including data visualization [14,15,17]. Log files are
used to analyse user chat logs via statistical analysis or data
mining. Ducheneaut et al. used recorded chat protocols of a
massively multiplayer online role-playing game
(MMORPG) to analyse social interactions. Public data
available in the game was used and quantitative metrics
were created for this analysis, such as amount of gestures
or signal to noise ratio [16]. In addition, the project PlayOn
focused on the social interaction in MMORPGs by
analyzing levelling behaviour from a top-down perspective.
Social Network Games (SNG)
In the HCI community, not much research has been done
on SNGs using player behaviour game analytics, although
companies such as Zynga track different user population
metrics for marketing purposes. SNGs make up a
community in which players and their information are
represented within a social graph which represents social
entities and their relationships [20].
Sung et al. [11] report results of a questionnaire launched
via Facebook with 253 participants about personality
factors and playing behaviour in SNGs. They found that
people with lower self-efficacy spend more time playing
SNGs. Correlations to playing frequency or customization
of characters were also reported. Finally, female players
gift more items to their friends than male players.
Games for Health
Digital games usually have a high level of interactivity and
are possibly able to close the gap between applications such
as wellness diaries and motivation trainers. A GWAP or a
serious game is often used for teaching knowledge or
training skills [8]. It includes health applications [12].
One Area of health games is to help people fulfilling their
goal by supporting behaviour change to a healthier life
style. The game OrderUp [9] for example, is a mobile
game for teaching adults how to make better meal choices.
The player takes the role of a waiter and their task is to
keep guests healthy by helping them to find the healthiest
meal from some choices. The reported study observed 12
Participants playing the game for three weeks. Within this
time, participants wrote short diary entries and answered
questionnaires about the gaming and food habits. Playing
the game OrderUp encouraged the participants to think
more about their ordering behaviour and to choose healthier
meals based on gained knowledge [9].
Another aspect of behaviour change is often the social
context through support groups, family and friends or other
people in the same condition [4]. The connection and
influence of the interactivity of a game and the social
influences on the success of behaviour changes with a
social health game application is not yet fully understood.
The game HealthSeeker was developed to use the social
support from the Facebook gaming community, especially
from friends. Besides posting results on Facebook walls,
players can interact with other HealthSeeker players and
show direct support for each other, through the Kudos
function. We had access to a player data subset but no
influence on data collection. Similar to Duchenaut [16], we
defined metrics for the social browser game and developed
definitions for player behaviour in the game.
METHOD
Measurements
Our first goal was to identify and create metrics, which we
consider as important because the HealthSeeker database
had no user logs or behaviour metrics. We wanted to
identify metrics that can be created by just looking at the
database and are important for the analysis of the game.
Together with the company and based on literature, we
created three different groups of metrics: success,
connectedness/social and virality. We used the information
of the HealthSeeker database to create variables for these
three groups. All these metrics were created by our
software and are based on parsing and filtering of the
HealthSeeker dataset. Therefore, we defined different
actions as separate activities with a time and necessary
information about the activity (e.g., writing a message,
commenting on a post, starting/finishing a mission). To
identify a valid player, we created a definition together with
the company and game designers of HealthSeeker:
Connectedness/Social Metrics. We define social metrics
within the game as variables which measure player actions
which leads to interactions with other players.
Success Metrics. The success metrics should express the
success a player has within the game. For example: the
number of completed missions or challenges. This category
includes all actions a player made during the game, which
we associated with success in the game.
Viral Metrics. The number of sent invitations is already
defined in our success metrics but is also an interesting
A valid player is a player who has
done at least two different actions in
the game on at least two different
days.
Definition:
Valid
Player
DESVIG 2013
40
value on its own. This category includes all actions
associated with virality.
Other
Other than success and viral metrics, we also identified
several descriptive variables, which should represent the
player and the behaviour in the game in general. All actions
during the game, which are not related to success and
virality, but include information about player behaviour.
Hypotheses
To analyze the data from the game HealthSeeker, we
formulated the following hypotheses toward predicting user
success within the game.
Assumption: The received sample size represents a group
of players/users who are well connected within the game.
H1: The number of friends has a strong influence on the
behaviour within the game.
H2: Users who are socially active become more involved
(successful) within the game
Explorative: Besides these hypotheses our analysis tries to
find relationships between the social and success variables
to find important implications for the companys future
games. Therefore, we will try to find specific player
behaviour using the recorded metrics in the game.
RESULTS
Population
To define our population, we searched for a definition to
separate active from inactive users. The definition should
help us find users who actually play the game. This method
excludes test accounts and other non-real accounts.
Facebook provides developers with a definition for active
users which they use in the statistics overview for
Facebook applications. For Facebook, people are
considered active users if they interact with the app in
any way or if they have the webpage with the app in focus.
The structure and concept of HealthSeeker is different to
other games. Facebook games encourage users to come
online every day (or more often), but HealthSeeker
missions require actions only once every couple of days.
So, a highly active user might be online less than once
every other day.
For this study, we got a subsample of valid users in the
game which we could use for analysis. The data was
collected by the company as a part of the HealthSeeker
game. In total our subsample size of valid users is 803 (N =
803, 532 female, 202 male, 69 without gender information)
registered user accounts. The most accounts were created
between May, 2010 and September, 2011.
Within the game, a total of 9358 missions were started and
3423 missions were completed. Players sent 1012 wall
posts. Almost 1000 kudos have been sent and players
challenged one another with healthy missions 94 times.
Statistical Analysis
With the knowledge of the user base connectedness, we
used statistical methods to analyze the user base. We
focused on the social factors in our analysis. Besides the
two hypotheses, based on the companys previous findings,
we used an exploratory approach to analyze the dataset step
by step to find important aspects between the social
behaviour and successful behaviour change.

Figure 1: Average "number of friends" and "completed
missions.
According to our first hypothesis the connectedness of a
player has a strong influence on their motivation or
behaviour in the game. We conducted a correlation analysis
that supports the connection between having friends and
success in the game. The effect gets stronger by looking at
the number of friends instead of category of just having
friends. Both effects are highly significant but only
moderate strong. Further the number of successful
accomplished missions seems to increase positively with
the number of friends (F (27,775) = 3.20; p < .001; =
.10; = .26). This indicates that having friends could be
responsible for at least a small amount of the motivation to
succeed in the game. Thus, our data supports that an active
player with friends solves twice as many missions as a
player without friends (with friends: N = 458; M = 3.15, SD
= .329 | without friends: N = 345; M = 1.74; SD = .294).
When we look at the amount of friends, we found that
people with more than 10 Friends solve even more
missions (more than 10 friends: M = 47; M = 6.32; SD =
1.512) and stay longer in the game than other users (less
than 10 friends: M = 2.3 SD = .2; more than 10 friends: M =
100.2 SD = 21.1).
Social Interactions Success in the Game
Our second hypothesis connects social activity with
motivation to succeed in the game and Table 1 shows that
this assumption has stronger effects than the number of
friends. While the social activities for sending challenges,
kudos or wall posts have strong effects, the receiving social
actions seem to have a good correlation, on the success
metrics (i.e., completed missions). A one-way analysis of
variance (ANOVA) shows highly significant results for
sending challenges (F (1, 775) = 152,043; p < 0.001; =
.14; = .372) and sending Kudos (F (27,780) = 13.799; p <
.001; = .283; = .230; = .456) for active social
actions. This shows that players who use the social
0
5
10
15
20
25
30
1
.
7
2
.
1
2
.
9
5
.
4
1
.
9
1
.
7
1
2
.
5
1
.
0
1
0
.
8
3
.
5
6
.
0
1
.
0
1
.
5
9
.
0
2
.
5
DESVIG 2013
41
functions of the game actively are more motivated than
players who do not. The receiving actions (e.g., Kudos)
show a good significant effect on the motivation to be
successful in the game (F (27,780) = 13.799; p < .001; =
.283; = .230; = .456).
Both, the effect of active and passive social interactions on
the motivation to be successful in the game underline the
importance of social interactions in a game and their power
to support the goal to for a behaviour change to a healthier
lifestyle. We could identify that both passive and active
social interaction metrics seem to have a highly significant
and strong effect on the user motivation in the game.
Therefore we declared the following variables as important
(with the strongest effects): send/receive challenges;
send/receive Kudos; wall-posts; comments;
To further describe the effect and analyze the impact of the
single variables regarding passive and active social actions
regression analysis were used for each of the important
variables. Also each variable had a fairly strong effect on
the success in the game but the social interaction between
players will hardly be describable with a single variable.
Therefore we focused on the combination of the found
strong effects to discover a model which explains combined
effect of social behaviour on motivation to succeed.
Model 1 and 2 show models for the combined active and
passive social actions recorded for each user. Model 1
includes strong variables for social active behaviour and
especially the number of posted messages on the players
wall has a big influence on the completed missions.
numbcr0ComplctcJHissions = 1.Su7 + u.S11
Numbcr0ScnJKuJos + u.198
Numbcr0ScnJCbollcngcs + 1.984 Numbcr0wollposts
Model 1: Model for send social activities
influence of social active behaviour on success in HealthSeeker
(r = .282 p < .001
( = .145 for step 1 (p < .001), = .106 (.251) (p < .001); p < .001)
The second model contains the combined results from
recorded passive social behaviour. Here the comments
seem to stick out of the model and have an important
influence on the completed missions.
numbcr0ComplctcJHissions
= u.2u6 + u.2u9 Numbcr0Rccci:cJCbollcngcs +u.4S9
Numbcr0Rccci:cJKuJos + 2.416
Numbcr0Rccci:cJCommcnts
Model 2: Model for received social activities
I nfluence of received social actions on success in HealthSeeker
( = .175 p < .001;
(Note: = .078 for step 1 (p < .001), = .045 (.123) (p < .001) ; p <
.001)
Both models have a strong and significant effect on the
depended variable and show that these values seem to be
highly connected. As noticed before the active social
behaviour has again a much stronger effect on the
motivation to succeed.
This effect is also visible in Model 3 when we analyze all
social actions (active and passive) together in one model.
Numbcr0ComplctcJHissions
= 1.29u +1.191 Numbcr o Commcnts + u.464
Numbcr o ScnJ KuJos + .186
numbcr o scnJ Cbollcngcs + 1.7u9
Numbcr o wollpost
Model 3: Combined model with relevant social activities.
I nfluence of social activities (send & received) on success in
HealthSeeker
(r = .299, p < .001;
(Note: = .299 for step 1 (p < .001), = .0 (.299) (p < .001) ; p <
.001)
Model 3 included all variables stepwise backwards and
kept these variables with a significant effect on the success
metric: completed missions. The variables with the highest
importance to predict the success or motivation in the
games according to the developed model are: Comments,
Sent Kudos, Sent Challenges and Wall Posts.
The combined Model 3 shows the important recorded
interactions with a great influence (r = .299; p < .001) on
the motivation to succeed in the game. Using this
information we created another filter to identify the user
groups from our subsample which can be described as
social active users: A user that uses the social features
comments, kudos, challenges and wall-posts.
With this definition for a social active user based on Model
3 we are able to look at the average behaviour of a user
based on his social activity status during his live as a
HealthSeeker player. We identified three user profiles with
this definition:
Highly Social Active Player: uses all of the identified social
variables (N = 16)

N = 803
Compl.
Missions
Compl.
Rate (in
%)
Days
Spent in
Game
Sent I nvit.
Has
Friends
p = .001;
r = .108
p = .000;
r = .161
p = .000;
r = .149
p = .000;
r = .165
Number of
Friends
p = .000;
r = .157
p = .000;
r = .181
p = .002;
r = .195
p = .000;
r = .133
Number of
Wall Posts
p = .000;
r = .325
p = .000;
r = .169
p > .005 p = .000;
r = .156
Send
Challenges
p = .000;
r = .374
p = .000;
r = .155
p > .005 p = .001;
r = .113
Send
Kudos
p = .000;
r = .381
p = .000;
r = .183
p = .002;
r = .110
p = .000;
r = .200
Received
Kudos
p = .000;
r = .287
p = .000;
r = .159
p = .000;
r = .015
p = .000;
r = .098
Receive
Challenges
p = .000;
r = .007
p = .000;
r = .035
p > .005 p = .000;
r = .002
Table 1: Correlation Table for Social Metrics (N =cases; r
=effect strength)
DESVIG 2013
42
Non-Social Player: has never used either of these functions (N
= 457)
Partially-Social Active Player: uses some functions but not all
of them (N = 330)
Table 2 shows some of the average values for the recorded
success metrics separated into the three user types. While
highly social active users complete more than five times as
many missions as none-social active users, they also spend
more than twice as many days in the game.
Completed
missions
Number of
actions in the
game
Days spend
in game
Highly social
active user
(N =16)
M = 19.94;
SD = 4.073
M = 870.31;
SD = 506.403
M = 125.56;
SD = 46.719
None-social
active user
(N =457)
M = 1.39;
D = .209
M = 6.46;
SD = .686
M = 71.17;
D = 6.262
Partial social
active user
(N =330)
M = 3.31;
SD = .367
M = 38.85;
SD = 5.023
M = 47.92;
SD = 5.092
Table 2: Display the Means of success metrics spitted into
three discovered groups based on the Model 3
Also users who use some of the discovered social functions
show more motivation to do healthy missions and action in
the game. Even if this is only an overview about the
average recorded behaviour for the overall live time of a
HealthSeeker user this shows the difference in behaviour
and motivation between these three groups.

DISCUSSION
Within this paper we defined metrics which described the
recorded actions and information of the game
HealthSeeker. We categorized these metrics as success,
social and player metrics and analyzed them for
relationships to describe the behaviour of a user.
Based on these metrics we could show that an important
factor is to have friends in the game. Having social
connections increased the motivation to be successful and
user tends to start and complete more healthy activities
over a longer time period. That friends are important for a
social game is clearly an important yet maybe obvious
result for a social health game. But it seems that just the
fact of heaving friends in the same game seems to be
important and can increase the outcome in the game.
Concepts like awareness or sociability can play an
important role but couldnt be included in this study. Future
studies could try to look deeper into this question if the fact
of just the knowledge of people playing the same game can
have positive effects on users behaviour.
Instead of just looking at a users social connections we
analyzed a users social interaction intensity and the effect
on the success in the game. The effect of a social active
support group within the game increases a users
motivation to complete healthy missions. Especially
sharing tasks in form of challenges or talking about success
and problems seem important for users. So, we could show
that the number of challenges explains about 14% of users
accomplished missions in the game. Challenges are
HealthSeekers way of creating cooperative or competitive
quests, which can be solved together with family, friends or
other users. While the effect of challenges is already quite
high the effect of direct moral support seems to have an
even higher impact on users behaviour in the social health
game. Therefore HealthSeeker offers the Kudos-system
with which users can send supporting messages for specific
goals or actions to other users. Based on our analysis the
Kudos-system can explain more than 28% of the variance
of successful missions. These two features (challenge and
Kudos-system) show that the moral support and
collaborative features are important for a social network.
Another important function of the social game seems to be
the possibility to discuss topics within the community.
To describe the relationship for the different success
influencing values we used regression models. We could
develop three models which show the influence of
receiving- and sending- social actions and a combined
model. While the models specifically about sending or
receiving actions show perfectly how social behaviour
influences the motivation to succeed in the game the
combined model can explain 30% of the completed mission
by itself. The three models underline the previous findings
that the Challenge, Kudos and Communication system
seem to have a big influence on the users behaviour and
build a good support mechanism. Though it seems that
especially social active users, who use the social functions
actively by sending challenges or other activities to other
user, are more engaged in the game than those who do not.
By developing definitions based on important metrics we
defined a social active user and showed that those users
spend a lot more time and effort playing the game and
accomplishing healthy missions. Based on our regression
analysis we could show that the most important metrics to
describe a users social behaviour are comments, wall-
posts, kudos and challenges.
While the user base we got from the company was only a
subsample and the users within the subsample only spend
62 days in the game in average, a social active user keep
playing the game for several years. A user who does not
use social functions (N = 330) stops playing after the first
day. Even if we compare missions as the success metric
with the social activeness of a user we could show those
users who partly use the social functions accomplish three
times and those fully social active users even almost 20
times more missions.
With these results we show important functions to describe
a users success or motivation to succeed in the game.
These metrics can be used to increase the success of a
social health game by encouraging users to use social
functions. Many of those functions seem to be common
DESVIG 2013
43
social network site functionality like posting messages on a
user wall or commenting actions in the game. However,
social interactions between users in particular (challenges,
kudos) seem to have great influence on user behaviour.
These findings are similar to the system of interventions in
traditional offline support groups. Also here direct moral
support from a group of people (family, friends or relatives)
seems to be important. In future studies, the differences of
these influences on demographic values could bring results
that are more interesting for this topic. The following are
our takeaways from this study:
To analyze social games metrics can be categorized into
the following groups: Social Metrics, Success Metrics,
Viral Metrics and Other
The number of friends is an important factor to motivate
users to be active and successful in the game
Using social functions increases motivation to succeed
The most important social active functions are: wall-
posts, comments, sending kudos and sending challenges.
REFERENCES
1. A. Holzinger, S. Dorner, M. Fdinger, A. Valdez, and
M. Ziefle, Chances of increasing youth health
awareness through mobile wellness applications, Proc.
of USAB10, pp. 71-81, ACM, 2010.
2. E. Koskinen and J. Salminen, A customizable mobile
tool for supporting health behaviour interventions,
Proc. of EMBS 2007, pp. 5907-5910. IEEE, 2007.
3. Y. Yaojin, The design and implementation of a web
mobile-based behaviour change application system,
Proc. of ITAB 2008, pp. 491494. IEEE, 2008.
4. C. M. Renders, G. D. Valk, S. J. Griffin, E. H. Wagner,
J. ThM, and W. J. J. Assendelft, "Interventions to
improve the management of diabetes in primary care,
outpatient, and community settings," Diabetes Care,
vol. 24, no. 10, pp. 1821-1833, Oct. 2001.
5. D. Walters, A. Sarela, A. Fairfull, K. Neighbour, C.
Cowen, B. Stephens, T. Sellwood, B. Sellwood, M.
Steer, M. Aust, R. Francis, C. K. Lee, S. Hoffman, G.
Brealey, and M. Karunanithi, "A mobile phone-based
care model for outpatient cardiac rehabilitation: the care
assessment platform (CAP)," BMC Cardiovascular
Disorders, vol. 10, no. 1, pp. 5+, Jan. 2010.
6. E. Mattila, I. Korhonen, J. H. Salminen, A. Ahtinen, E.
Koskinen, A. Sarela, J. Parkka, and R. Lappalainen,
"Empowering citizens for well-being and chronic
disease management with wellness diary," IEEE
Transactions on Information Technology in
Biomedicine, vol. 14, no. 2, pp. 456-463, Mar. 2010.
7. H. A. van Dam, F. G. van der Horst, L. Knoops, R. M.
Ryckman, H. F. J. M. Crebolder, and B. H. W. van den
Borne, "Social support in diabetes: a systematic review
of controlled intervention studies," Patient Education
and Counseling, vol. 59, no. 1, pp. 1-12, Oct. 2005.
8. Gee, James Paul. What video games have to teach us
about learning and literacy. Palgrave: MacMillan,
2007.
9. A. Grimes, V. Kantroo, and R. E. Grinter, "Let's play!:
mobile health games for adults," Proc. of Ubicomp '10.
New York, NY, USA: ACM, 2010, pp. 241-250.
10. N. B. Ellison, C. Steinfield, and C. Lampe, "The
benefits of facebook friends: social capital and
college students' use of online social network sites,"
Journal of Computer-Mediated Communication, vol.
12, no. 4, pp. 1143-1168, Jul. 2007.
11. J. B. Sung, "Social network games: Exploring audience
traits," Proc. of CHI 2010, Apr. 2010, pp. 3649-3654.
12. M. Klauser, A. Ktteritzsch, J. Niesenhaus, and S.
Budweg, "Analysis and classification of serious games
for elderly," Proc. of Mensch & Computer 2012: 12.
13. K. Renaud and P. Gray, "Making sense of low-level
usage data to understand user activities," Proc. of
SAICSIT '04. 2004, pp. 115-124.
14. A. Tychsen and A. Canossa, "Defining personas in
games using metrics," Proc. of Future Play08. ACM,
2008, pp. 73-80.
15. J. H. Kim, D. V. Gunn, E. Schuh, B. Phillips, R. J.
Pagulayan, and D. Wixon, "Tracking real-time user
experience (TRUE): a comprehensive instrumentation
solution for complex systems," Proc. of CHI '08. ACM,
2008, pp. 443-452.
16. Ducheneaut, N. ; Moore, R. J. ; Nickell, E. Designing
for sociability in massively multiplayer games: an
examination of the third places of SWG. Proc. of
Other Players 2004. Copenhagen; Denmark.
Copenhagen: IT University of Copenhagen; 2004.
17. A. Canossa, A. Drachen, and J. R. Srensen,
"Arrrgghh!!!: blending quantitative and qualitative
methods to detect player frustration," Proc. of FDG '11.
ACM, 2011.
18. Renault, K.; Gray, P. 2004. Making sense of low-level
usage data to understand user activities. Proc. of
SAICSIT, 115-134.
19. DeRosa, P. Tracking Player Feedback to Improve Game
Design. Gamasutra, August 7, 2007.
20. I. Guy, M. Jacovi, E. Shahar, N. Meshulam, V. Soroka,
and S. Farrell, "Harvesting with SONAR: the value of
aggregating social network information," Proc. of CHI
'08. ACM, 2008, pp. 1017-1026

DESVIG 2013
44

Towards a Tripartite Theoretical Framework for the
Sociality of Digital Games: Psycho-Social-Biological
Effie Lai-Chong Law
Department of Computer
Science
University of Leicester, UK
elaw@mcs.le.ac.uk
Mark Springett
Computing and Multimedia
Technology
Middlesex University, UK
m.springett@mdx.ac.uk
Martin Mihajlov
Faculty of Economics
Ss. Cyril and Methodius
University, Macedonia
martinmihajlov@gmail.com

ABSTRACT
There is much debate as to whether user experience (UX) is
inherently individual or socially constructed. Games
provide fertile contexts to explore and investigate the nature
of UX., Several attempts have been undertaken to
operationalize the sociality of digital games. We
recommend that a tripartite theoretical framework from the
psychological, biological and social perspective is needed
to substantiate the operationlization. Several notions are
deemed particularly relevant, including affective practice,
neo-J amesian theory, emotional contagion, activity theory,
social identity, and discursive theory. It is considered a
grand challenge for us to integrate them to develop a
coherent and useful framework for understanding UX in
general and sociality of digital games in particular.
Author Keywords
Sociality; Digital games; Activity theory; Body; Contagion;
Affect; Emotion; Affective practice; Discourse
ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g., HCI):
Miscellaneous.
General Terms
Theory; Design;
INTRODUCTION
The soar of interest in affect and emotion in the field of
Human-Computer Interaction (HCI) has been instigated by
the shift of emphasis from usability to user experience (UX)
at the turn of millennium. As experience goals such as fun,
challenge, surprise and tension are hailed as key, if not the,
main motivation for playing computer games, analysing the
process of gaming and feelings arising is deemed highly
relevant for understanding UX. There exists a broad range
of computer games and different game classifications
(Elverdam & Aarseth 2008). Therefore, selecting a specific
game genre to study is already not a simple task. One of the
classification criteria is modality: single-player versus
multiplayer. This seemingly straightforward classifier is
complicated by the observation that rules of games are
often adapted as well as created ad hoc by gamers,
changing from an individual to a social mode, or vice versa,
even within one game session (e.g. Consalvo 2009).
Another classification criterion is arguably the goal of
digital games. With the advent of serious games (Ritterfeld
2009), game with a purpose (GWAP)
1
and digital
educational games (DEG) (Prensky 2001), increasing
research efforts have been invested in exploiting such
games for learning and training. Games have strong
potential in sustaining gamers motivation to engage in
gainful tasks.. However, designing and evaluating games
that can simultaneously serve the pragmatic goal of
learning/training and hedonic goal of playing (Hassenzahl
2011) is challenging. For this particular game genre (i.e.
games more than pure entertainment), we argue that
sociality or sociability (which we regard as synonymous)
could play a particular important role. The notion of
sociality is defined as the tendency of groups and persons
to develop social links and live in communities
2
Inter-
subjectivity resulting from social interactions can enhance
sense-making (Fuchs & De J aegher 2009), rendering the
content of the game more comprehensible or persuasive
than otherwise.
One of the main aims of this workshop and its forerunner -
the Special Interest Group (SIG) in CHI11 (Christou et al.
2011) is to operationalize the notion of sociality in digital
games. We interpret it as the extent to which a digital game
supports social interactions among or between gamers in
online as well as physical settings (co-located gamers).
Furthermore, we emphasize that it is essential to ground the
operationalization in a robust conceptual framework and
that it is promising to understand the sociality of games
from a tripartite perspective: psychological, social, and
biological. Specifically, we explore several related
concepts: Practice, Activity, Body, Contagion, and Identity.
The concept Activity is anchored in the first authors
earlier work on applying Activity Theory to analyse the
interactions of pairs of children playing a digital
educational game (DEG) called 80Days (Law & Sun, 2012)
The concepts Practice, Body, and Contagion are
inspired by Wetherells (2012) recent monograph where she
presents intriguing arguments about affect and emotion
from the psycho-biological and psycho-social perspective.

1
http://www.gwap.com
2
http://dictionary.reference.com/browse/sociality?s=t/
[Collins English Dictionary]
DESVIG 2013
45

We first explore each of the above mentioned concepts.
Next we present a case study on the evaluation of the DEG
80Days to see how the tripartite framework can shed light
onto understanding sociality of digital games.
EMOTION, AFFECT AND FEELING
As mentioned earlier, the research interest in understanding
the nature of emotion has been fuelled by the recent
movement on UX. Based on the results of a survey with the
HCI community (Law et al., 2009), UX is commonly
understood as highly subjective, dynamic and contextual.
Furthermore, the most controversial statement (out of a
total of 23) concerning the nature of UX (with the highest
standard deviation in the extent of agreement on a 5-point
Likert scale) was Only an individual person can have an
experience. An experience is something personal,
something within a person. (statement #19). In the
qualitative comments on this particular statement, some
respondents remark that experience is very much bound to
peoples social groups and that a community may share the
experience. Another stance is that while social others in a
group can shape how one experiences with an artifact, only
an individual can have feelings and experience. This
resonates with the high extent of agreement with statement
#5 UX occurs in. and is dependent on the context in which
the artifact is experienced. Nevertheless, while it seems
indisputable that subjective experiences are not static
mental states and that they are shaped by contextual factors,
whether they are inherently individual or socially
constructed remains controversial (Roto et al 2011). The
debates can be attributed to divergent interpretations what
emotion is and how it can be operationalized (e.g., Boehner
et al 2007).
Echoing Wetherells (2012) view, Burkitt (2002) furnishes
an elegant account of emotion and its relation with affect
and feeling. Built upon the work of earlier philosophers
such as Spinoza, Bateson and Vygotsky, Burkitt argues for
relationality of emotion, implying that emotion is not an
object within a person but is a relation arising in response to
a situation, directly or indirectly involving social others.
Emotion, in this view, is socially constructed. In corollary,
affect is not exclusively confined to an animate or
inanimate object. Burkhitt also argues that emotions rest on
bodily responses that somehow turn into feelings. He
describes feelings as practical consciousness, a kind of
action-oriented intuition. Furthermore, he remarks that
feeling is not expressed in discourse as much as completed
in discourse (Wetherell 2012). It may imply that culturally
accepted terms and narratives enable feelings to be
communicated. This social constructionist perspective of
emotion is gaining acceptance in the field of UX and can
contribute to understanding the sociality of games.
AFFECTIVE PRACTICE
An intriguing notion related to the social constructionist
approach to emotion is affective practice, as proposed by
Wetherell. Specifically, affective practice focuses on the
emotional as it appears in social life and tries to follow
what participants do. (p.4). It is continually evolving in
multiple and divergent directions. Wetherell specifies a key
advantage of practice thinking as a way of conceptualizing
social action as constantly in motion while yet recognizing
too that the past, and what has been done before, constrains
the present and the future (p.23). This draws in the socio-
historical base of emotion. Practice can be an activity for
participants, an established reference point, an
improvisation, and also a form of training with discipline
and control. While practice tends to result in some form of
habit, it helps people identify patterns not only in mundane
and recurring events but also in uncommon and one-off
activities.
In affective practice of playing digital games socially, affect
of gamers can arise from scratch (e.g. discovering a new
feature) and from a process of repeatedly mimicking their
own as well as other gamers past practice (e.g. shouting
when being defeated by an opponent).
Furthermore, according to Wetherell, in affective practice,
our body parts such as facial muscles, neurons, heart and
sweat glands get patterned together with feelings and
thoughts, which are shaped by our interpretative repertoires
of social relations and personal histories. Patterns
constituted by different combinations of bodily,
phenomenological, social, cultural, historical, and
developmental components are resulted and interact with
each other in a complicated manner. Some affective
patterns might involve only a small number of contributing
patterns, when loosely connected, they might dissipate
rapidly.
For instance, an online GWAP (footnote 1) called Verbosity
randomly connects two gamers, whose identities are
unknown to each other, to engage in a turn-taking word
guessing game. A gamer may get annoyed when the other
gamer is perceived as incompetent in providing good hints
in one instance, but she may get excited when the other
gamer makes a correct guess in no time. However, the
emotional ups and downs within a three-minute game
episode are inconsequential, especially when the
partnership normally is not carried over to subsequent
rounds of the game. Other affective practice might tightly
be coupled with social practices, rendering the related affect
persistent. For instance, sharing narratives during regular
meal times in a family where members affect is mutually
shaped by each others verbal and non-verbal behaviours.
In an affective pattern the intensity of different elements
can vary over time and thus their dominance in the whole.
For instance, in a competitive social digital game of car
racing in a geographically grotesque terra, the somatic
responses (e.g. clammy hands, fast heart rate) get patterned
with feelings of excitement and tension, the attractive
photorealistic landscape, knowledge of the opponents
strength, and assessment of personal competence. This
affective practice can be knotted in with the cheering of
DESVIG 2013
46

non-playing observers, rendering the affect of engagement
so strong that no other ordering forces (e.g. a scheduled
closing time of a social gathering) can disrupt it.
As shown by the above examples, affective practice can
vary largely in scale, ranging from solo subjects, dyads,
nuclear families to an entire country. Affect can be
distributed across in shared jokes, social games (co-
presence or online), mass protests, and national celebrations
of victory in a worldwide sport event. Wetherell (2012)
asserts that affective practice is observable and explicit, but
factors shaping affect in practice are implicit and largely
non-conscious. Such underlying factors include our deeply
entrenched values, habitual way of thinking and acting, and
internalized relational patterns and semiotic associations.
This is consistent with the phenomenon that sometimes we
are not aware of what we are doing as we do it, and we only
become conscious of how our bodies and minds have been
deployed after the event. Indeed, affect can be highly
automatic.
While we see the potential of affective practice in analyzing
the sociality of digital games, its underlying multi-
dimensional bases, including somatic, discursive, situated,
historical, social, psychological and cultural aspects, can be
overwhelmingly complex.
PSYCHOBIOLOGICAL VIEW: BODY & CONTAGION
The controversial J ames-Lange theory can be seen as a
precursor for the ongoing debates about the role of bodily
or somatic activities in affect and emotion, which is
assumed to be mediated by a cognitive process of
interpretation. Broadly speaking, such bodily activities
range from explicit verbal as well as non-verbal behaviours
such as dancing, crying, screaming and panting to implicit
patterns of neural network in human brains.
Recent neurological findings have further fuelled this age-
old debate on the precedence of body over affect, or vice-
versa. Damasios contemporary work (1999, 2004) is
frequently cited by social scientists who aspire to identify
interdisciplinary arguments or facts to settle the debate.
Specifically, Damasios stance on this issue is referred to as
neo-J amesian as they both insist that the body precedes the
feeling. In other words, in response to a stimulus, our body
and brain first perform certain functions and our registering
of such performance then leads to our experiencing of
specific emotions. One major difference between J ames
and Damasios stance is that the latter also takes the role of
cognitive events such as memories into consideration.
J ames illustrative example that: seeing a bear escaping
(sensorimotor reactions to an observable external stimulus)
being aware of self escaping feeling scared can be
reframed as seeing a bear triggering internalized
knowledge about the danger of a bear body response
feeling. We argue that both directions (i.e. affect leads to
body response or body responses results in affect) are
plausible and even necessary for survival.
Furthermore, Damasio describes a phenomenon called as if
body loop where no actual bodily or physical change takes
place in response to a cognitive event, which nevertheless
triggers some particular affective experience in a person.
This observation resonates with an increasingly referenced
notion dubbed mirror neurons (Rizzolatti & Sinigaglia
2008) and the associated concept emotional contagion
(Hatfield et al 2009; Wild, Erb & Bartels 2001).
Accordingly, when we observe someone whom we care
feels the pain, say, caused by a cut in her finger or feels
frustrated, say, by repeated losses in a video game, our
brains fire neurons in the same pattern as the brains of those
actually in pain or frustration. This intriguing observation
suggests the neurobiological bases of human inter-
subjectivity and cognitive empathy (Preston et al 2007).
Furthermore, Damasio and other researchers like Tomkins
Ekman and Panksepp differentiate primary emotions (or
emotional primes) from secondary (or social) emotions.
The former (i.e. anger, sadness, fear, disgust, surprise and
happiness; Ekman 1992) are considered as innate,
universal, pre-organized and hard-wired in our brains in the
form of default patterns of neural firing, whereas the latter
such as shyness and contempt are seen as variations or
derivatives of the former. However, there have always been
arguments against the representativeness of the six basic or
primary emotions (Scherer 1994; further arguments see
Wetherell 2012). For instance, in lieu of basic emotions
theory, Scherer (2009) puts forward the component process
model. Accordingly, experiencing results from appraising
and integrating information from multiple components
(e.g., cognitive schemata, neural events, bodily movements)
at multiple levels. This controversy over the basic emotions
does not defy the unanimous view on the embodiment of
human affect.
In summary, from the psycho-biological perspective,
subjective-experiential feelings (e.g. tension and challenge
during gameplay) are accompanied not only by physical
changes in our body/brain and but also by other cognitive
(e.g., adapting gaming strategies) and motivational-
behavioural (i.e. sustaining gameplay; taking over input
control) changes.
PSYCHOSOCIAL VIEW: ACTIVITY & IDENTITY
Two theoretical constructs are regarded as highly pertinent
to the understanding the sociality of digital games: activity
and identity.
Acti vity in Activity Theory
Rooted in the work of Vygotsky and his student Leontiev,
Activity Theory (AT) is a framework aiming to understand
individual human beings (i.e. subjects) and social systems
they constitute through an analysis of the origin, structure
and processes of their activities (Kaptelinin & Nardi, 2006).
Activity is defined as a unit of subject-object interaction
with the object having the status of a motive that meets a
certain need of the subject. Engestrm (1987) expanded AT
to address systems of activity at the collective level. He
DESVIG 2013
47

defined the activity system model with seven components:
subject, object, tools, division of labour, community, rules,
and outcome (Figure 1):
Subject: an individual engaging in an activity;
Object: an objective or motive held by the subject for
performing an activity;
Tool: material or mental agency mediating the subject-object
interaction;
Rule: regulations of actions and interactions in an activity;
Community: one or more people sharing the objective with the
subject;
Division of labour: how tasks, power and status are divided
among cooperating members of the community;
Outcome: an idea, a situational status, or a specific
positive/negative emotional response.

Figure 1. Engestrms (1987) activity system model
AT accounts for both moment-to-moment interactions that
constitute gameplay (including the gamers goals and
intentions) and the broader socio-cultural contexts that
situate the gaming activity. For Activity Theorists, it is not
the presence of these components in isolation that makes
meaningful analysis, but rather the interactions among
them. Engestrm (1993) refers to such relations as primary
and secondary contradictions. Primary contradictions are
those that occur within a component of a system (e.g.
tools), while secondary contradictions are those that occur
between components of a system (e.g. subjects and tools).
In a situation where a game is used in formal learning
environments as a DEG, the tension between winning the
game and learning the subject matter is the object of an
activity system, depending on whether the student or the
teacher is the subject of the activity system. Activity
Theorists argue that characterizing the tension of an activity
system can help participants understand and react to
changes in the system.
AT has been used to study the activity of an organization
that involves computer use (Kaptelinin, 1996). From an AT
perspective, the computer is the tool mediating the
interaction of humans with their environment (Bannon &
Kaptelinin, 2000). Barr et al. (2006) recommended the
applicability of AT in analysing conduct in video games.
Roussou (2004) and her colleagues (2008) applied AT to
study the interaction in learning through virtual reality and
concluded that AT is a useful framework to understand the
relationship between the user and virtual environment.
However, their analyses did not explicitly address user
experience (UX); they focused primarily on the behavioural
aspect of gameplay.
According to Rogers (2008), there are 57 varieties of AT,
and we are not yet to add another one. Instead, we apply
Engestrm (1987) extended framework of AT (Figure 1),
which is normally used for analysing activities in a system
of a scale larger than a collaborative gameplay setting to be
described subsequently. Nevertheless, we are aware that
there are some limitations of AT. As an open framework,
AT enables customisation, which, on one hand, may ease its
application to a problem domain, but, on the other hand,
may lead to a variant that deviates much from the original
AT, causing incoherence and confusion. Furthermore, with
its root in cognitive psychology, AT does not explicitly
address affective constructs. However, we see it as an
opportunity rather than a hindrance, and therefore extend it
to cover the experiential aspect of human-technology
interaction.
Identity
Games and gamification transposes well on to a number of
more instrumental, goal-directed areas of life. They have
motivational, cognitive and social advantages in education
and also in fitness and health applications, where multiple
participants compete or co-operate with strategies and skills
to complete game quests or simply share gameplay
experience. The use of gaming to aid learning is of
particular interest in our work. Key features such as rules
and conventions, reward structures, feedback and hints have
significant utility for learning. In addition to the intrinsic
engagability of games, a sense of sociability is an important
motivator to the gamer, who is simultaneously a learner.
Gaming and a sense of self seem to have a very strong link.
Sociologists have often pointed to ways in which
gamification is spontaneously used to enhance the
tolerability of working tasks on factory floors (Nelson
2012). Equally, psychological studies of gamers suggest
that the individuals identity seem to be strongly mediated
through games (Coulson et al 2012). In turn gaming seems
to be linked to a sense of social identity.
It can be argued that the sociability factor in gaming is a
phenomenon that extends beyond the actual presence of a
fellow gamer, either as an opponent or a collaborator. A
sense of the presence of others, or of community seems to
be equally as strong. The nature of the self-conscious
individual in relation to gaming or a game-mediated
community seems to be less dependent on the actual
presence of others than the perceived presence. This
phenomenon seems to reflect the concepts of socio and ideo
pleasure in Jordans taxonomy of pleasure (1998). A game
may engender a sense of social belonging and identity
conveyed through visual design and the game concept.
A further theoretical construct relevant is that of
persuasion, often a key tent of educational software. The
Tool
Subject Object
Rules
Community
Division of
Labour
OUTCOME
DESVIG 2013
48

use of gaming is a way of producing approach where a
tendency for withdrawal was prevailing. Considering the
Theory of Reasoned Actions (Azjen & Fishbein 1980) we
can consider the way in which the perceived social or
ideological dimension can be utilised to produce positive
actions. The use of the social dimension weakens resistance
to educational activities that learners may be reluctant to
engage in. This could be by reversing a social stigma (for
example in school education where learning may have a
stigma attached), or in learning of digital technologies for
older users (where there is a sense of disenfranchisement, a
sense that it is not their technology). The overlay of a
sense of social positivity through gaming produces
behaviour change.
Hence, sociability in a conveyed but not necessarily literal
sense may be central to the effectiveness of a digital
educational game (DEG). Access to education is often as
much an issue in overcoming reluctance as in cognitive
progress of learning, and the social aspect of games is a
major element of this.
Subsequently we briefly describe a case study to illustrate
the possible applications of the theoretical constructs
discussed above.
CASE STUDY: TOUCHTABLE GAMES & OLDER USERS
In this section we briefly describe a case study to illustrate
the application of Activity Theory (AT) in the context of a
recent empirical study using games to familiarise older
citizens with digital technology. In describing the games,
we map the related observations to the seven basic
components of the activity system model (cf. Figure 1).
The example we use is of participants between 58 and 79
years old playing simple touch-table games. In this example
the game-playing was the players first exposure to digital
technology. The aim was to study the relationship between
instrumental and affective aspects of learning. Instrumental
and affective factors are intertwined for those represented
by the participant group. Many older people express the
belief that they relate to digital technology as something
that is made by and for a culture and population that they
exist outside. As a result there is a reluctance to attempt
learning. The study used a touch-table to overcome
reluctance, using the game and simple touch-table
interaction to encourage participation.
The games selected for the study included familiar objects
such as world maps and playing cards (Figure 2). There
were also short games involving math puzzles, geometry
and picture sorting. The tasks were selected for coverage of
the basic manipulation set and the varied nature of object
and action metaphors.
The Subject, regardless of the collaborative nature of the
game-playing task, is the single individual player.
Utterances and behaviours from the other player are an
input to the actors cognitive system, and seem therefore to
fit more comfortably as a Tool.

Figure 2. Two older participants were playing touch-table
games on world maps, using both verbal and non-verbal
cues to communicate and collaborate.
The study showed verbal utterances ranging from explicit
suggestions, assertions of a hypothesis, confirmation,
building, and information relating to a collaborators
movement. Non-verbal contributions included pointing to
an object, making dummy gestures to demonstrate a
technique or a perception of a correct action. The usual
pattern emerging was that players were usually willing to
share roles, taking turns to be the active player. This would
often be a rapid switch of roles, in which one player would
demonstrate an alternative theory in reaction to the previous
action. Occasional exceptions included one session in
which a participant made no attempt to perform any
manipulations, and others where one partner emerged as a
lead player, with the other adding selections and making
occasional physical demonstrations of a hypothesis or
action suggestion. The generation of hypothesis was often
in reaction to a wrong or unsuccessful action for the active
player. It was frequently linked to recognition of some
aspect of a system response that that was probably not
intended by the active player, such as the slight changing of
an objects position as a result of the wrong action being
attempted. The most frequent example of information was
where the active player was observed performing an
involuntary input. For example one subject made accidental
inputs with his elbow when leaning across the table and was
unable to explain the system response, until the co-player
noticed and pointed out the cause of the system behaviour.
The Rule component had the function of yoking the
principles of touch-table interaction to the rules of the
particular games. As the game was intended to familiarise
players with baseline knowledge and skills for digital
interaction, the intention was that violations of rules
(unsuccessful actions) should facilitate learning. This
reflects theories of display-based learning and learning by
generalisation of examples.
DESVIG 2013
49

For instance, the card-playing game violated some typical
principles of manual card-playing. Players tried
unsuccessfully to turn cards over if they were facing
downwards by a twisting motion (Figure 3). The lack of
success forced a number of switches of position as players
took it in turns to generate hypotheses, critique the each
others verbal and physical turns. In all cases this led
eventually to the successful action being performed. This
appears to suggest that positive valence and motivation to
pursue the game task are not solely facilitated by successful
action, but by a sense that positive outcomes remain
possible, and that players are capable of achieving them.

Figure 3. A dyad of older participants were playing cards
on the touch-table
The role of collaboration in this is partly that it widens the
space of exploration, and makes more resources available to
participants. In this example the mutual willingness to
explore, generating testing and critiquing each others
hypotheses and exploratory actions had observable benefits
in learning. It is less demonstrable but nonetheless likely
that positive experience and motivation was enhanced
through the collaboration. Where frustration at a players
actions was observed it usually led to a critique, the
generation of an alternative hypothesis of a spontaneous
exploratory action.
In the following we summarise the mapping from the
observations pertaining to the players and touch-table
games to the basic components of the activity system model
derived from AT:
Subject:
o Player
Object:
o Solving a game-task;
o Familiarisation with the technology;
o Confidence building;
Tool:
o Game metaphor;
o Affordances of the touch-table interface;
o Knowledge-base in long-term memory;
o Physical interaction skills;
o Utterances and gestures of co-player;
Rule:
o Actual principles of the game;
o Behaviour of objects (in contrast to expectations
generated from metaphor interpretation and
experience of non-digital manipulation)
Community:
o The other participant in mutual problem-solving and
mental model building;
Division of labour
o The active player in any instant;
o The initiator of an action (through action of a
verbal/gestural directive)
Outcome:
o Mental model development;
o Skill acquisition;
o Adding to stack of known principles of operation;
o Stronger expectations of success;
o Confidence and motivation (willingness to explore;
o Emotional reactions (positive and negative);
In applying AT to understand the mediated interactions
between the dyadic players, we could only rely on explicit
non-verbal and verbal behaviours. The underlying
physiological changes, down to the neural level, were
impossible for us to capture, given the constraints of the
empirical study setup. Furthermore, if some highly sensitive
and sophisticated devices were available to measure time
gaps between the expression of feelings and the exhibition
of emotional behaviour (e.g. changing the sitting posture
because of excitement; taking input control from ones
partner because of frustration), we might be able to identify
the sequential relationship between body and feeling (cf.
Damasiao-neo-J amesians conjecture). Nonetheless, we
assume the schematic representation depicted in Figure 1
could be used to illustrate affective practice where
interactions between the components occur at both the
behavioural as well as emotional level. This could be
analogous to Scherers (2001) multi-level sequential check
model, but less complicated.
CONCLUDING REMARK
Apparently the theoretical constructs we have addressed in
foregoing text is not exhaustive. One obvious omission is
discourse, which plays an indispensable role in negotiating
affect and enabling sense-making. As emphatically pointed
out by Wetherell (2012) human affect is inextricably
linked with meaning-making and with the semiotic and the
discursive. Nonetheless, discursive theories are covered by
volumes of publication, and their application to affective
practice and experiences entail a separate article. The
inclusion of the discursive perspective into the tripartite
framework is one of the challenges that we aim to tackle in
our future work. In addition, the loose connections among
the theoretical constructs, namely, psychological, biological
and social, need to be further concretized in order to
develop a coherent and workable framework that enables us
to deepen the understanding of UX in general and the
DESVIG 2013
50

sociality of digital games in particular. We anticipate that
the challenge will be tough but rewarding.
REFERENCES
1. Ajzen, I. & Fishbein, M. (1980). Understanding
attitudes and predicting social behavior. Prentice-Hall.
2. Bannon, L., & Kaptelinin, V. (2000). From human-
computer interaction to computer-mediated activity. In
C. Stephanidis (Ed.), User interfaces for all: Concepts,
methods, and tools (pp. 183-202). Erlbaum.
3. Barr, P., Noble, J ., & Riddle, R. (2007). Video game
values: Humancomputer interaction and games.
Interacting with Computers, 19, 180-195.
4. Boehner, K., DePaula, R., Dourish, P., & Sengers, P.
(2007). How emotion is made and measured.
International J ournal of Human-Computer Studies,
65(4), 275-291.
5. Burkitt, I. (2002). Complex emotions: Relations,
feelings and images in emotional experience. In J.
Barbalet (ed.), Emotions and sociology. Blackwell.
6. Christou,G., Zaphiris, P., Ang, C.S., Law, E.L-C.
(2011). Designing for the user experience of sociability
in massively multiplayer online games. In Proc. CHI
Extended Abstracts : 879-882
7. Consalvo, M. (2009). There is no magic circle. Games
and Culture, 4(4), 408-417
8. Coulson, M., Barnett, J., Ferguson, C. J., & Gould, R. L.
(2012). Real feelings for virtual people: Emotional
attachments and interpersonal attraction in video games.
Psychology of Popular Media Culture, 1(3), 176-184.
9. Damasio, A.R. (1999). The feeling of what happens:
Body and emotin in the making of consciousness.
Harcourt Brace.
10. Damasio, A.R. (2004). Looking for Spinoza: J oy,
sorrow and the feeling brain. Papermac.
11. Ekman, P. (1992). An argument for basic emotions.
Cognition & Emotion, 6(3-4), 169-200.
12. Elverdam, C. and Aarseth, E. (2007). Game
classification and game design: Construction through
critical analysis. Games and Culture, 2(2007):322.
13. Engestrm, Y. (1987). Learning by expanding: An
activity-theoretical approach to developmental research.
Helsinki: Orienta-Konsultit.
14. Engestrm, Y. (1993). Developmental studies of work
as a testbench of activity theory: The case of primary
care medical practice. In S. Chaiklin & J. Lave (Eds.)
Understanding practice: perspectives on activity and
context (pp. 64-103). CUP.
15. Fuchs, T., & De J aegher, H. (2009). Enactive
intersubjectivity: Participatory sense-making and mutual
incorporation. Phenomenology and the Cognitive
Sciences, 8(4), 465-486.
16. Hassenzahl, M. (2011). User experience and experience
design. Encyclopedia of Human-Computer Interaction.
17. Hatfield, E., Rapson, R.L., & Le, Y.L. (2009).
Emotional contagion and empathy. In J . Decety, J .
(Ed.), The social neuroscience of empathy (19-30). MIT
Press.
18. J ordan, P. W. (1998). Human factors for pleasure in
product use. Applied ergonomics, 29(1), 25-33.
19. Kaptelinin, V. (1996). Computer-mediated activity:
Functional organs in social and developmental contexts.
In B. Nardi (Ed.), Context and consciousness: Activity
theory and human-computer interaction (45-68). MIT
20. Kaptelinin, V., & Nardi, B.A. (2006). Acting with
technology: Activity theory and interaction design. MIT
press.
21. Nelson, M. (2012). Soviet and American Precursors to
the Gamification of Work. In Proc. of the 16th
International Academic MindTrek Conference.
22. Prensky, M. (2005). Computer games and learning:
Digital game-based learning. Handbook of computer
game studies, 97-122.
23. Preston, S.D., Bechara, A., Damasio, H., Grabowski,
T.J ., Stansfield, R.B., Mehta, S, Damasio, A.R.(2007).
The neural substrates of cognitive empathy. Social
Neuroscience, 2(3-4):254-75.
24. Ritterfeld, U. (2009). Serious games: Mechanisms and
effects. Routledge
25. Rizzolatti, G, Sinigaglia, C. (2008). Mirrors in the
Brain. How We Share our Actions and Emotions.
Oxford University Press.
26. Rogers, Y. (2008). 57 varieties of Activity Theory.
Interacting with Computers, 20, 247-250.
27. Roussou, M., Oliver, M., and Slater, M.(2008).
Exploring activity theory as a tool for evaluating
interactivity and learning in virtual environments for
children. Cognition, Technology and Work, 10,141-153
28. Roussou, M. (2004). Examining Young Learners'
Activity Within Interactive Virtual Environments. In
Proc. Interaction Design & Children (pp. 167-168)
29. Roto, V., Law, E. L-C., Vermeeren, A.P.O.S. &
Hoonhout, J . (2011). UX White Paper. Available online:
http://www.allaboutux.org/
30. Scherer, K. (1994). Towards a concept of modal
emotions. In P. Ekman and R.J . Davidson (Eds.), The
nature of emotion: fundamental quesitons. OUP.
31. Scherer, K. R. (2001). Appraisal considered as a process
of multilevel sequential checking. In K.R. Scherer,
A.Shorr, & T. J ohnstone (Eds.), Appraisal processes in
emotion: Theory, methods, research (pp. 92, 120).
32. Scherer, K. (2009). The dynamce architecture of
emotion: Evidence for the component process model.
Cognition and Emotion, 23(7), 1307-1351.
33. Wetherell, M. (2012). Affect and emotion: A new social
science understanding. Sage.
34. Wild, B., Erb, M., & Bartels, M. (2001). Are emotions
contagious? Evoked emotions while viewing
emotionally expressive faces: quality, quantity, time
course and gender differences. Psychiatric Research,
102, 109-124.

DESVIG 2013
51
The Relationship Between Game Rules And Team
Cohesion: An Empirical Study
Eleanor Martin
University of Sussex
Falmer, Brighton, United
Kingdom
e.martin@sussex.ac.uk
J udith Good
University of Sussex
Falmer, Brighton, United
Kingdom
j.good@sussex.ac.uk

ABSTRACT
Multiplayer games (both face-to-face and online) frequently
feature teams. This study investigates whether it is possible
to use the rules of a game to alter the team cohesion. Game
rules from two face-to-face games were analyzed using
Social Identity Theory to predict which will lead to more
cohesive teams. Significant differences in team cohesion
ratings given by the players after the games suggest that the
game rules may indeed affect the team cohesion as
predicted.
Author Keywords
Game design; Teamwork; Social interaction
ACM Classification Keywords
H.5.3 Group and Organization Interfaces: Theory and
models
INTRODUCTION
Since 1999, there has been an explosive growth in online
multiplayer gaming [8]. However, there has been little
consideration of the ways in which one might design so as
to positively promote teamwork and cooperation. Instead,
it would appear that teamwork and cooperation are
enforced by making the games either too difficult or
physically impossible to complete alone [7]. This does not
necessarily lead to strong, cohesive and long lasting teams;
in World of Warcraft, the longevity of many guilds has
been shown to be less than a month [3]. Rather than
requiring teamwork, the research discussed in this paper
considers whether it is possible to promote teamwork and
cooperation through the rules of a game and, as such, to
increase the cohesion of the groups or teams that form
within the game. If this were the case, game designers could
have much more control over the strength of the teams that
form in their games.
An online, multiplayer game called African Farmer is
being designed as a learning tool for students of
international development and future policy makers. The
game has been designed so as to give players first-hand
experience of the complex decision making processes
involved in small scale farming in inhospitable
environments. The game also aims to encourage debate on
the effects of large-scale policy decisions on individuals,
their families, and their general well being. This game will
be run by a game manager and played by 15-35 players.
The players will form small teams of up to 3, with each
team responsible for a farming household consisting of
themselves and a growing number of non-playing
characters (NPCs). The NPCs range from babies to adults
and, if insufficiently cared for, they may get sick or even
die. Over a series of annual cycles the players make
decisions about what to plant, grow, buy and sell to sustain
their households through a variety of hazards, such as
drought or pest attack.
Although the farming element is important, player
interactions are also a key learning opportunity [10]. Both
inter- and intra-household interactions can have a large
impact on the strategies chosen and therefore the success
(or otherwise) of the players.
African Farmer is loosely based on two pre-existing face-
to-face games:
1. The Green Revolution Game (GRG) [1] provides a
simple yet sophisticated model of rice farming in
Bihar, India. The game allows players to choose
between planting normal or high yield rice to sustain
their household members.
2. Africulture [2] provides a model of gender roles in an
African rural community. The players have a greater
range of crops to choose from, but the farming model is
less sophisticated.
These games have very different social models, differing in
terms of the point at which teams form, the permeability of
the team boundaries, and the types of goals that players will
have (see Table 1). Social Identity Theory (SIT) considers
the way that group membership affects personal identity,
and how different factors affect the strength of group
identification. The social psychologist Henri Tajfel initially
developed SIT in the 1970s as a result of his interest in
prejudice, discrimination and inter-group conflict [6].
Social identity is defined as the individuals awareness of



Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.
DESVIG 2013
52
belonging to particular groups, and the value of those group
memberships to that person. Each individual has many
different group memberships, e.g. female, researcher,
knitter, European, etc. At any given moment these will vary
in significance and importance, based on the situation and
the accessibility of the identity. If the person frequently
identifies as a researcher (for example) it will be easier for
them to do so in a wide range of situations, whereas quite
specific circumstances may be required to make them
identify as something else. Experimental research suggests
that only one identity is salient at any given moment, and
this allows the individual to judge how to behave, their
social standing and so on [9]. GRG rates players as a team
rather than individually as Africulture does, which will
make the team identity more salient than the players
identities as individuals. This should increase the team
cohesion in players of GRG [11].
In general it is assumed that people want to associate
themselves with positive social identities whenever
possible, so where it is feasible to choose a social group,
people will be happier to associate with successful groups
rather than unsuccessful ones. In games this would imply
that better performing teams would generate more
commitment. However, it is not always possible to change
which group one belongs to; in GRG players are not
allowed to change teams once the game has begun, unlike
Africulture. The inability to change groups reduces the
impact of a low-status group on team identification [4],
suggesting that GRG players should, on average have a
stronger group cohesion level.
The group formation stage also has an effect on the
commitment of the individual to the group. If group
members are allowed to select their own groups (versus
being externally allocated) the commitment to the group
increases. However, the work done by Ellemer et al [5]
suggests that this only has an effect when the performance
of the group is poor (e.g. people feel strongly committed to
a group if they selected the group or if the group does well).
In Africulture the team formation happens as part of the
first stage of the game, after the players have discovered the
roles (male or female) they will be playing and the assets
they have been allocated as an individual. In GRG the team
formation happens before the assets have been allocated to
the team, and with less knowledge of the consequences of
the team formation. There is an element of self-selection in
both cases, but in Africulture the team selection is better
informed which may increase the players commitment to
the team. However, as this only affects the teams that
perform badly this will be counter-balanced by the effect
(explained above) of being unable to change teams in GRG.
These factors are summarized in Table 1. Based on the
combination of the three factors above, it is predicted that
players of GRG will experience a higher level of team
cohesion than those playing Africulture.
METHOD
The research investigates the relationship between game
rules and team cohesion in both GRG and Africulture,
looking specifically at the differences in team cohesion that
might exist as a result of their differing rules. This study
used a between-subjects design. In order to measure the
team cohesion in each game an instrument was constructed
consisting of 10 items, based on a pre-existing in-group
identification scale [c.f. 4]. This included statements such
as I would like to play another game with this team. and
I had a lot in common with the other team members..
Players were asked to rate how strongly they agreed with
these statements on a 7-point Likert scale ranging from
Strongly disagree to Strongly agree.
In addition to the team cohesion measures, the
questionnaire also comprised two further questions: one
querying players preferences for working in groups versus
working individually, and a second question gauging the
number of other players who were known to the individual
before the session commenced. For the first question,
players had the option of choosing between Yes, I work
best in groups, It depends on the work and the group and
No, I much prefer working alone. The second question
used a five-point Likert scale ranging I knew everyone to
I knew no one.
The participants in both games were students of
International Development and form part of the target
audience for these games. Prior to the start of the game the
participants were given an information sheet that provided
details of the study, and explained that if players decided to
take part in the study, they would be asked to complete a
questionnaire at the end of the game. The information sheet
also made it clear that player would be able to take part in
the game without participating in the study. Both games
were played to the rules contained in the respective
managers handbook. The games were each played for 3
hours. A paper questionnaire was then distributed to the
players, who completed and returned it while the game
managers prepared for a post-game discussion that allowed
the players to reflect on the events of the game. This
discussion is a standard part of the game experience with
these games, and not specifically part of the study.
Table 1: Differences in the social model
GRG Africulture
Teams form before
resource allocation.
Teams form during first
round after resource
allocation.
Teams fixed for game
duration
Team members may
leave and join a different
team.
Players have team goals Players have individual
goals
DESVIG 2013
53
GRG was offered as an extra-curricular event as part of a
summer school. 16 players signed up (10 female, 6 male).
The players were asked to sit at any of the pre-arranged
tables before the initial family and farm sizes had been
allocated. They completed 5 annual cycles of 4 seasons,
where each annual cycle allowed for the full growth cycle
of a crop from sowing through to harvesting.
20 students from the School of International Development
at a university in the South of England played Africulture
(14 female, 6 male). This was an optional activity with no
direct benefits to their grades. The players were again asked
to sit at any of the pre-arranged places, with those playing
men kept separate from those playing women (note that
gender roles in the game bore no relation to real-life
gender). The resources were allocated at this stage,
followed by a period of negotiation where players formed
teams. They completed 3 annual cycles of 4 seasons per
cycle.
RESULTS
The 10 items of the team cohesion instrument were
combined to produce a single rating for team cohesion for
each player. This has a maximum value of 70.
Table 2 shows the median and range for each game: the
median score for GRG was 61, while for Africulture, it was
58. The spread of scores was narrower for GRG (14 points)
than for Africulture (19 points).
Game played Median Minimum Maximum
GRG 61.00 56.00 70.00
Africulture 58.00 48.00 67.00
Table 2: Team cohesion scores as a function of game played

Figure 1: Frequency distribution of the number of group
members each player felt they knew before playing.

A one-tailed Mann-Whitney test was conducted on these
data. This revealed that the rating of team cohesion
recorded by the GRG players was significantly higher than
those who played Africulture (U=99.5, p<.05, r=.31).
In response to the question on group versus individual
working preference, most (N=21) said it depended on the
work, but a number (N=15) said they always prefer working
in groups. No one said they always prefer to work alone.
Using a Kruskal-Wallis test the team cohesion rating was
not found to be significantly affected by the players
reported liking for group work (H(1)=0.084, ns).
Figure 1 displays the results of how many people each
player knew prior to the game session. A Kruskal-Wallis
test showed that the number of players known did not
significantly affect the team cohesion rating (H(4)=7.908,
ns).
DISCUSSION
The results of the study suggest that there is a relationship
between the team cohesion experienced by a team member
and the particular rules of the game played, as predicted by
SIT. Players of GRG reported a higher level of team
cohesion than players of Africulture. The team cohesion
rating was not found to correlate with the number of players
known before playing, or with the players preference for
working in groups or individually
This finding suggests that it may well be possible to use
SIT to create game rules that will affect the way that
players interact in a game situation.
Although these results are encouraging, they should
nonetheless be treated with some caution. For a start,
although the two games used in this study are similar, and
are often used in similar teaching contexts to convey the
same core content, they are not identical. It was felt that
using two existing and proven games was less problematic
than modifying an existing game as, firstly, the effort and
costs involved in producing a modified version of a
commercial game would have been overly onerous and
secondly, changing the rules of the existing game may have
cause unanticipated problems during game play. However,
a further experiment is planned with the online African
Farmer game, which will use two versions of the game that
differ only in the social model.
The different number of annual cycles completed within the
playing time hints at the differences between the two
games. Africulture contains a much more complex set of
decisions than GRG, and this resulted in each annual cycle
taking longer. For this experiment a decision was made to
keep the playing time between the two games constant,
rather than the number of cycles. This resulted in the
players having the same amount of game play, and hence
contact with their team members, but it may have been
more difficult for the players of Africulture to gauge the
relative success of their team as compared to GRG.. This in
DESVIG 2013
54
turn may have had an effect on the results, given that group
success is shown to have an impact on team cohesion [4].
Again, in the follow-up experiment with African Farmer,
we plan to keep the time taken for each annual cycle similar
between the two versions.
Some of the differences in social model are quite difficult to
bring out in the course of the game-play. For example, the
ability to change teams was not spelt out by the Africulture
game manager. It was clear from the discussion at the end
of the game that players had assumed they were able to
move teams (one player mentioned selling her wife to
another household!). GRG does not allow players to change
teams, but again this was not explicitly stated. Unlike the
Africulture players, none of the GRG players expressed any
interest in changing teams. This is not dissimilar to the way
a number of online game rules are discovered by players,
through hints gleaned during play e.g. from the aesthetics
rather than the manual [12]. It may well be that making this
difference explicit at the start of the game would actually
have increased the difference seen between the two games,
however it is not always practical to discuss all of the rules
at the start of the game.
These initial findings suggest a relationship that could be
further explored in future research by examining the
different factors individually, although some factors will be
more difficult to isolate than others when creating a
coherent and enjoyable game. From a game design
perspective, this research suggests that game designers may
be able to make conscious choices in terms of their game
rules which will have a positive impact on team cohesion
during the game.
ACKNOWLEDGMENTS
Support for the development of African Farmer comes from
the UK Department for International Development through
a grant to the Future Agricultures Consortium. Thanks also
to the organizers of the STEPS summer school and to Nitya
Rao of UEA for allowing these games to take place.
REFERENCES
1. Chapman, G. & Dowler, E. The Green Revolution
Game, Marginal Context Ltd. (1982).
2. Chapman, G., Zeeuw, H. de & J iggins, J ., Africulture,
ETC Foundation/IIED (1993).
3. Ducheneaut, N., Yee, N., Nickell, E., Moore, R.J .,
Alone together? Exploring the Social Dynamics of
Massively Multiplayer Online Games. In Proc. CHI
2006, ACM Press (2006), 407416.
4. Ellemers, N., The Influence of Socio-structural
Variables on Identity Management Strategies. European
Review of Social Psychology 4, 1 (1993), 27-57.
5. Ellemers, N., Kortekaas, P., Ouwerkerk, J . Self-
categorisation, commitment to the group and group self-
esteem as related but distinct aspects of social identity.
European J ournal of Social Psychology 29, 2-3 (1999),
371-389.
6. Hogg, M.A., Social Identity Theory. In P. J . Burke, ed.
Contemporary social psychological theories. Stanford
University Press (2006), 111-136.
7. Knizia, R. The design and testing of the board game
Lord of the Rings. In K. Salen & E. Zimmerman (Eds.),
Rules of play: Game design fundamentals, The MIT
Press (2004), 22-27.
8. MMOData Charts v3.8, http://mmodata.net (2012)
9. Oakes, P. The Salience of Social Categories. In J . C.
Turner (Ed.), Rediscovering the Social Group: A Self-
Categorization Theory, Basil Blackwell (1987), 117
141.
10. Park, W., English, B., Gray, M., Cleland, C.
Experiential learning through participation in Green
Revolution/Exaction: a role-playing simulation. Review
of Agricultural Economics, 17 (1995), 257-265.
11. Reicher, S., Spears, R., Postmes, T. A Social Identity
Model of Deindividuation Phenomena. European review
of Social Psychology 6, 1 (1995), 161-198.
12. Shirinian, A., Intuition, Expectations and Culture:
learning from psychology to build better game
interfaces.
http://www.gamasutra.com/view/feature/134989/intuitio
n_expectations_and_.php (2012)

DESVIG 2013
55

S
o
c
i
a
b
i
l
i
t
y

i
n

S
o
c
i
a
l

N
e
t
w
o
r
k

G
a
m
e
s

A
b
s
t
r
a
c
t

S
o
c
i
a
l

n
e
t
w
o
r
k

g
a
m
e
s

i
n

F
a
c
e
b
o
o
k

h
a
v
e

b
e
c
o
m
e

p
o
p
u
l
a
r

p
a
s
t
-
t
i
m
e

f
o
r

m
i
l
l
i
o
n
s

o
f

p
l
a
y
e
r
s

a
r
o
u
n
d

t
h
e

w
o
r
l
d
.

A
n

i
n
t
e
r
e
s
t
i
n
g

a
s
p
e
c
t

o
f

t
h
e
s
e

s
o
c
i
a
l

g
a
m
e
s

i
s

t
h
e
i
r

a
c
t
u
a
l

s
o
c
i
a
b
i
l
i
t
y
.

I
n

t
h
i
s

p
o
s
i
t
i
o
n

p
a
p
e
r
,

w
e

e
x
a
m
i
n
e

t
h
e

s
o
c
i
a
b
i
l
i
t
y

o
f

s
o
c
i
a
l

g
a
m
e
s

f
r
o
m

t
h
r
e
e

d
i
f
f
e
r
e
n
t

p
e
r
s
p
e
c
t
i
v
e
s
:

a
c
a
d
e
m
i
a
,

i
n
d
u
s
t
r
y

a
n
d

p
l
a
y
e
r
s
.

O
u
r

a
i
m

i
s

t
o

u
n
d
e
r
s
t
a
n
d

t
h
e

c
u
r
r
e
n
t

s
t
a
t
e

o
f

s
o
c
i
a
b
i
l
i
t
y

i
n

s
o
c
i
a
l

g
a
m
e
s

a
n
d

r
e
c
o
g
n
i
z
e

t
h
e

n
e
g
l
e
c
t
e
d

a
r
e
a
s

t
h
a
t

c
o
u
l
d

b
e

u
s
e
d

i
n

t
h
e

d
e
s
i
g
n

o
f

s
o
c
i
a
l

g
a
m
e
s
.


A
u
t
h
o
r

K
e
y
w
o
r
d
s

V
i
d
e
o

G
a
m
e
s
;

S
o
c
i
a
l

N
e
t
w
o
r
k
s
;

S
o
c
i
a
l

G
a
m
e
s
;

F
a
c
e
b
o
o
k
;

S
o
c
i
a
b
i
l
i
t
y

A
C
M

C
l
a
s
s
i
f
i
c
a
t
i
o
n

K
e
y
w
o
r
d
s

H
.
5
.
m
.

I
n
f
o
r
m
a
t
i
o
n

i
n
t
e
r
f
a
c
e
s

a
n
d

p
r
e
s
e
n
t
a
t
i
o
n

(
e
.
g
.
,

H
C
I
)
:

M
i
s
c
e
l
l
a
n
e
o
u
s
.

S
e
e
:

h
t
t
p
:
/
/
w
w
w
.
a
c
m
.
o
r
g
/
a
b
o
u
t
/
c
l
a
s
s
/
1
9
9
8
/

M
a
n
d
a
t
o
r
y

s
e
c
t
i
o
n

t
o

b
e

i
n
c
l
u
d
e
d

i
n

y
o
u
r

f
i
n
a
l

v
e
r
s
i
o
n
.

G
e
n
e
r
a
l

T
e
r
m
s

S
e
e

l
i
s
t

o
f

t
h
e

l
i
m
i
t
e
d

A
C
M

1
6

t
e
r
m
s

i
n

t
h
e

i
n
s
t
r
u
c
t
i
o
n
s

a
n
d

a
d
d
i
t
i
o
n
a
l

i
n
f
o
r
m
a
t
i
o
n
:

h
t
t
p
:
/
/
w
w
w
.
s
h
e
r
i
d
a
n
p
r
i
n
t
i
n
g
.
c
o
m
/
s
i
g
c
h
i
/
g
e
n
e
r
a
l
t
e
r
m
s
.
h
t
m


O
p
t
i
o
n
a
l

s
e
c
t
i
o
n

t
o

b
e

i
n
c
l
u
d
e
d

i
n

y
o
u
r

f
i
n
a
l

v
e
r
s
i
o
n
.

C
o
p
y
r
i
g
h
t

i
s

h
e
l
d

b
y

t
h
e

a
u
t
h
o
r
/
o
w
n
e
r
(
s
)
.

C
H
I

1
3
,

A
p
r
i
l

2
7

M
a
y

2
,

2
0
1
3
,

P
a
r
i
s
,

F
r
a
n
c
e
.

A
C
M

9
7
8
-
1
-
X
X
X
X
-
X
X
X
X
-
X
/
X
X
/
X
X
.

J
a
n
n
e

P
a
a
v
i
l
a
i
n
e
n

U
n
i
v
e
r
s
i
t
y

o
f

T
a
m
p
e
r
e

K
a
n
s
l
e
r
i
n
r
i
n
n
e

1

T
a
m
p
e
r
e
,

F
I
-
3
3
0
1
4

F
i
n
l
a
n
d

j
a
n
n
e
.
p
a
a
v
i
l
a
i
n
e
n
@
u
t
a
.
f
i

H
a
n
n
u

K
o
r
h
o
n
e
n

U
n
i
v
e
r
s
i
t
y

o
f

T
a
m
p
e
r
e

K
a
n
s
l
e
r
i
n
r
i
n
n
e

1

T
a
m
p
e
r
e
,

F
I
-
3
3
0
1
4

F
i
n
l
a
n
d

h
a
n
n
u
.
j
u
h
a
n
i
.
k
o
r
h
o
n
e
n
@
u
t
a
.
f
i

K
a
t
i

A
l
h
a

U
n
i
v
e
r
s
i
t
y

o
f

T
a
m
p
e
r
e

K
a
n
s
l
e
r
i
n
r
i
n
n
e

1

T
a
m
p
e
r
e
,

F
I
-
3
3
0
1
4

F
i
n
l
a
n
d

k
a
t
i
.
a
l
h
a
@
u
t
a
.
f
i


J
a
a
k
k
o

S
t
e
n
r
o
s

U
n
i
v
e
r
s
i
t
y

o
f

T
a
m
p
e
r
e

K
a
n
s
l
e
r
i
n
r
i
n
n
e

1

T
a
m
p
e
r
e
,

F
I
-
3
3
0
1
4

F
i
n
l
a
n
d

j
a
a
k
k
o
.
s
t
e
n
r
o
s
@
u
t
a
.
f
i

F
r
a
n
s

M

y
r

U
n
i
v
e
r
s
i
t
y

o
f

T
a
m
p
e
r
e

K
a
n
s
l
e
r
i
n
r
i
n
n
e

1

T
a
m
p
e
r
e
,

F
I
-
3
3
0
1
4

F
i
n
l
a
n
d

f
r
a
n
s
.
m
a
y
r
a
@
u
t
a
.
f
i


DESVIG 2013
56
I
n
t
r
o
d
u
c
t
i
o
n

S
o
c
i
a
l

n
e
t
w
o
r
k

g
a
m
e
s

h
a
v
e

b
e
c
o
m
e

h
u
g
e
l
y

p
o
p
u
l
a
r

s
i
n
c
e

t
h
e
i
r

l
a
u
n
c
h

i
n

2
0
0
7
,

w
h
e
n

t
h
e

d
o
m
i
n
a
n
t

s
o
c
i
a
l

n
e
t
w
o
r
k

p
l
a
t
f
o
r
m
,

F
a
c
e
b
o
o
k
,

r
e
l
e
a
s
e
d

i
t
s

a
p
p
l
i
c
a
t
i
o
n

p
r
o
g
r
a
m
m
i
n
g

i
n
t
e
r
f
a
c
e

(
A
P
I
)

f
o
r

t
h
e

t
h
i
r
d

p
a
r
t
y

d
e
v
e
l
o
p
e
r
s
.

C
o
i
n
e
d

a
s

s
o
c
i
a
l

g
a
m
e
s

b
y

t
h
e

i
n
d
u
s
t
r
y
,

t
h
e
s
e

g
a
m
e
s

a
r
e

p
l
a
y
e
d

b
y

m
i
l
l
i
o
n
s

o
f

p
l
a
y
e
r
s

o
n

d
a
i
l
y

b
a
s
i
s

m
a
k
i
n
g

s
o
c
i
a
l

g
a
m
e
s

o
n
e

o
f

t
h
e

f
a
s
t
e
s
t

g
r
o
w
i
n
g

s
e
g
m
e
n
t
s

i
n

t
h
e

b
u
s
i
n
e
s
s

[
1
1
]
.

S
o
c
i
a
l

g
a
m
e
s

a
r
e

b
a
s
e
d

o
n

s
o
c
i
a
l

n
e
t
w
o
r
k

p
l
a
t
f
o
r
m
s

a
n
d

t
h
e
y

u
t
i
l
i
z
e

f
r
e
e
-
t
o
-
p
l
a
y

r
e
v
e
n
u
e

m
o
d
e
l
,

p
r
o
v
i
d
i
n
g

e
a
s
y

a
c
c
e
s
s

t
o

p
l
a
y
.

S
i
n
c
e

t
h
e
i
r

l
a
u
n
c
h
,

s
o
c
i
a
l

g
a
m
e
s

h
a
v
e

g
a
i
n
e
d

a

l
o
t

o
f

h
y
p
e

a
n
d

a
t
t
e
n
t
i
o
n

i
n
s
i
d
e

t
h
e

g
a
m
e

i
n
d
u
s
t
r
y
.

T
h
e
i
r

n
a
t
u
r
e

a
n
d

v
a
l
u
e
s

h
a
v
e

b
e
e
n

d
i
s
c
u
s
s
e
d

w
i
d
e
l
y
.

O
n
e

d
i
s
c
u
s
s
i
o
n

t
o
p
i
c

i
s

t
h
e
i
r

s
o
c
i
a
b
i
l
i
t
y
.

S
o
c
i
a
l

g
a
m
e
s

a
r
e

s
o
c
i
a
l

i
n

t
h
e

s
e
n
s
e

t
h
a
t

t
h
e
y

a
r
e

b
a
s
e
d

o
n

a

s
o
c
i
a
l

n
e
t
w
o
r
k

p
l
a
t
f
o
r
m

a
n
d

o
f
t
e
n

u
t
i
l
i
z
e

t
h
e

p
l
a
y
e
r

s

f
r
i
e
n
d
s

f
o
r

p
l
a
y

p
u
r
p
o
s
e
s
.

H
o
w
e
v
e
r
,

t
h
e
r
e

h
a
v
e

b
e
e
n

d
i
s
c
u
s
s
i
o
n
s

s
a
y
i
n
g

t
h
a
t

t
h
e
s
e

g
a
m
e
s

a
r
e

n
o
t

t
r
u
l
y

s
o
c
i
a
l

a
n
d

t
h
a
t

t
h
e
r
e

i
s

m
u
c
h

m
o
r
e

p
o
t
e
n
t
i
a
l

t
o

i
n
c
r
e
a
s
e

t
h
e
i
r

s
o
c
i
a
b
i
l
i
t
y

[
2
,

9
]
.

A
s

t
h
e
s
e

g
a
m
e
s

a
r
e

n
e
w

a
n
d

t
h
e
i
r

d
e
s
i
g
n

p
r
a
c
t
i
c
e
s

a
r
e

e
v
o
l
v
i
n
g

r
a
p
i
d
l
y
,

t
h
e

s
o
c
i
a
l

s
i
d
e

o
f

s
o
c
i
a
l

g
a
m
e
s

d
e
s
e
r
v
e
s

a

c
l
o
s
e
r

l
o
o
k
.

I
n

t
h
i
s

p
o
s
i
t
i
o
n

p
a
p
e
r

w
e

e
x
a
m
i
n
e

t
h
e

s
o
c
i
a
b
i
l
i
t
y

o
f

s
o
c
i
a
l

g
a
m
e
s
.

W
e

a
p
p
r
o
a
c
h

t
h
e

i
s
s
u
e

f
r
o
m

t
h
r
e
e

d
i
s
t
i
n
c
t

p
e
r
s
p
e
c
t
i
v
e
s
:

a
c
a
d
e
m
i
a
,

i
n
d
u
s
t
r
y

a
n
d

p
l
a
y
e
r
s
.

T
h
e

p
u
r
p
o
s
e

i
s

t
o

m
a
p

t
h
e

d
i
s
c
u
s
s
i
o
n

a
n
d

d
e
s
i
g
n

p
r
a
c
t
i
c
e
s
,

t
h
u
s

t
o

u
n
d
e
r
s
t
a
n
d

t
h
e

c
u
r
r
e
n
t

s
t
a
t
e

o
f

s
o
c
i
a
b
i
l
i
t
y

i
n

s
o
c
i
a
l

g
a
m
e
s
.

A
l
s
o
,

t
h
i
s

a
n
a
l
y
s
i
s

a
i
m
s

t
o

r
e
c
o
g
n
i
z
e

t
h
e

n
e
g
l
e
c
t
e
d

a
r
e
a
s

o
f

s
o
c
i
a
b
i
l
i
t
y

t
h
a
t

c
o
u
l
d

b
e

u
s
e
d

i
n

t
h
e

d
e
s
i
g
n

o
f

s
o
c
i
a
l

g
a
m
e
s
.

W
e

b
e
l
i
e
v
e

o
u
r

a
n
a
l
y
s
i
s

i
s

b
e
n
e
f
i
c
i
a
l

f
o
r

t
h
e

d
e
s
i
g
n

a
n
d

e
v
a
l
u
a
t
i
o
n

o
f

s
o
c
i
a
b
i
l
i
t
y

i
n

s
o
c
i
a
l

g
a
m
e
s
.

A
c
a
d
e
m
i
a

A
p
p
r
o
a
c
h

S
t
e
n
r
o
s

e
t

a
l
.

h
a
v
e

c
r
e
a
t
e
d

a

f
r
a
m
e
w
o
r
k

f
o
r

s
o
c
i
a
l

i
n
t
e
r
a
c
t
i
o
n

i
n

g
a
m
e
s

[
1
4
]
.

I
n

g
e
n
e
r
a
l
,

s
o
c
i
a
b
i
l
i
t
y

c
a
n

e
m
e
r
g
e

b
o
t
h

i
n

a
n
d

o
u
t
s
i
d
e

a

g
a
m
e

a
n
d

i
t

i
s

r
e
l
a
t
e
d

t
o

p
r
e
s
e
n
c
e
,

i
n
t
e
r
a
c
t
i
o
n

a
n
d

c
o
m
m
u
n
i
c
a
t
i
o
n

b
e
t
w
e
e
n

t
h
e

p
l
a
y
e
r
s
.

S
o
c
i
a
b
i
l
i
t
y

i
s

n
a
t
u
r
a
l

p
a
r
t

o
f

g
a
m
e
s
,

w
h
i
c
h

f
e
a
t
u
r
e

t
w
o

o
r

m
o
r
e

p
l
a
y
e
r
s
,

b
u
t

i
t

c
a
n

b
e

f
o
u
n
d

i
n

s
i
n
g
l
e

p
l
a
y
e
r

g
a
m
e
s

a
s

w
e
l
l

[
1
4
]
.


O
n
e

o
f

t
h
e

i
n
t
e
r
e
s
t
i
n
g

a
s
p
e
c
t
s

o
f

t
h
e

f
r
a
m
e
w
o
r
k

i
s

s
o
c
i
a
b
i
l
i
t
y

o
f

t
h
e

m
a
s
s
i
v
e
l
y

s
i
n
g
l
e

p
l
a
y
e
r

g
a
m
e
s
,

i
.
e
.

s
o
c
i
a
l

g
a
m
e
s
.

I
n

t
h
e
s
e

g
a
m
e
s
,

n
u
m
e
r
o
u
s

p
l
a
y
e
r
s

a
r
e

p
l
a
y
i
n
g

t
h
e

g
a
m
e

a
s
y
n
c
h
r
o
n
o
u
s
l
y

i
n

p
a
r
a
l
l
e
l
.

E
a
c
h

p
l
a
y
e
r

h
a
s

t
h
e
i
r

o
w
n

p
e
r
s
o
n
a
l

g
a
m
e

s
p
a
c
e
,

a
n
d

t
h
e
y

c
a
n

v
i
s
i
t

t
h
e
i
r

f
r
i
e
n
d
s

(
u
s
u
a
l
l
y

c
a
l
l
e
d

n
e
i
g
h
b
o
r

)

g
a
m
e

s
p
a
c
e
s

a
n
d

e
n
a
c
t

s
i
m
p
l
e

t
a
s
k
s

s
u
c
h

a
s

s
e
n
d
i
n
g

g
i
f
t
s

o
r

h
e
l
p
i
n
g

w
i
t
h

c
h
o
r
e
s

t
o

a
i
d

t
h
e
i
r

f
r
i
e
n
d
s

p
r
o
g
r
e
s
s
i
n
g

i
n

t
h
e

g
a
m
e
.

[
1
4
,

1
5
]

S
o
c
i
a
l

g
a
m
e
s

c
a
n

a
l
s
o

u
t
i
l
i
z
e

t
h
e

p
l
a
y
e
r

s

s
o
c
i
a
l

n
e
t
w
o
r
k

t
o

c
r
e
a
t
e

p
e
r
s
o
n
a
l

c
o
n
t
e
n
t

i
n

t
h
e

g
a
m
e
.

W
a
n
d
e
r
i
n
g

n
o
n
-
p
l
a
y
e
r

c
h
a
r
a
c
t
e
r
s

m
i
g
h
t

b
e

n
a
m
e
d

a
f
t
e
r

p
l
a
y
e
r

s

f
r
i
e
n
d
s

o
r

t
h
e

p
l
a
y
e
r

c
a
n

a
s
s
i
g
n

f
r
i
e
n
d
s

f
o
r

d
i
f
f
e
r
e
n
t

r
o
l
e
s

i
n

t
h
e

g
a
m
e

s
p
a
c
e
,

w
i
t
h
o
u
t

l
e
t
t
i
n
g

h
e
r

f
r
i
e
n
d
s

k
n
o
w

a
b
o
u
t

i
t

[
1
1
]
.

A
l
t
h
o
u
g
h

t
h
e
r
e

m
i
g
h
t

n
o
t

b
e

r
i
c
h

a
n
d

s
y
n
c
h
r
o
n
o
u
s

s
o
c
i
a
l

i
n
t
e
r
a
c
t
i
o
n

b
e
t
w
e
e
n

t
h
e

p
l
a
y
e
r

a
n
d

h
e
r

f
r
i
e
n
d
s
,

t
h
e
r
e

i
s

t
h
e

s
e
n
s
e

o
f

p
e
r
s
o
n
a
l

p
r
e
s
e
n
c
e

i
n

t
h
e

g
a
m
e
.

A
n

i
n
t
e
r
e
s
t
i
n
g

o
b
s
e
r
v
a
t
i
o
n

i
s

t
h
a
t

a
l
t
h
o
u
g
h

s
o
c
i
a
b
i
l
i
t
y

s
e
e
m
s

t
o

b
e

s
h
a
l
l
o
w
,

p
r
o
g
r
e
s
s
i
n
g

i
n

a

g
a
m
e

m
i
g
h
t

b
e
c
o
m
e

d
e
p
e
n
d
e
n
t

o
n

t
h
e
s
e

s
h
a
l
l
o
w

s
o
c
i
a
l

a
c
t
i
o
n
s
.


A
t

t
h
e

b
e
g
i
n
n
i
n
g

t
h
e

p
l
a
y
e
r

c
a
n

p
l
a
y

t
h
e

g
a
m
e

a
l
o
n
e
,

b
u
t

a
s

t
h
e

g
a
m
e

a
d
v
a
n
c
e
s

t
h
e

p
l
a
y
e
r

n
e
e
d
s

t
o

h
a
v
e

m
o
r
e

f
r
i
e
n
d
s

i
n
v
o
l
v
e
d
,

w
h
o

w
i
l
l

h
e
l
p

i
n

c
o
m
p
l
e
t
i
n
g

t
a
s
k
s
,

o
r

t
h
e

p
l
a
y
e
r

n
e
e
d
s

t
o

h
e
l
p

o
t
h
e
r
s

[
1
5
]
.

T
h
i
s

c
a
n

c
r
e
a
t
e

a

DESVIG 2013
57
b
u
r
d
e
n

f
o
r

s
o
m
e

p
l
a
y
e
r
s
,

a
s

t
h
e
y

m
i
g
h
t

n
o
t

h
a
v
e

f
r
i
e
n
d
s

a
v
a
i
l
a
b
l
e

w
h
o

w
o
u
l
d

p
l
a
y

t
h
e

g
a
m
e
,

t
h
u
s

t
h
e

p
r
o
g
r
e
s
s

s
t
a
g
n
a
t
e
s

[
1
1
]
.

S
o
c
i
a
l

g
a
m
e
s

d
i
s
c
u
s
s
i
o
n

f
o
r
u
m
s

a
r
e

u
s
u
a
l
l
y

f
i
l
l
e
d

u
p

w
i
t
h

A
d
d

m
e
!

t
h
r
e
a
d
s

w
h
e
r
e

e
n
t
h
u
s
i
a
s
t
i
c

p
l
a
y
e
r
s

s
e
e
k

g
a
m
i
n
g

f
r
i
e
n
d
s

t
o

b
e

i
n
c
l
u
d
e
d

i
n
t
o

t
h
e
i
r

s
o
c
i
a
l

n
e
t
w
o
r
k
.


S
o
c
i
a
b
i
l
i
t
y

i
n
c
l
u
d
e
s

c
o
m
m
u
n
i
c
a
t
i
o
n

b
e
t
w
e
e
n

i
n
d
i
v
i
d
u
a
l
s
.

I
n

t
r
a
d
i
t
i
o
n
a
l

g
a
m
e
s
,

t
h
i
s

c
a
n

m
e
a
n

c
h
a
t
t
i
n
g

a
m
o
n
g

p
l
a
y
e
r
s

i
n

t
h
e

s
a
m
e

r
o
o
m

o
r

t
h
r
o
u
g
h

a
n

i
n
-
g
a
m
e

c
o
m
m
u
n
i
c
a
t
i
o
n

c
h
a
n
n
e
l

[
1
4
]
.

S
o
c
i
a
l

g
a
m
e
s

d
o

n
o
t

u
s
u
a
l
l
y

f
e
a
t
u
r
e

i
n
-
g
a
m
e

c
o
m
m
u
n
i
c
a
t
i
o
n

c
h
a
n
n
e
l
s

(
a
s
i
d
e

f
r
o
m

c
o
m
m
u
n
i
c
a
t
i
n
g

b
y

g
i
v
i
n
g

g
i
f
t
s

e
t
c
.
)
,

b
u
t

t
h
e
y

c
a
n

u
t
i
l
i
z
e

e
x
t
e
r
n
a
l

c
h
a
n
n
e
l
s
,

l
i
k
e

t
h
e

F
a
c
e
b
o
o
k

c
h
a
t
.

C
o
m
m
u
n
i
c
a
t
i
o
n

b
e
t
w
e
e
n

f
r
i
e
n
d
s

t
h
r
o
u
g
h

t
h
e

g
a
m
e

i
s

m
a
i
n
l
y

d
o
n
e

v
i
a

m
e
s
s
a
g
e
s

t
h
a
t

c
a
n

b
e

p
o
s
t
e
d

t
o

t
h
e

p
l
a
y
e
r

s

o
r

f
r
i
e
n
d
s

f
e
e
d
s

i
n

t
h
e

s
o
c
i
a
l

n
e
t
w
o
r
k

(
e
.
g
.

F
a
c
e
b
o
o
k

W
a
l
l
)
.

T
h
e
s
e

f
e
a
t
u
r
e
s

a
r
e

e
s
s
e
n
t
i
a
l

f
r
o
m

t
h
e

v
i
r
a
l

m
a
r
k
e
t
i
n
g

p
e
r
s
p
e
c
t
i
v
e
,

b
u
t

t
h
e
y

c
a
n

a
l
s
o

b
e

c
o
n
s
i
d
e
r
e
d

s
p
a
m
m
i
n
g

w
h
i
c
h

i
s

n
o
t

d
e
s
i
r
a
b
l
e

s
o
c
i
a
b
i
l
i
t
y

i
n

t
h
e

g
a
m
e

[
1
0
,

1
1
,

1
5
]
.

I
n
d
u
s
t
r
y

I
n
s
i
g
h
t
s

T
h
e

g
a
m
e

i
n
d
u
s
t
r
y

h
a
s

h
a
d

h
a
r
d

t
i
m
e

t
o

d
e
f
i
n
e

p
r
e
c
i
s
e
l
y

w
h
a
t

s
o
c
i
a
l

g
a
m
e
s

a
r
e

[
8
]
.

P
a
r
t
l
y
,

t
h
i
s

i
s

b
e
c
a
u
s
e

o
f

t
h
e

(
u
n
)
i
n
t
e
n
d
e
d

c
o
n
f
u
s
i
o
n

a
b
o
u
t

t
h
e

s
c
o
p
e

o
f

t
h
e

t
e
r
m

s
o
c
i
a
l

g
a
m
e
s

.

W
h
i
l
e

s
o
m
e

c
o
n
s
i
d
e
r

i
t

s
y
n
o
n
y
m
o
u
s

t
o

F
a
c
e
b
o
o
k

g
a
m
e
s
,

o
t
h
e
r
s

f
r
a
m
e

i
t

t
o

a

l
a
r
g
e
r

c
o
n
t
e
x
t

o
f

a
n
y

g
a
m
e

w
i
t
h

s
o
c
i
a
l

i
n
t
e
r
a
c
t
i
o
n

[
8
]
.


T
h
e
r
e

a
r
e

a
l
s
o

t
h
o
s
e

w
h
o

d
o

n
o
t

c
o
n
s
i
d
e
r

t
h
e
m

a
s

g
a
m
e
s

a
t

a
l
l
,

b
u
t

m
e
r
e
l
y

a
s

m
o
c
k
e
r
y

o
f

r
e
a
l

g
a
m
e
s

[
8
]
.

J

r
v
i
n
e
n

h
a
s

p
r
o
p
o
s
e
d

t
h
e

f
o
l
l
o
w
i
n
g

d
e
f
i
n
i
t
i
o
n

f
o
r

s
o
c
i
a
l

g
a
m
e
s
:

O
n
l
i
n
e

g
a
m
e
s

t
h
a
t

a
d
a
p
t

y
o
u
r

f
r
i
e
n
d
s
h
i
p

t
i
e
s

f
o
r

p
l
a
y

p
u
r
p
o
s
e
s
,

w
h
i
l
e

a
c
c
o
m
m
o
d
a
t
i
o
n

y
o
u
r

d
a
i
l
y

r
o
u
t
i
n
e
s

[
4
]
.

H
e
r
e
,

s
o
c
i
a
b
i
l
i
t
y

i
s

m
a
n
i
f
e
s
t
e
d

t
h
r
o
u
g
h

t
h
e

f
r
i
e
n
d
s
h
i
p

t
i
e
s

r
e
m
a
r
k
.

G
a
m
e

d
e
s
i
g
n
e
r
s

h
a
v
e

p
r
o
p
o
s
e
d

v
a
r
i
o
u
s

m
o
d
e
l
s

[
3
,

1
6
]

f
o
r

u
n
d
e
r
s
t
a
n
d
i
n
g

s
o
c
i
a
l

g
a
m
e
s

d
e
s
i
g
n
,

a
n
d

s
o
c
i
a
b
i
l
i
t
y

h
a
s

b
e
e
n

i
d
e
n
t
i
f
i
e
d

a
s

a

k
e
y

f
e
a
t
u
r
e

o
f

t
h
e
s
e

g
a
m
e
s
.

T
y
p
i
c
a
l
l
y
,

s
o
c
i
a
b
i
l
i
t
y

i
s

d
i
s
c
u
s
s
e
d

i
n

r
e
l
a
t
i
o
n

t
o

a
s
y
n
c
h
r
o
n
o
u
s

g
a
m
e
p
l
a
y
,

r
e
c
i
p
r
o
c
i
t
y
,

c
o
l
l
a
b
o
r
a
t
i
o
n

a
n
d

c
o
m
p
e
t
i
t
i
o
n
.

A
l
t
h
o
u
g
h

s
o
c
i
a
b
i
l
i
t
y

h
a
s

b
e
e
n

s
e
e
n

a
s

a
n

i
m
p
o
r
t
a
n
t

a
s
p
e
c
t

i
n

s
o
c
i
a
l

g
a
m
e
s
,

i
n
t
e
r
e
s
t
i
n
g
l
y

t
h
e

i
n
d
u
s
t
r
y

d
i
s
c
u
s
s
i
o
n

h
a
s

b
e
e
n

f
o
c
u
s
e
d

o
n

t
h
e

s
h
a
l
l
o
w

s
o
c
i
a
b
i
l
i
t
y

i
n

t
h
e
s
e

g
a
m
e
s
.

T
h
e

l
a
c
k

o
f

r
e
a
l

s
o
c
i
a
b
i
l
i
t
y

h
a
s

b
e
e
n

c
o
n
n
e
c
t
e
d

t
o

t
h
e

a
s
y
n
c
h
r
o
n
o
u
s

g
a
m
e
p
l
a
y
,

w
h
i
c
h

a
l
l
o
w
s

t
h
e

p
l
a
y
e
r
s

t
o

p
l
a
y

w
i
t
h
o
u
t

a
l
l

p
a
r
t
i
e
s

b
e
i
n
g

p
r
e
s
e
n
t

i
n

t
h
e

g
a
m
e

a
t

t
h
e

s
a
m
e

t
i
m
e

[
1
3
]
.

A
b
i
l
i
t
y

t
o

p
l
a
y

t
o
g
e
t
h
e
r

i
n

a

s
h
a
r
e
d

p
h
y
s
i
c
a
l

s
p
a
c
e

o
r

i
n

r
e
a
l
-
t
i
m
e

c
o
n
c
u
r
r
e
n
t
l
y

h
a
s

b
e
e
n

s
e
e
n

a
s

t
h
e

f
u
n
d
a
m
e
n
t
a
l

a
s
p
e
c
t
s

o
f

a

t
r
u
l
y

s
o
c
i
a
l

g
a
m
e

[
7
,

1
2
]
.

R
e
c
e
n
t
l
y

t
h
e
r
e

h
a
v
e

b
e
e
n

a
m
b
i
t
i
o
u
s

a
t
t
e
m
p
t
s

t
o

b
r
i
n
g

s
y
n
c
h
r
o
n
o
u
s

g
a
m
e
p
l
a
y

i
n
t
o

t
h
e

s
o
c
i
a
l

g
a
m
e
s

s
p
a
c
e

a
s

w
e
l
l

[
6
]
.

T
h
e
r
e

i
s

a
l
s
o

a
n

e
v
o
l
u
t
i
o
n
a
r
y

p
e
r
s
p
e
c
t
i
v
e

t
h
a
t

s
o
c
i
a
l

g
a
m
e
s

a
r
e

s
t
i
l
l

i
n

t
h
e

i
n
f
a
n
c
y

a
n
d

b
e
t
t
e
r

p
r
a
c
t
i
c
e
s

a
r
e

p
i
c
k
e
d

u
p

a
s

t
h
e

g
e
n
r
e

e
v
o
l
v
e
s

[
2
,

9
]
.

T
h
i
s

d
i
s
c
u
s
s
i
o
n

h
a
s

b
e
e
n

a
p
p
a
r
e
n
t

i
n

e
a
r
l
i
e
r

g
a
m
e

r
e
l
e
a
s
e
s
,

w
h
i
c
h

h
a
v
e

b
e
e
n

m
o
r
e

s
o
c
i
a
l

t
h
e
n

t
h
e
i
r

p
r
e
d
e
c
e
s
s
o
r
s

[
1
,

1
5
]
.

H
o
w
e
v
e
r
,

n
o
t

e
v
e
r
y
o
n
e

a
g
r
e
e
s

t
h
a
t

s
o
c
i
a
l

g
a
m
e
s

s
h
o
u
l
d

b
e

m
o
r
e

s
o
c
i
a
l
,

a
s

i
t

m
i
g
h
t

l
e
a
d

t
o

l
o
s
i
n
g

t
h
e

c
a
s
u
a
l

f
e
e
l

o
f

t
h
e

g
a
m
e
p
l
a
y
,

i
f

t
h
e
y

a
r
e

b
u
r
d
e
n
e
d

w
i
t
h

t
o
o

m
a
n
y

s
o
c
i
a
l

f
e
a
t
u
r
e
s
.

[
5
]

DESVIG 2013
58
P
l
a
y
e
r
s

P
e
r
s
p
e
c
t
i
v
e

I
n

a
n
o
t
h
e
r

s
t
u
d
y

[
1
1
]

w
e

i
n
t
e
r
v
i
e
w
e
d

1
8

F
a
c
e
b
o
o
k

u
s
e
r
s

(
1
4

m
a
l
e
,

4

f
e
m
a
l
e
,

a
v
e
r
a
g
e

a
g
e

3
1

y
e
a
r
s
)
,

f
o
c
u
s
i
n
g

o
n

h
o
w

s
o
c
i
a
l

g
a
m
e
s

a
r
e

p
e
r
c
e
i
v
e
d

a
n
d

p
l
a
y
e
d
.

T
h
e

i
n
t
e
r
v
i
e
w
e
e
s

s
a
w

s
o
c
i
a
b
i
l
i
t
y

a
s

a
n

i
m
p
o
r
t
a
n
t
,

y
e
t

s
h
a
l
l
o
w
,

p
a
r
t

o
f

g
a
m
i
n
g

a
n
d

f
o
r

s
o
m
e

p
l
a
y
e
r
s

i
t

w
a
s

c
o
n
s
i
d
e
r
e
d

t
o

b
e

m
o
r
e

i
m
p
o
r
t
a
n
t

t
h
a
n

i
n

o
t
h
e
r

t
y
p
e
s

o
f

g
a
m
e
s
.

W
e
l
l
-
d
e
s
i
g
n
e
d

s
o
c
i
a
b
i
l
i
t
y

c
a
n

i
n

s
o
m
e

c
a
s
e
s

e
v
e
n

o
u
t
w
e
i
g
h

p
o
o
r

g
a
m
e

d
e
s
i
g
n

i
f

t
h
e
r
e

i
s

a

n
i
c
e

g
r
o
u
p

o
f

f
r
i
e
n
d
s

p
l
a
y
i
n
g

t
h
e

g
a
m
e
.


T
h
e

p
l
a
y
e
r
s

p
e
r
c
e
p
t
i
o
n
s

o
n

s
o
c
i
a
b
i
l
i
t
y

f
o
l
l
o
w

t
h
e

i
n
d
u
s
t
r
y

d
i
s
c
u
s
s
i
o
n
:

s
o
c
i
a
l

g
a
m
e
s

h
a
v
e

l
i
t
t
l
e

r
e
a
l

s
o
c
i
a
b
i
l
i
t
y

i
n

r
e
g
a
r
d
s

t
o

s
o
c
i
a
l

i
n
t
e
r
a
c
t
i
o
n

t
h
r
o
u
g
h

g
a
m
e

m
e
c
h
a
n
i
c
s
.

I
n

a
d
d
i
t
i
o
n
,

t
h
e

g
a
m
e
s

w
e
r
e

p
e
r
c
e
i
v
e
d

a
s

s
i
n
g
l
e

p
l
a
y
e
r

g
a
m
e
s

w
i
t
h

a

s
o
c
i
a
l

t
w
i
s
t
.

H
o
w
e
v
e
r
,

e
v
e
n

t
h
o
u
g
h

t
h
e

l
e
v
e
l

o
f

s
o
c
i
a
b
i
l
i
t
y

o
n

g
a
m
e

m
e
c
h
a
n
i
c
s

l
e
v
e
l

w
a
s

p
e
r
c
e
i
v
e
d

a
s

l
o
w
,

i
t

s
t
i
l
l

h
a
d

v
a
l
u
e

f
o
r

t
h
e

p
l
a
y
e
r
s

i
n

r
e
g
a
r
d
s

t
o

s
o
c
i
a
l

r
e
l
a
t
i
o
n
s
h
i
p
s
.

F
o
r

e
x
a
m
p
l
e
,

a

t
y
p
i
c
a
l

t
a
s
k

o
f

s
e
n
d
i
n
g

g
i
f
t
s

t
o

f
e
l
l
o
w

p
l
a
y
e
r
s

c
a
n

l
e
a
d

t
o

s
t
r
o
n
g
e
r

t
i
e
s

b
e
t
w
e
e
n

t
h
e
m
,

c
r
e
a
t
i
n
g

t
r
u
s
t

a
n
d

s
o
c
i
a
l

p
r
e
f
e
r
e
n
c
e
.

H
e
l
p
i
n
g

t
r
u
s
t
e
d

f
e
l
l
o
w

p
l
a
y
e
r
s

i
n

t
h
e

g
a
m
e

c
a
n

b
e

a

m
o
t
i
v
a
t
i
n
g

f
a
c
t
o
r

i
n

i
t
s
e
l
f

a
n
d

t
h
e

p
l
a
y
e
r

k
e
e
p
s

o
n

p
l
a
y
i
n
g

t
h
e

g
a
m
e
,

a
l
t
h
o
u
g
h

t
h
e

p
l
a
y
e
r

i
s

n
o

l
o
n
g
e
r

i
n
t
e
r
e
s
t
e
d

i
n

t
h
e

g
a
m
e

a
s

s
u
c
h
.

A
n

i
n
t
e
r
e
s
t
i
n
g

n
o
t
i
o
n

w
a
s

t
h
a
t

p
l
a
y
e
r
s

s
a
w

s
o
c
i
a
b
i
l
i
t
y

m
a
i
n
l
y

s
u
p
p
o
r
t
i
n
g

t
h
e

c
o
l
l
a
b
o
r
a
t
i
v
e

s
i
d
e

o
f

g
a
m
i
n
g

a
n
d

s
o
m
e

i
n
t
e
r
v
i
e
w
e
e
s

w
o
u
l
d

h
a
v
e

l
i
k
e
d

t
o

s
e
e

m
o
r
e

c
o
n
f
l
i
c
t

m
e
c
h
a
n
i
c
s
,

w
h
i
c
h

c
o
u
l
d

b
e

u
s
e
d

a
g
a
i
n
s
t

o
t
h
e
r

p
l
a
y
e
r
s
.

T
h
e

i
n
t
e
r
v
i
e
w
e
e
s

a
l
s
o

c
o
m
m
e
n
t
e
d

t
h
a
t

s
o
c
i
a
l

i
n
t
e
r
a
c
t
i
o
n

i
s

n
o
t

a

p
r
e
r
e
q
u
i
s
i
t
e

f
o
r

s
o
c
i
a
l

f
u
n
.

S
t
e
n
r
o
s

e
t

a
l
.

s
t
a
t
e
d

t
h
a
t

o
n
e

o
f

t
h
e

k
e
y

f
o
r
m
s

o
f

s
o
c
i
a
b
i
l
i
t
y

w
a
s

t
h
e

s
o
c
i
a
l

p
r
e
s
e
n
c
e

i
n

t
h
e

g
a
m
e

a
n
d

u
s
i
n
g

s
o
c
i
a
l

g
a
m
e

c
o
n
t
e
n
t
,

w
h
i
c
h

e
m
e
r
g
e
s

f
r
o
m

t
h
e

s
o
c
i
a
l

n
e
t
w
o
r
k

[
1
4
]
.

I
n
t
e
r
v
i
e
w
e
e
s

i
n

t
h
e

s
t
u
d
y

s
t
a
t
e
d

t
h
a
t

a
s
s
i
g
n
i
n
g

f
r
i
e
n
d
s

t
o

n
o
m
i
n
a
l

r
o
l
e
s

i
n

t
h
e

g
a
m
e

w
a
s

c
o
n
s
i
d
e
r
e
d

f
u
n

a
n
d

i
t

a
d
d
e
d

v
a
l
u
e

t
o

t
h
e

g
a
m
e
,

b
u
t

i
t

w
a
s

n
o
t

s
e
e
n

a
s

v
e
r
y

s
o
c
i
a
l
.

I
n
c
r
e
a
s
i
n
g

d
e
m
a
n
d

o
f

i
n
v
i
t
i
n
g

f
r
i
e
n
d
s

t
o

j
o
i
n

t
h
e

g
a
m
e

a
n
d

r
e
c
e
i
v
i
n
g

s
u
c
h

r
e
q
u
e
s
t
s

f
r
o
m

f
r
i
e
n
d
s

w
a
s

d
e
s
c
r
i
b
e
d

s
o
m
e
t
i
m
e
s

a
s

h
e
l
l
i
s
h

a
n
n
o
y
a
n
c
e

.

P
r
o
g
r
e
s
s
i
o
n

i
n

s
o
c
i
a
l

g
a
m
e
s

w
a
s

c
o
n
s
i
d
e
r
e
d

t
o

b
e

t
o
o

d
e
p
e
n
d
e
n
t

o
n

t
h
e

a
m
o
u
n
t

o
f

p
l
a
y
i
n
g

f
r
i
e
n
d
s
,

a
n
d

s
p
a
m
m
i
n
g

f
r
i
e
n
d
s

w
i
t
h

r
e
q
u
e
s
t
s

w
a
s

s
e
e
n

a
s

i
m
p
o
l
i
t
e
.

O
n
e

e
x
p
l
a
n
a
t
i
o
n

f
o
r

s
u
c
h

o
p
i
n
i
o
n

m
i
g
h
t

b
e

t
h
e

p
e
r
c
e
p
t
i
o
n

o
f

s
o
c
i
a
l

g
a
m
e
s

b
e
i
n
g

m
a
i
n
l
y

s
i
n
g
l
e

p
l
a
y
e
r

g
a
m
e
s
,

a
s

s
t
a
t
e
d

e
a
r
l
i
e
r
.

A
s

v
i
r
a
l

d
i
s
t
r
i
b
u
t
i
o
n

o
f

t
h
e

g
a
m
e

i
s

a
n

e
s
s
e
n
t
i
a
l

f
e
a
t
u
r
e

f
r
o
m

t
h
e

b
u
s
i
n
e
s
s

p
e
r
s
p
e
c
t
i
v
e
,

i
t

w
a
s

a
l
s
o

t
h
e

m
o
s
t

c
o
m
m
o
n

c
a
u
s
e

o
f

f
r
u
s
t
r
a
t
i
o
n

f
o
r

t
h
e

p
l
a
y
e
r
s
.

D
i
f
f
e
r
e
n
t

n
o
t
i
f
i
c
a
t
i
o
n
s
,

r
e
q
u
e
s
t
s
,

n
e
w
s

f
e
e
d

i
t
e
m
s

a
n
d

w
a
l
l

p
o
s
t
s

w
e
r
e

c
o
n
s
i
d
e
r
e
d

a
s

s
p
a
m
.

I
t

w
a
s

e
s
p
e
c
i
a
l
l
y

i
r
r
i
t
a
t
i
n
g

w
h
e
n

n
o
t
i
f
i
c
a
t
i
o
n
s

c
a
m
e

f
r
o
m

a

g
a
m
e

t
h
a
t

a

p
l
a
y
e
r

d
i
d

n
o
t

e
v
e
n

p
l
a
y
.

I
t

r
e
s
u
l
t
e
d

t
h
a
t

t
h
e

p
l
a
y
e
r
s

b
l
o
c
k
e
d

t
h
e

w
h
o
l
e

g
a
m
e

t
o

s
t
o
p

t
h
o
s
e

n
o
t
i
f
i
c
a
t
i
o
n
s
.

O
n

t
h
e

o
t
h
e
r

h
a
n
d
,

i
n
t
e
r
v
i
e
w
e
e
s

a
d
m
i
t
t
e
d

t
h
a
t

v
i
r
a
l

m
e
c
h
a
n
i
s
m
s

h
a
d

a

p
o
s
i
t
i
v
e

i
n
f
l
u
e
n
c
e

a
s

w
e
l
l
.

R
e
c
o
m
m
e
n
d
a
t
i
o
n
s

f
r
o
m

t
r
u
s
t
w
o
r
t
h
y

p
l
a
y
e
r
s

w
e
r
e

a
n

i
m
p
o
r
t
a
n
t

s
o
u
r
c
e

o
f

f
i
n
d
i
n
g

n
e
w

g
a
m
e
s
.

I
n

o
v
e
r
a
l
l
,

i
t

s
e
e
m
s

t
h
a
t

i
n
f
l
u
e
n
c
e

o
f

o
t
h
e
r

p
l
a
y
e
r
s

i
s

a

s
t
r
o
n
g

f
a
c
t
o
r

o
f

s
o
c
i
a
b
i
l
i
t
y
,

w
h
i
c
h

h
a
s

n
o
t

b
e
e
n

u
t
i
l
i
z
e
d

i
n

f
u
l
l

e
x
t
e
n
t
.

I
n
t
e
r
v
i
e
w
e
e
s

d
e
s
c
r
i
b
e
d

t
h
a
t

t
h
e

b
e
h
a
v
i
o
r

o
f

t
h
e
i
r

f
r
i
e
n
d
s

w
i
l
l

h
a
v
e

a
n

i
n
f
l
u
e
n
c
e

o
n

t
h
e
i
r

g
a
m
i
n
g

p
r
a
c
t
i
c
e
s
.

I
f

f
r
i
e
n
d
s

q
u
i
t

p
l
a
y
i
n
g

t
h
e

g
a
m
e
,

t
h
e
r
e

m
i
g
h
t

n
o
t

b
e

e
n
o
u
g
h

a
c
t
i
v
e
l
y

p
l
a
y
i
n
g

f
r
i
e
n
d
s

t
o

c
o
n
t
i
n
u
e

p
r
o
g
r
e
s
s
i
n
g

i
n

t
h
e

g
a
m
e
.

T
h
e
r
e
f
o
r
e
,

i
t

i
s

l
i
k
e
l
y

f
o
r

t
h
e

p
l
a
y
e
r

t
o

q
u
i
t

p
l
a
y
i
n
g

a
s

w
e
l
l
.

S
o
m
e
t
i
m
e
s

t
h
i
s

w
a
s

c
o
u
n
t
e
r
e
d

b
y

a
d
d
i
n
g

s
t
r
a
n
g
e
r
s

i
n
t
o

t
h
e

s
o
c
i
a
l

n
e
t
w
o
r
k

DESVIG 2013
59
f
o
r

p
l
a
y

p
u
r
p
o
s
e
s
,

b
u
t

t
h
e
n

t
h
e

p
e
r
s
o
n
a
l

s
o
c
i
a
l

p
r
e
s
e
n
c
e

w
a
s

l
o
s
t

[
1
1
]
.

D
i
s
c
u
s
s
i
o
n

T
h
e

s
o
c
i
a
l

f
e
a
t
u
r
e
s

i
n

s
o
c
i
a
l

g
a
m
e
s

h
a
v
e

e
v
o
l
v
e
d

s
i
n
c
e

t
h
e
i
r

l
a
u
n
c
h

i
n

2
0
0
7
.

T
h
e
y

s
t
a
r
t
e
d

o
u
t

a
s

v
i
r
a
l

c
h
a
i
n
-
l
e
t
t
e
r

g
a
m
e
s

w
h
e
r
e

t
h
e

o
n
l
y

a
c
t
i
o
n

w
a
s

t
o

b
i
t
e

y
o
u
r

f
r
i
e
n
d
,

t
h
u
s

c
r
e
a
t
i
n
g

a

p
y
r
a
m
i
d

s
c
h
e
m
e

g
a
m
e
p
l
a
y
.

N
o
w

t
h
e
r
e

i
s

m
o
r
e

s
o
c
i
a
l

d
e
p
t
h
,

w
h
i
c
h

i
n

s
o
m
e

g
a
m
e
s

i
s

e
v
e
n

m
e
a
s
u
r
e
d

b
y

t
h
e

s
o
c
i
a
l

e
x
p
e
r
i
e
n
c
e

l
e
v
e
l

,

w
h
i
c
h

i
n
d
i
c
a
t
e
s

h
o
w

h
e
l
p
f
u
l

t
h
e

p
l
a
y
e
r

h
a
s

b
e
e
n

f
o
r

o
t
h
e
r
s
.


S
o
c
i
a
b
i
l
i
t
y

i
n

s
o
c
i
a
l

g
a
m
e
s

h
a
s

b
e
c
o
m
e

s
o
m
e

k
i
n
d

o
f

a

p
a
r
a
d
o
x
.

O
n

o
n
e

h
a
n
d
,

t
h
e

f
r
e
e
-
t
o
-
p
l
a
y

m
o
d
e
l

r
e
q
u
i
r
e
s

c
e
r
t
a
i
n

v
i
r
a
l

g
r
o
w
t
h

f
o
r

t
h
e

g
a
m
e

t
o

b
e

s
u
c
c
e
s
s
f
u
l
.

T
h
e
n

o
n

t
h
e

o
t
h
e
r

h
a
n
d
,

p
l
a
y
e
r
s

d
i
s
l
i
k
e

e
n
f
o
r
c
e
d

s
o
c
i
a
b
i
l
i
t
y
,

w
h
i
c
h

r
e
s
u
l
t
s

t
o

s
e
n
d
i
n
g

a
n
d

r
e
c
e
i
v
i
n
g

a
n
n
o
y
i
n
g

r
e
q
u
e
s
t
s

o
r

s
t
a
g
n
a
t
i
o
n

d
u
e

t
o

t
h
e

l
a
c
k

o
f

f
r
i
e
n
d
s
.

B
a
s
e
d

o
n

t
h
i
s

f
i
n
d
i
n
g
,

t
h
e
r
e

s
h
o
u
l
d

b
e

a

b
a
c
k

d
o
o
r

m
e
c
h
a
n
i
c

f
o
r

a
l
l
o
w
i
n
g

p
l
a
y
e
r
s

t
o

p
r
o
g
r
e
s
s

i
n

t
h
e

g
a
m
e

i
n

t
h
e

c
a
s
e

w
h
e
r
e

t
h
e
r
e

a
r
e

n
o

f
r
i
e
n
d
s

a
v
a
i
l
a
b
l
e
.

I
n

s
o
m
e

g
a
m
e
s
,

p
l
a
y
e
r
s

c
a
n

b
u
y

t
h
e
m
s
e
l
v
e
s

o
u
t

o
f

t
h
e
s
e

s
i
t
u
a
t
i
o
n
s

b
u
t

a
s

t
h
e

m
a
j
o
r
i
t
y

o
f

p
l
a
y
e
r
s

a
r
e

r
e
l
u
c
t
a
n
t

t
o

p
a
y

m
i
c
r
o
p
a
y
m
e
n
t
s
,

t
h
e
r
e

s
h
o
u
l
d

b
e

o
t
h
e
r

o
p
t
i
o
n
s

a
s

n
o
n
-
p
a
y
e
r
s

h
a
v
e

b
e
e
n

f
o
u
n
d

t
o

b
e

i
m
p
o
r
t
a
n
t

p
a
r
t

o
f

t
h
e

p
l
a
y
e
r

b
a
s
e

[
1
5
]
.


A
d
d
i
n
g

m
o
r
e

s
o
c
i
a
l

f
e
a
t
u
r
e
s

i
n

g
a
m
e
p
l
a
y

c
a
n

b
e

a

d
e
s
i
g
n

t
r
a
d
e
-
o
f
f

a
s

t
h
e

m
o
r
e

c
o
m
p
l
e
x

t
h
e

g
a
m
e

i
s
,

t
h
e

m
o
r
e

i
t

m
i
g
h
t

l
o
s
e

i
t
s

c
a
s
u
a
l
,

e
a
s
y

f
e
e
l

[
5
]
.

U
l
t
i
m
a
t
e
l
y

t
h
e

g
a
m
e

c
o
n
c
e
p
t

a
n
d

t
h
e

c
o
n
t
e
x
t

o
f

p
l
a
y

d
e
f
i
n
e

t
h
e

s
w
e
e
t

s
p
o
t

f
o
r

s
o
c
i
a
b
i
l
i
t
y
.

T
h
e
r
e

i
s

d
e
f
i
n
i
t
e
l
y

r
o
o
m

f
o
r

a
l
l

k
i
n
d
s

o
f

s
o
c
i
a
l

g
a
m
e
s
,

s
h
a
l
l
o
w

a
n
d

d
e
e
p
.

S
o
c
i
a
b
i
l
i
t
y

i
n

i
t
s
e
l
f

i
s

n
e
i
t
h
e
r

g
o
o
d

n
o
r

b
a
d
,

b
u
t

i
t

c
a
n

b
e

d
e
s
i
g
n
e
d

w
e
l
l

o
r

p
o
o
r
l
y
.

B
a
s
e
d

o
n

o
u
r

i
n
t
e
r
v
i
e
w
s

a
n
d

e
x
p
e
r
i
e
n
c
e
s

i
n

a
n
a
l
y
z
i
n
g

s
o
c
i
a
l

g
a
m
e
s

[
1
0
,

1
1
]
,

t
h
e

c
o
m
p
e
t
i
t
i
v
e

s
i
d
e

o
f

s
o
c
i
a
l

g
a
m
e
s

i
s

s
o
m
e
t
h
i
n
g

t
h
a
t

c
o
u
l
d

b
e

s
t
u
d
i
e
d

f
u
r
t
h
e
r
.

G
e
n
e
r
a
l
l
y

s
p
e
a
k
i
n
g

c
o
n
f
l
i
c
t

m
e
c
h
a
n
i
c
s

h
a
v
e

n
o
t

b
e
e
n

p
o
p
u
l
a
r

i
n

p
o
p
u
l
a
r

w
e
s
t
e
r
n

s
o
c
i
a
l

g
a
m
e
s
,

b
u
t

i
n

o
t
h
e
r

r
e
g
i
o
n
s
,

l
i
k
e

C
h
i
n
a
,

t
h
e
y

a
r
e

p
a
r
t

o
f

t
h
e

c
o
r
e

m
e
c
h
a
n
i
c
s

[
1
7
]
.

I
n

t
h
i
s

p
o
s
i
t
i
o
n

p
a
p
e
r

w
e

h
a
v
e

f
u
n
n
e
l
e
d

f
r
o
m

t
o
p

t
o

b
o
t
t
o
m

b
y

l
o
o
k
i
n
g

s
o
c
i
a
l

g
a
m
e
s

w
i
t
h

d
i
f
f
e
r
e
n
t

p
e
r
s
p
e
c
t
i
v
e
s
.

T
h
e

a
c
a
d
e
m
i
a

p
r
o
v
i
d
e
s

t
h
e

c
o
m
p
r
e
h
e
n
s
i
v
e

t
h
e
o
r
y

o
n

s
o
c
i
a
b
i
l
i
t
y

i
n

g
a
m
e
s
,

t
h
e

i
n
d
u
s
t
r
y

f
o
c
u
s
e
s

o
n

t
h
e

c
e
r
t
a
i
n

a
s
p
e
c
t
s

w
i
t
h
i
n

a

g
e
n
r
e
,

a
n
d

t
h
e

p
l
a
y
e
r
s

p
r
o
v
i
d
e

t
h
e
i
r

o
w
n

p
e
r
c
e
p
t
i
o
n
s

o
n

i
n
d
i
v
i
d
u
a
l

g
a
m
e
s
.

T
h
e

n
e
x
t

l
o
g
i
c
a
l

s
t
e
p

i
s

t
o

a
n
a
l
y
z
e

t
h
e

g
a
m
e
s

a
n
d

t
h
e
i
r

m
e
c
h
a
n
i
c
s

i
n

d
e
t
a
i
l
.

A

u
s
e
f
u
l

a
p
p
r
o
a
c
h

w
o
u
l
d

b
e

t
o

u
n
d
e
r
s
t
a
n
d

t
h
e

c
o
r
e

c
o
n
c
e
p
t
s

a
n
d

t
h
e

c
o
m
m
o
n

g
a
m
e

m
e
c
h
a
n
i
c
s

w
h
i
c
h

a
r
e

r
e
l
a
t
e
d

t
o

s
o
c
i
a
b
i
l
i
t
y
.

T
h
e

c
o
r
e

c
o
n
c
e
p
t
s

a
r
e

h
i
g
h

l
e
v
e
l

d
e
s
i
g
n

p
a
r
a
d
i
g
m
s

l
i
k
e

c
o
m
p
e
t
i
t
i
o
n
,

c
o
l
l
a
b
o
r
a
t
i
o
n
,

r
e
c
i
p
r
o
c
i
t
y

e
t
c
.

T
h
e

c
o
m
m
o
n

m
e
c
h
a
n
i
c
s

i
n
c
l
u
d
e

s
i
n
g
l
e

s
o
c
i
a
l

a
c
t
i
o
n
s

l
i
k
e

s
e
n
d
i
n
g

a

g
i
f
t
,

a
d
d
i
n
g

a

n
e
i
g
h
b
o
r
,

p
o
s
t
i
n
g

a

m
e
s
s
a
g
e

e
t
c
.

E
a
c
h

i
n
d
i
v
i
d
u
a
l

s
u
b
-
g
e
n
r
e

a
n
d

g
a
m
e

h
a
s

t
h
e
i
r

o
w
n

d
i
s
t
i
n
c
t

s
o
c
i
a
l

m
e
c
h
a
n
i
c
s

a
s

w
e
l
l
.

A
n
a
l
y
z
i
n
g

s
o
c
i
a
b
i
l
i
t
y

i
n

t
h
i
s

w
a
y

i
s

b
e
n
e
f
i
c
i
a
l

f
o
r

t
w
o

r
e
a
s
o
n
s
.

F
i
r
s
t
,

i
t

s
t
r
u
c
t
u
r
e
s

t
h
e

s
o
c
i
a
l

d
e
s
i
g
n

i
n

s
o
c
i
a
l

g
a
m
e
s
.

S
e
c
o
n
d
,

s
u
c
h

a
n
a
l
y
s
i
s

c
a
n

b
e

u
s
e
d

a
s

i
n
s
p
i
r
a
t
i
o
n

f
o
r

d
e
s
i
g
n
i
n
g

s
o
c
i
a
b
i
l
i
t
y
.

T
h
i
s

a
n
a
l
y
s
i
s

r
e
q
u
i
r
e
s

f
u
r
t
h
e
r

r
e
s
e
a
r
c
h

a
n
d

i
s

n
o
t

w
i
t
h
i
n

t
h
e

s
c
o
p
e

o
f

t
h
i
s

p
o
s
i
t
i
o
n

p
a
p
e
r
.

D
u
e

t
h
e

n
o
v
e
l
t
y

o
f

t
h
e

g
e
n
r
e
,

t
h
e
r
e

i
s

n
o

r
i
g
o
r
o
u
s

r
e
s
e
a
r
c
h

d
o
n
e

o
n

t
h
e

s
o
c
i
a
b
i
l
i
t
y

o
f

s
o
c
i
a
l

g
a
m
e
s
,

t
h
u
s

m
a
k
i
n
g

t
h
i
s

r
e
s
e
a
r
c
h

t
o
p
i
c

i
n
t
e
r
e
s
t
i
n
g
,

t
i
m
e
l
y

a
n
d

n
e
e
d
e
d
.

C
r
e
a
t
i
n
g

s
u
c
h

b
a
s
i
s

i
s

n
e
c
e
s
s
a
r
y

f
o
r

d
e
v
e
l
o
p
i
n
g

a

s
o
c
i
a
b
i
l
i
t
y

e
v
a
l
u
a
t
i
o
n

t
o
o
l

f
o
r

t
h
e

a
c
a
d
e
m
i
a

a
n
d

i
n
d
u
s
t
r
y
.

DESVIG 2013
60
R
e
f
e
r
e
n
c
e
s

[
1
]

E
A
.

S
i
m
C
i
t
y

S
o
c
i
a
l

B
r
e
a
k
s

G
r
o
u
n
d

o
n

F
a
c
e
b
o
o
k
.

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m

i
n
v
e
s
t
o
r
.
e
a
.
c
o
m
:

h
t
t
p
:
/
/
i
n
v
e
s
t
o
r
.
e
a
.
c
o
m
/
r
e
l
e
a
s
e
d
e
t
a
i
l
.
c
f
m
?
R
e
l
e
a
s
e
I
D
=
6
7
9
8
3
5

[
2
]

B
r
i
g
h
t
m
a
n
,

J
.

Z
y
n
g
a
:

S
o
c
i
a
l

g
a
m
e
s

a
r
e

s
t
i
l
l

n
o
t

t
r
u
l
y

s
o
c
i
a
l
.

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m

G
a
m
e
i
n
d
u
s
t
r
y
.
b
i
z
:

h
t
t
p
:
/
/
w
w
w
.
g
a
m
e
s
i
n
d
u
s
t
r
y
.
b
i
z
/
a
r
t
i
c
l
e
s
/
2
0
1
2
-
0
4
-
1
1
-
z
y
n
g
a
-
s
o
c
i
a
l
-
g
a
m
e
s
-
a
r
e
-
s
t
i
l
l
-
n
o
t
-
t
r
u
l
y
-
s
o
c
i
a
l

[
3
]

J

r
v
i
n
e
n
,

A
.

G
a
m
e

D
e
s
i
g
n

f
o
r

S
o
c
i
a
l

N
e
t
w
o
r
k
s
:

I
n
t
e
r
a
c
t
i
o
n

D
e
s
i
g
n

f
o
r

P
l
a
y
f
u
l

D
i
s
p
o
s
i
t
i
o
n
.

I
n

P
r
o
c
.

S
a
n
d
b
o
x

0
9
,

A
C
M

P
r
e
s
s

(
2
0
0
9
)
,

9
5
-
1
0
2
.


[
4
]

J

r
v
i
n
e
n
,

A
.

S
t
a
t
e

o
f

S
o
c
i
a
l

i
n

S
o
c
i
a
l

G
a
m
e
s
.

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m
:

h
t
t
p
:
/
/
c
a
s
u
a
l
c
o
n
n
e
c
t
.
o
r
g
/
l
e
c
t
u
r
e
s
/
c
o
m
m
u
n
i
t
y
-
s
o
c
i
a
l
/
s
t
a
t
e
-
o
f
-
s
o
c
i
a
l
-
i
n
-
s
o
c
i
a
l
-
g
a
m
e
s
/

[
5
]

J

r
v
i
n
e
n
,

A
.

T
h
e

m
o
r
e

s
o
c
i
a
l

s
c
h
o
o
l

o
f

s
o
c
i
a
l

g
a
m
e
s

d
e
v
e
l
o
p
m
e
n
t
.

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m
:

h
t
t
p
:
/
/
g
a
m
e
s
4
n
e
t
w
o
r
k
s
.
p
o
s
t
e
r
o
u
s
.
c
o
m
/
t
h
e
-
m
o
r
e
-
s
o
c
i
a
l
-
s
c
h
o
o
l
-
o
f
-
s
o
c
i
a
l
-
g
a
m
e
s
-
d
e
v
e
l
o

[
6
]

J

r
v
i
n
e
n
,

A
.

3

B
e
n
e
f
i
t
s

o
f

s
y
n
c
h
r
o
n
o
u
s

g
a
m
e
p
l
a
y

f
o
r

s
o
c
i
a
l

g
a
m
e
s
.

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m
:

h
t
t
p
:
/
/
g
a
m
e
s
4
n
e
t
w
o
r
k
s
.
p
o
s
t
e
r
o
u
s
.
c
o
m
/
3
-
b
e
n
e
f
i
t
s
-
o
f
-
s
y
n
c
h
r
o
n
o
u
s
-
g
a
m
e
p
l
a
y
-
f
o
r
-
s
o
c
i
a
l

[
7
]

K
e
o
g
h
,

B
.

N
i
n
t
e
n
d
o

s

I
w
a
t
a
:

S
o
c
i
a
l

N
e
t
w
o
r
k

G
a
m
e
s

a
r
e

N
o
t

S
o
c
i
a
l

G
a
m
e
s
.

R
e
t
r
i
e
v
e
d

1
7
t
h

J
a
n

2
0
1
3

f
r
o
m

i
n
d
u
s
t
r
y
g
a
m
e
s
.
c
o
m
:

h
t
t
p
:
/
/
w
w
w
.
i
n
d
u
s
t
r
y
g
a
m
e
r
s
.
c
o
m
/
n
e
w
s
/
n
i
n
t
e
n
d
o
s
-
i
w
a
t
a
-
s
o
c
i
a
l
-
n
e
t
w
o
r
k
-
g
a
m
e
s
-
a
r
e
-
n
o
t
-
s
o
c
i
a
l
-
g
a
m
e
s
/

[
8
]

L
o
v
e
l
l
,

N
.

W
h
a
t

i
s

a

S
o
c
i
a
l

G
a
m
e
?

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m

G
a
m
e
s

B
r
i
e
f
:

h
t
t
p
:
/
/
w
w
w
.
g
a
m
e
s
b
r
i
e
f
.
c
o
m
/
2
0
1
1
/
0
1
/
w
h
a
t
-
i
s
-
a
-
s
o
c
i
a
l
-
g
a
m
e
/

[
9
]

N
u
t
t
,

C
.

F
a
c
e
b
o
o
k
:

S
o
c
i
a
l

g
a
m
e

d
e
v
s

m
u
s
t

m
a
k
e

t
h
e
i
r

g
a
m
e
s

a
c
t
u
a
l
l
y

s
o
c
i
a
l
.

R
e
t
r
i
e
v
e
d

1
7
t
h

J
a
n

2
0
1
3

f
r
o
m

G
a
m
a
S
u
t
r
a
:

h
t
t
p
:
/
/
g
a
m
a
s
u
t
r
a
.
c
o
m
/
v
i
e
w
/
n
e
w
s
/
1
8
0
3
1
0
/
F
a
c
e
b
o
o
k
_
S
o
c
i
a
l
_
g
a
m
e
_
d
e
v
s
_
m
u
s
t
_
m
a
k
e
_
t
h
e
i
r
_
g
a
m
e
s
_
a
c
t
u
a
l
l
y
_
s
o
c
i
a
l
.
p
h
p

[
1
0
]

P
a
a
v
i
l
a
i
n
e
n
,

J
.
,

A
l
h
a
,

K
.

a
n
d

K
o
r
h
o
n
e
n
,

H
.

E
x
p
l
o
r
i
n
g

P
l
a
y
a
b
i
l
i
t
y

o
f

S
o
c
i
a
l

N
e
t
w
o
r
k

G
a
m
e
s
.

I
n

A
d
v
a
n
c
e
s

i
n

C
o
m
p
u
t
e
r

E
n
t
e
r
t
a
i
n
m
e
n
t
,

v
o
l
u
m
e

7
6
2
4

o
f

L
e
c
t
u
r
e

N
o
t
e
s

i
n

C
o
m
p
u
t
e
r

S
c
i
e
n
c
e

(
2
0
1
2
)
,

S
p
r
i
n
g
e
r
,

p
.

3
3
6
-
3
5
1
.

[
1
1
]

P
a
a
v
i
l
a
i
n
e
n
,

J
.
,

H
a
m
a
r
i
,

J
.
,

S
t
e
n
r
o
s
,

J
.

a
n
d

K
i
n
n
u
n
e
n
,

J
.

S
o
c
i
a
l

G
a
m
e
s

o
n

F
a
c
e
b
o
o
k
:

P
l
a
y
e
r
s

P
e
r
s
p
e
c
t
i
v
e
.

A
c
c
e
p
t
e
d

f
o
r

S
i
m
u
l
a
t
i
o
n

&

G
a
m
i
n
g

(
f
o
r
t
h
c
o
m
i
n
g
)
.

[
1
2
]

R
a
d
d
,

D
.

I
d
l
e

G
a
m
e
s

s
a
y
s

y
o
u

c
a
n

t

h
a
v
e

a

s
o
c
i
a
l

g
a
m
e

w
i
t
h
o
u
t

r
e
a
l
-
t
i
m
e

p
l
a
y
.

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m

g
a
m
e
i
n
d
u
s
t
r
y
.
b
i
z
:

h
t
t
p
:
/
/
w
w
w
.
g
a
m
e
s
i
n
d
u
s
t
r
y
.
b
i
z
/
a
r
t
i
c
l
e
s
/
2
0
1
2
-
0
4
-
0
7
-
i
d
l
e
-
m
i
n
d
s
-
s
a
y
s
-
y
o
u
-
c
a
n
t
-
h
a
v
e
-
a
-
s
o
c
i
a
l
-
g
a
m
e
-
w
i
t
h
o
u
t
-
r
e
a
l
-
t
i
m
e
-
p
l
a
y

[
1
3
]

R
o
s
e
,

M
.

B
r
a
i
d

s

B
l
o
w
:

S
o
c
i
a
l

G
a
m
e
s

a
r
e

N
o
t

V
e
r
y

S
o
c
i
a
l

.

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m

G
a
m
a
S
u
t
r
a
:

h
t
t
p
:
/
/
w
w
w
.
g
a
m
a
s
u
t
r
a
.
c
o
m
/
v
i
e
w
/
n
e
w
s
/
1
2
3
4
9
0
/
B
r
a
i
d
s
_
B
l
o
w
_
S
o
c
i
a
l
_
G
a
m
e
s
_
A
r
e
_
N
o
t
_
V
e
r
y
_
S
o
c
i
a
l
.
p
h
p
#
.
U
O
h
y
6
n
f
I
V
o
M

[
1
4
]

S
t
e
n
r
o
s
,

J
.
,

P
a
a
v
i
l
a
i
n
e
n
,

J
.

&

M

y
r

,

F
.

S
o
c
i
a
l

I
n
t
e
r
a
c
t
i
o
n

i
n

G
a
m
e
s
.

I
n
t
e
r
n
a
t
i
o
n
a
l

J
o
u
r
n
a
l

o
f

A
r
t
s

a
n
d

T
e
c
h
n
o
l
o
g
y

4
,

3
,

(
2
0
1
1
)
,

3
4
2
-
3
5
8
.

[
1
5
]

T
y
n
i
,

H
.
,

S
o
t
a
m
a
a
,

O
.
,

&

T
o
i
v
o
n
e
n
,

S
.

H
o
w
d
y

p
a
r
d
n
e
r
!
:

o
n

f
r
e
e
-
t
o
-
p
l
a
y
,

s
o
c
i
a
b
i
l
i
t
y

a
n
d

r
h
y
t
h
m

d
e
s
i
g
n

i
n

F
r
o
n
t
i
e
r
V
i
l
l
e
.

I
n

P
r
o
c
.

A
c
a
d
e
m
i
c

M
i
n
d
T
r
e
k
,

(
2
0
1
1
)
,

A
C
M

P
r
e
s
s
,

2
2
-
2
9
.


[
1
6
]

V
e
n
t
r
i
c
e
,

T
.

B
u
i
l
d
i
n
g

t
h
e

F
o
u
n
d
a
t
i
o
n

o
f

a

S
o
c
i
a
l

F
u
t
u
r
e
.

R
e
t
r
i
e
v
e
d

J
a
n
u
a
r
y

1
0
,

2
0
1
3
,

f
r
o
m

G
a
m
a
s
u
t
r
a
:

h
t
t
p
:
/
/
w
w
w
.
g
a
m
a
s
u
t
r
a
.
c
o
m
/
v
i
e
w
/
f
e
a
t
u
r
e
/
4
2
1
0
/
b
u
i
l
d
i
n
g
_
t
h
e
_
f
o
u
n
d
a
t
i
o
n
_
o
f
_
a
_
.
p
h
p

[
1
7
]

C
h
e
n
,

M
.

C
r
o
p
-
s
t
e
a
l
i
n
g

o
n

H
a
p
p
y

F
a
r
m
:

a
n

a
d
d
i
c
t
i
o
n

t
o

a
f
f
i
l
i
a
t
i
o
n
.

R
e
t
r
i
e
v
e
d

J
a
n

1
7
t
h

2
0
1
3

f
r
o
m

C
h
i
n
a
.
o
r
g
.
c
n
:

h
t
t
p
:
/
/
w
w
w
.
c
h
i
n
a
.
o
r
g
.
c
n
/
c
h
i
n
a
/
2
0
0
9
-
1
2
/
1
0
/
c
o
n
t
e
n
t
_
1
9
0
4
4
4
7
8
.
h
t
m

DESVIG 2013
61
World of Warcraft Machinima:
Para-Game Communities as Cultures of Participation
Tyler Pace, Shad Gross
Indiana University
901 E. 10
th
St, Bloomington, IN 47408 USA
{tympace, shagross} @indiana.edu

ABSTRACT
This paper leverages the framework of cultures of participa-
tion to construct a history of the World of Warcraft (WoW)
machinima community. In constructing this history, we
identify a number of mechanisms by which WoW machin-
ima was leveraged to support collaboration among and
within the larger WoW game community. WoW machinima
presents a case study of a culture of participation that serves
as a backbone for a larger, creative, and collaborative para-
game community that continues to shape the broader game
community. In this way, the theories and frameworks of
cultures of participation may provide game designers with
inspiration and tools for designing for collaboration and
sociability within and surrounding their games.
Author Keywords
Cultures of Participation, Machinima, World of Warcraft
ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g., HCI)
INTRODUCTION
Internet-based special interest communities have demon-
strated enormous creativity and productivity in recent years,
from Wikipedia and open source communities to art, craft,
and multimedia communities, such as Second Life and De-
viant Art. Increasingly, such communities are the subjects
of HCI research, as scholars seek to understand the struc-
tures and processes that support these often bottom-up, non-
traditional collaborations [8]. Such research can influence
the design of technologies intended to support social col-
laborations in online communities [1].
The present work focuses on the World of Warcraft (WoW)
machinima community. World of Warcraft is a massively
multiplayer online game, with an estimated 10 million ac-
tive subscribers. Machinima refers to digital videos that are
primarily produced (i.e., set, acted out, and recorded) in
video games. Most of these videos are 3-5 minutes in
length, and in addition to showing actual or staged game-
play may also feature any combination of in-game sounds,
voice acting, music, and recordings taken from popular
culture (e.g., pop songs and movies). As noted by Lowood
and Nitsche [9], machinima has emerged from an amateur-
driven community on the way to a history, an aesthetic,
and forms of reception and marketability that it can call its
own. The WoW machinima community includes the ma-
chinimators themselves (i.e., machinima producers), their
audiences, and those who produce the software tools that
support and expand the capabilities of machinimators.
Further, studying the history of an online creative commu-
nity provides a unique set of methodological opportunities
and challenges. The study highlighted in this paper employs
a diachronic historical analysis of the development of the
WoW machinima community leveraging the data, re-
sources, and remnants created by the community from its
earliest days to the present. Chiefly, these resources exist in
the form of forum posts, developer blogs, web sites, instruc-
tional guides, and patch notes. In relying on data generated
by the community itself during its development, our study
works to relate the history of the community as it was artic-
ulated and experienced in-the-moment by its members. We
conclude this paper with a discussion of how the connection
between machinima, its tools, and its diverse community
work together to form a culture of participation that sup-
ports a para-game [2] community.
Para-Game Communities and Cultures of Participation
Carter et al. describe para-game activities as those where
the meaning and purpose of these play activities is con-
structed through player motivations found beyond what can
be experienced through the [normal game] or is relevant to
the content of play [2]. The communities that develop
around para-game activities (i.e., para-game communities)
are valuable as they provide a place for extended forms of
interaction and socialization with respect to the games from
which they are derived. In terms of WoW, its machinima
community can be observed as a para-game function.
HCI has not yet dealt with the para-game in those specific
terms, but there are a number of theoretical analogues that
may be used to help game designers think about para-game
communities. Notably, the well-documented theories of
communities of practice [11] and interest [5] offer a strong
starting point. These models work to explain community
collaborations exemplified in networks of practitioners and
professionals who coalesce around a specific problem and,
ultimately, solve the problem and disband the community.






Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.


DESVIG 2013
62
The key contributions of these communities are most often
the knowledge and knowledge-management systems pro-
duced in the pursuit of the communitys problem solving.
However, the problem-focused orientation of these theories
may limit their applicability to game design.
Recently, research on community collaborations in HCI has
expanded its interest beyond that of professional problem-
solving communities to networks of amateur producers en-
gaged in everyday practices [1, 3,4]. In this vein, the locus
of collaboration extends to cultures of participation [3,4]
which, instead of emphasizing discrete professional prob-
lems, embrace unfinished design opportunities for which
there is not an intent to find a single solution, but rather to
engage in informal, practice-based, and experience-oriented
activities with others [3]. A key characteristic of cultures of
participation is that all members have an opportunity for
active engagement, to make contributions, and share those
contributions with others [3]. In this way, cultures of partic-
ipation, and the theoretical frameworks surrounding them,
may provide game designers with inspiration on how to
support thriving para-game communities.
Notably, the WoW machinima community embodies a cul-
ture of participation. The WoW machinima community is
primarily comprised of amateur multimedia producers with
varying technical competencies, prior training, and socio-
cultural contexts. Further, instead of focusing on a unifying
problem to be solved, the WoW machinima community
operates from a shared interest in a specific game (i.e.,
WoW). Critically, the product of the WoW machinima
community is not just its tools or videos, but includes the
community itself which supports a desirable and collabora-
tive para-game space that enriches WoW for all players.
METHODOLOGY
We employed a diachronic, historical analysis to inform our
understanding of the culture of participation in the WoW
machinima community. A diachronic analysis aims to cap-
ture the way that events internal and external to the com-
munity combine forces to build upon one another in shap-
ing the community. To pursue this analysis, we gathered
information reflecting the thoughts of the community mem-
bers at the time those thoughts occurred. As those beliefs
are subject to change, it is critical that a diachronic analysis
acquire evidence that spans the timeline of the WoW ma-
chinima community. This approach is partially influenced
by Sus discourse analysis constructing the Street Fighter
IV player, wherein textual data was captured as a means to
help access the meanings embedded in the Street Fighter
community [10]. Specifically, online documents, the texts
of this study, were collected as a means of understanding
how the WoW machinima community developed.
While working on a related project, in which we are analyz-
ing a large corpus of WoW machinima, we noticed a trend
of machinimators listing two tools in their credits and video
descriptions. This trend inspired us to examine those tools,
as they appeared to play an important role both within the
community and in the production of WoW machinima. The
two tools that we encountered are the WoW Model Viewer
and the WoW Machinima Tool. These tools served as the
starting points of our data collection.
We applied a systematic process of gathering data, first
casting a large net with sequential searches on Google,
bracketed by year from the release of WoW to the present
(2004-2013), of the names of the two tools along with relat-
ed words/phrases such as change log, release, and
download. These searches returned approximately 7,200
hits, which were reviewed for relevance. The results were
categorically organized into sources that were focused on
change logs, tutorials, individual machinimators, Blizzard,
and the general WoW community.
The change logs and machinima tutorials were used to de-
termine when tool releases and features occurred within the
period of observation. For the remaining categories, we
identified frequently mentioned sites, blogs, and forums,
such as the tools websites, machinima hosting sites, and
prominent machinimators blogs. These sites were subse-
quently used as sources to build a more focused timeline.
Further, the Internet Archive WayBack Machine
(http://archive.org/web/web.php) was used to view archived
versions of the sites thereby allowing access to content that
was removed over time.
Significant events identified through our searches were
chronologically organized, and further searches were em-
ployed to fill in any gaps with respect to specific events in
the timeline. This synthesis of events serves as the core of
the data used to reconstruct meaning in the history of WoW
machinima tools and their relationship to the community
and medium of WoW machinima.
FINDINGS
In this section, we cover preliminary findings and present
them within the framework of three aspects of cultures of
participation: ownership, control, and cohesion [3].
Ownership
The ability to cultivate a sense of ownership over an aspect
of a communitys work and creations is key to the devel-
opment of a culture of participation [3]. Fischer describes
ownership as working best when participants are directly
engaged and receive support for intrinsic and extrinsic
motivation [3]. If the community members have a sense
that what they are doing is tied to their desires as well as
representative of them, then the outcomes of the community
will be optimized. A common strategy for maximizing
ownership is to break projects into smaller parts. In the case
of WoW machinima, the development of the medium is
directed by the different ways that members of the commu-
nity take ownership of the content of the game, and the
means of doing so through tools.
DESVIG 2013
63
The defining component of WoW machinima, World of
Warcraft, was released in November 2004. However, the
earliest machinima date back to September of that year,
showing off the beta version of the game [i]
1
. Shortly after
the games release, warcraftmovies.com debuted, dividing
movies into categories of Player vs. Player (PVP), differ-
ent players fighting each other, Player vs. Environment
(PVE), players fighting monsters within the game, and
Other a catch-all for anything that did not fall into the
other categories [ii]. This categorization, with its primary
focus on footage of events within the game, approaches one
of the first expressions of ownership within the WoW ma-
chinima community exploration.
The beginnings of WoW presented an entirely new game
for players to explore. Oxhorn, a prominent machinimator,
began by making A PVP Laugh, a video about PVP [iii].
This video focused on showing the game as it was played.
However, there was also a different approach to exploring
the game looking into the actual 3-dimensional compo-
nents of the game the models. This was done using an
appropriately named tool: WoW Model Viewer.
Model Viewer is a tool that allows access to WoWs data
files, which contain all of 3-dimensional models that com-
prise the avatars, non-player characters (NPCs), and objects
in the game. This tool was originally created by program-
mers named Ufo_Z and Darjk [iv]. The initial description of
Model Viewer from wowmodelviewer.org provides a de-
scription of initial, exploration-based uses: You can dress
up and customise the character models or just view NPC
models [v]. One of the first references to Model Viewer,
far before the release of wowmodelviewer.com, came from
wowwiki.com in J uly of 2005, wherein Model Viewer is
listed as a User Interface tool. A frequent example of this
type of use was to look through the games data files as a
means of finding out what different models existed [vi].
Beyond the exploratory uses of Model Viewer, there are
early examples of ownership in the form of personalization.
Since Model Viewer provided access to all of the character
models and armor, clothing, and weapons from the game,
players could recreate their character or any other in-game
character that they wanted. One application of this person-
alization technique was to make images using the models
from the game [vii, viii, ix]. These images allow the crea-
tors to show off their avatars and their ability to manipulate
the contents of the game. This also meant that players could
explore options regarding outfits and combinations of items
to see what looked best as a new form of recreation involv-
ing the game [x, xi]. This use extended beyond just demon-
strating ownership of different aspects of the game. The

1
Our research is documented with many non-academic
resources. For reasons of space, we list these online at
[http://tinyurl.com/wowchi2013machinimarefs]. Non-
academic references are indicated with Roman numerals.
tool created the possibility of greater control, above and
beyond the confines of what the game normally offered.
Early WoW machinima, and the initial purpose of one of its
tools, was focused on the exploration of WoW as a game
and as a library of 3D models. Further, the creation of these
early artifacts, such as PVP movies or avatar images, pro-
vided the initial foundation for a sense of ownership over
the content of the game. This sense of ownership would
lead to a need for greater control by players, and ultimately
lay the foundation for the medium.
Control
Strong cultures of participation support users as active
contributors who can transcend the functionality and con-
tent of existing [sociotechnical] systems [3]. In other
words, the best sociotechnical systems allow for users to
exceed the original capabilities and conceptions of the sys-
tem. This extension of the original system (i.e., WoW) is at
the heart of WoW machinima, even in its earliest form, and
the machinima tools ability to grant control to the machin-
imators is an important aspect of the medium.
Personalization of the characters in WoW through the
WoW Model Viewer exhibited signs of taking ownership of
a character and other game assets, but also exhibited signs
of taking control of that character. Additionally, Model
Viewer was used to alter models in order to change the ex-
perience of gameplay: players could use Model Viewer to
find a model they liked and then go into the client-side code
and swap and existing model for a new custom model [xii].
The result of this process would be that the characters in the
game would appear differently on the players computer
than they would in the rest of the world. While this could be
done just as a means of changing the way the game looks to
suit a players taste, it also presented problems in terms of
gameplay. Swapping the skins (the graphics that wrap
around a three dimensional model) of models will only re-
sult in a different appearance for that player; however,
swapping models results in a change to the size of the play-
er, making them more difficult to hit or capable of moving
into places that would otherwise be impossible to enter
[xiii]. Due to the gameplay issues that this use of Model
Viewer caused, Blizzard, the company that created WoW,
set up obstacles in the game code to prevent players from
changing models, and banned players who continued to do
so after the obstacles had been set in place [xiii]. For the
purposes of changing the game while playing on a live
server, the WoW Model Viewer was considered an exploi-
tative tool by the games developers.
Model Viewer was designed to offer little capacity to func-
tion within the game and while this issue was not specific to
machinima, it did have an effect on the ease of use of Mod-
el Viewer for machinimators. As Mike Spiff Booth, an-
other notable WoW machinimator, notes in his guide to
making machinima: capturing video in game is quicker,
easier but youre stuck with what you can make happen
DESVIG 2013
64
in the game itself while using Model Viewer afforded lots
of control over the output but is more time-consuming
than doing it the easy(er) way [xiv]. As a result, the choice
to restrict Model Viewers use in-game was confronted by
the ease that came from operating in-game tools. This left a
gap in control with respect to WoW machinima production.
This gap in machinima control focused on the nature of
how the game appeared. Since changing the models was not
condoned by Blizzard, machinimators looked to another
part of the game that they could better control: the camera.
WoWs in-game camera normally follows set rules that
determine how it moves, giving limited control of angle,
distance, direction, etc. The solution to this problem came
in the form of the second WoW machinima production tool
on which this paper will focus: the WoW Machinima Tool.
Machinima Tool was created by a machinimator named
malu05, whose Exploration After 1.9 video dates back to
J anuary 2006 [xv]. Whereas Model Viewer was built in a
way that emphasized exploration of the game, Machinima
Tool was constructed by a seasoned machinimator, with the
context of WoW machinima as a guiding principal. Ma-
chinima Tool saw an initial demo release in J uly 2008 that
was advertised by way of a warcraftmovies.com video. Ini-
tially Machinima Tool was advertised as a tool that allowed
the player to take full control of the in-game camera in ad-
dition to taking control of a few other aspects of the visual
appearance of scenery during gameplay [xvi]. This ground-
breaking level of camera and visual control allowed for an
increase in what could be done in WoW while still allowing
for the afforded ease of recording in-game.
In terms of control, the capture of game footage can be seen
as the point of inception for taking control of the game in
ways that extend the original purpose of the game. Person-
alization of game content via Model Viewer yielded some
control, but, because of the potential for gameplay exploita-
tion, this form of control was met with disapproval and ban-
ishment. However, extraction of game assets opened new
potential for control in terms of analyzing game content and
more effectively documenting what is in the game. Alter-
nately, Machinima Tool offered the possibility of taking
control of specific mechanisms within the game, allowing
for manipulations specifically focused on the processes of
making machinima. The creation of the tools, along with
the nature of their initial contributions to the medium of
WoW machinima, grew, in part, from this desire for greater
control. However, in WoW machinima the nature of that
control, itself, is not static but influenced and developed in
conjunction with other aspects of the community.
Cohesion
Importantly, WoW machinima is not the product of an or-
ganization that was formed by a single need or desire to
create videos. It grew from the creative power and interests
of a heterogeneous group. This type of creative collabora-
tion is an important component for cultures of participation
[3]. Specifically, the cohesion between the members and
groups that make up the greater WoW community help
spurred the growth of the machinima community.
An important aspect of WoW machinima is its community
nature. This community nature is partially accounted for by
Fischers notion of creative collaboration. At the macro-
level, the different groups that make up the WoW machini-
ma community were divided into Blizzard, as the creator of
the game, the players of the game as part of the general
WoW community, the WoW machinima makers, and the
makers of Machinima tools. These groups have different
perspectives on WoW and the community as a whole. As an
example, Model Viewer enabled exploration of the game
for WoW fans who did not make machinima, a means of
creating new scenes for WoW machinima makers, and the
potential to create unfair advantages to some players, and
thus an issue regarding gameplay to Blizzard. In this exam-
ple, lack of cohesion between Blizzard and players who
used Model Viewer to change aspects of the game resulted
in players getting banned. However, the cohesion between
players interested in exploring the game and the machinima
makers who wanted to take control of the game resulted in
the adaptation of a tool. As Fischer [3] argues, diversity,
independence, decentralization, and aggregation increase
creative collaboration. The different groups described
above fulfill the first three qualities, but their ability to ag-
gregate, or the cohesion of the group, changes over time.
One of the first examples of cohesion between WoW ma-
chinima makers and Model Viewer came in the form of
recognition of the tools potential on wowmodelview-
er.com. A month after the launch of the page, with its de-
scription of the tool in terms of looking at models without
mention of machinima, a lengthier About WoW Model
Viewer section had been added with a note about the addi-
tion of customizable lighting to allow the users, imparticu-
larly machinima makers to improve the lighting conditions
to help blend the models with background scenes [xvii].
This not only showed recognition of WoW machinimators,
but also pointed to a specific change to the features of the
tool in order to be accommodating to them. This recogni-
tion of WoW machinimators would continue over time,
including a link to Machinima 101 a site that has all
the basics to get you started, including an explanation of
what machinima really is and how it came about [xviii].
While the connections between the WoW machinima com-
munity and Model Viewer grew over time in the beginning,
there was a major setback in terms of that cohesion. The
tool was programmed by Darjk, who also was responsible
for maintaining the website [xix]. During the tools devel-
opment, he had left and returned to working on it, having
intentions outside of that program [xx]. More disruptive
was a motorcycle accident in December 2007, which led
him to: a new perspective on whats important and, if suf-
ficient progress cant be made due to time constraints most
likely be discontinuing the project as well as giving up pro-
gramming as a hobby [xxi]. This led to a lengthy time
DESVIG 2013
65
away from the Model Viewer, with few updates focused on
bug fixes, particularly getting the tool to work with new
patches [xxii]. At a later point, a technical issue was met
with the response of The original author, J ohn Steele (aka
Darjk), has currently left the project and wont be back till
2010 [xxiii]. To add to this dilemma, in 2009, the size of
the community began to overwhelm Model Viewers web-
site, forcing a relocation that resulted in cross-platform is-
sues [xxiv]. While the Model Viewer still existed during
this period, the growth of cohesion between machinima
makers and Model Viewer had been reduced as the latter
continued to develop while the former slowed to a stop.
This process was not permanent, however, as another de-
veloper named chuanhsing ported the project over to
Google Code, where multiple members of the community
could collaboratively work on it [xxv].
Developed by a WoW machinimator, Machinima Tool be-
gan with a high level of cohesion with the WoW machinima
community. One example is the Machinima Tool webpage
(pxr.dk), which has the normal documentation of develop-
ment, but also has tutorials to help use the tool to create
machinima [xxvi]. In spite of this initial cohesion, Machin-
ima Tool was not without its own unique issues which
would temporarily distance it from the larger community
due to issues with the perception of Blizzards view of it.
Blizzards relationship with tools had shown considerable
fluctuation. While the model-swapping uses of Model
Viewer had gotten some players banned, Blizzard also
showed support for machinima with a letter regarding its
status, regarding terms of use. In this letter, released to the
public in August, 2007, Blizzard specifically made a state-
ment of support for outlined what was considered appropri-
ate use excluding videos only if they generated revenue
for some party other than themselves or contained content
that was pornographic in nature [6]. Then, with the release
of WotLK, Blizzard also began to shut down private serv-
ers; WoW Insider reported on the issue on December 5,
2008 stating that Blizzard is starting to shut down servers
and advising those who would fight this, Dont tempt fate
[xxvii]. In opposition to this warning one server, Scapegam-
ing, became the target of a Blizzard lawsuit based on cop-
yright infringement in October of 2009 [xxviii]. This law-
suit ended in Blizzard winning $88m in damages from Ali-
son Rees of Scapegaming [xxix]. While this did not relate
directly to WoW machinima, it did vex a number of players
and show that they were willing to take action with regard
to situations that they found unsuitable [xxx].
With this as a contextual background, concerns also raised
regarding Machinima Tool use. One player posted that they
were concerned with getting the ban hammer from this
software, or being disqualified from contests for using it
[xxxi]. Many users posted to the forums, but found few
responses from official sources and the answers from the
web to be murky at best [xxxii]. Although paranoia had
risen regarding Blizzards stance on copyright, WoW Insid-
er addressed this with a post that plainly restated Blizzards
stance on the rights of machinimators [xxxiii].
In WoW machinima, the community can have a profound
effect on output. The group is diverse, comprised of indi-
viduals from different geographical locations and perspec-
tives. The community is also completely decentralized, with
neither a clearly defined system of leadership nor a singular
driving purpose that applies to the entire group equally.
However, this is advantageous: by operating independent of
leadership, fluctuations in the groups aggregative capabili-
ties, due to its cohesion, are less likely to destroy the com-
munity due to the elimination of leaders.
DISCUSSION
Leveraging the framework of cultures of participation al-
lowed us to construct a history of the WoW machinima
community. In constructing this history, we identified a
number of mechanisms by which WoW and its machinima
were leveraged to support the surrounding community.
WoW machinima presents a case study of a culture of par-
ticipation wherein development of machinima, its commu-
nity members, and its support tools are all connected. These
connections were discussed with respect to the concepts of
ownership, control, and cohesion. However, these concepts
are all related through another: adaptation [7].
Early machinima was focused on the game of WoW, with
the earliest machinima focusing on the documentation, ex-
ploration, and personalization of the game. The content of
the game was appropriated as a means to create a link be-
tween the game and the members of the machinima com-
munity. This link was supported by the adoption of the
Model Viewer as it allowed players a means to access the
many models that comprised the game, and export them for
use in machinima. Later, the Machinima Tool would im-
prove the extent to which WoW was appropriated by intro-
ducing a new in-game camera system for machinimators.
Combined, Model Viewer and Machinima Tool improved
the affordance for adaptation, or adaptogenicity [7], of
WoW within the machinima community. This new level of
adaptation resulted in increasing personalization. For ex-
ample, Model Viewer allowed players to create events that
could not be created in-game, thereby appropriating the
whole of the contents of the game to map to the players
interests. Further, both Model Viewer and Machinima Tool
allowed for the creation of videos that would be prohibi-
tively difficult in-game, but mapped more to the idea of
what the player wanted from the game, such as a cinematic
battle between two bosses [liv]. Early use of Model Viewer
and Machinima Tool provided a starting point: as players
appropriated and personalized aspects of the game, they
began to have a greater understanding of the built-in con-
tents of the game and the limitations of what was available
and possible without modification. This tool-enabled explo-
ration of the game supported a common understanding
amongst machinimators of what was available within the
DESVIG 2013
66
game and the possibilities of how it could be used. Person-
alization through adaptation extended these possibilities,
and as machinimators developed a sense of ownership, they
further pushed the boundaries of their work.
The adaptation of 3D models and in-game camera controls
demonstrate how the design of game assets and logics are a
fundamental to the development of the WoW machinima
community. Model Viewer stemmed from a need to access
and use the assets of the game such as models, clothes, and
backgrounds. The models could then be composited onto
entirely new backgrounds, creating juxtaposition between
the games content and external elements. The elements
from the game could become aspects of the style of the vid-
eo, as in Mike Spiff Booths creations [xxxvii]. Machin-
ima Tool introduced elements taken from cinema, such as
depth of field, and brought those into the milieu. WoW ma-
chinima quickly became populated with examples of these
applications use and popularity [lv, lvi, lvii].
Adapting WoW for machinima was enabled by tools, but
stemmed from the diverse creative collaborations of its
supporting community. As noted by Fischer [3, 4], cultures
of participation, like the WoW machinima community,
support diverse perspectives that work to increase collabo-
ration and socialization within a community. Within the
WoW community, the diversity of adaptations becomes a
metric by which innovation is understood and measured.
Further, the WoW machinima community celebrates and
promotes the introduction of tool-mediated innovations, as
evidenced by Darjks presentation and commentary on ex-
emplary movies and frequent machinima contests such as
those held by Blizzard and XFire.com.
Subsequently, the ability of WoW-as-a-game and as soft-
ware to be so adaptable may have helped to foster the cul-
tures of participation that support WoWs many para-game
communities (of which machinima is just one). Further, the
history of WoW machinima demonstrates how para-game
communities may change in terms of their capabilities and
coherence, but remain engaged with WoW in no small part
because of its adaptability and its ability to provide for pa-
ra-game activities that provide places for extended forms of
participation and socialization with respect to the game.
CONCLUSION
This paper leverages the notion of cultures of participation
to construct a history of the World of Warcraft (WoW) ma-
chinima community. In constructing this history, we identi-
fied how issues of ownership, control, and cohesion helped
the machinima community to grow into a prominent part of
the WoW para-game. Further, we preliminarily identified
adaptation as a key theme that may be leveraged by design-
ers interesting in supporting future cultures of participation.
Thus, we argue that the theories and frameworks of cultures
of participation may provide game designers with inspira-
tion and tools for designing for collaboration and sociability
for their game and supporting para-game communities.
ACKNOWLEDGMENTS
This research was funded in part by NSF IIS Creative IT
(#1002772) and the Intel Science and Technology Center
for Social Computing programs.
REFERENCES
1. Bardzell, J . Creativity in Amateur Multimedia: Popular
Culture, Critical Theory, and HCI. Human Technology,
3 (2007), 12-33.
2. Carter, M., Gibbs, M., & Harrop, M. Metagames, para-
games and orthogames: a new vocabulary. Proc. of
FDG 2012. ACM Press (2012), 11-17.
3. Fischer, G. Beyond Interaction: Meta-Design and Cul-
tures of Participation. Proc. of OzCHI. ACM Press
(2011), 112-121.
4. Fischer, G. Exploiting the Powers of Cultures of Partici-
pation. Proc. of SKG2011. Beijing, China, 1-8.
5. Fischer, G. and Herrmann, T. Socio-Technical Systems:
A Meta-Design Perspective. Intl. J . for Sociotechnology
and Knowledge Development, 3 (2011), 1-33.
6. Hayes, C. Changing the Rules of the Games: How Vid-
eo Game Publishers are Embracing User-Generated De-
rivative Works. Harvard J ournal of Law & Technology,
21 (2008), 567-587.
7. Hutcheon, L. A Theory of Adaptation. Routledge, 2006.
8. Kow, Y. M. and Nardi, B. Culture and Creativity: World
of Warcraft Modding in China and the US. In Bain-
bridge, W.S. (Ed.), Online Worlds: Convergence of the
Real and the Virtual. Springer, London, 2010.
9. Lowood, H. and Nitsche, M. Introduction. In Lowood,
H. and Nitsche, M. (Ed.), The Machinima Reader. MIT
Press, Cambridge, 2011.
10. Su, N. Street Fighter IV: Braggadocio Off and On-Line.
Proc. of CSCW 2012. ACM Press (2012), 361-370.
11. Wenger, E. Communities of Practice: Learning, Mean-
ing, and Identity. Cambridge University, UK, 1998.

DESVIG 2013
67
All Work No Social Play: Contextualizing the Missing
Element in Exergames
Tamara Peyton
Pennsylvania State University
University Park, PA 16802
tpeyton@psu.edu


ABSTRACT
This research note discusses the possibility for gaming-
related interventions into the health contexts of obesity
among children and youth. The paper argues that the
problem with existing exergames is that they are too
focused on their health agenda and as such are not fun.
Through a discussion of the impact of designing for flow on
engagement, and the adoption of an idea of flow within a
case study of a commercial exergame, Dance Central 2, the
paper suggests that a synthesized approach that adopts the
best practices from commercial social activity games with
those from the medical and academic research around
health games and healthy physical activity might yield
better, more immersive, more socially aware and more
successful exergames, capable of a greater impact on their
target audience.
Author Keywords
Games; Health; Sociality; Youth; Gamification: Workshop

ACM Classification Keywords
H.5.m.
General Terms
Human Factors
INTRODUCTION
American children and youth lead increasingly screen-
oriented sedentary lives. Studies suggest that children spend
more time at home watching television and playing video
games than they do pursuing school lessons or engaging
with family and friends. This change in activity type and
levels is due to a number of factors. One factor is the
increase of either single parent or two-income earner
parents, which often leaves children as young as 7 at home
alone between daily school end and bedtime, or from dawn
to dusk on weekends. Another contributing but
contradictory factor to increased screen time and sedentary
pursuits among children is the increasing urbanization with
a concommitant sense of risk in society. As more children
and youth grow up in densely populated urban areas,
concerns for child safety increase. When the constant
bombardment of negative media messages of risk, around
crime, traffic dangers, and abuse are considered, the general
civic feeling now is that it is irresponsible to allow ones
child to run loose in the streets alone. Children are to be
kept indoors in order to keep them safe.
That safety comes at a severe health cost however. As will
be discussed in this paper, rates of child and youth obesity
are climbing at alarming rates, and with them, lifestyle
diseases previously associated primarily with adults, such
as cardiovascular issues, are now being found in children as
young as 5.
APPROACH
This paper examines the possibility for gaming-related
interventions into the health contexts of obesity among
children and youth. Working from the position that there is
a lot of work yet to be done to design and deploy positive
exergames for health, the paper argues that the problem
with existing exergames is that they are too focused on their
health agenda and as such are not fun. Through a discussion
of the impact of designing for flow on engagement, and
adoption within exergames, the paper suggests that a
synthesized approach that adopts the best practices from
commercial social activity games with those from the
medical and academic research around health games and
healthy physical activity might yield better, more
immersive and more successful exergames, capable of a
greater impact on their target audience.
CHILD & YOUTH OBESITY
Human life has become increasingly less involved with
physical labour. In the United States, concurrent with the
labour reduction has been an increase in financial income
and the widespread availability of a cornucopia of food.
This relationship between increased availability of food and
the decrease in corporeal labour and physical exertion has
led to what some call an obesity epidemic, and this is seen
most prominently in the child (5-13) and youth populations
(14-24) populations. Thirty-four to forty percent of youth
are overweight or obese [4, 8, 28]. When the sedentary
leisure lifestyles of children and youth are factored in, the
obesity phenomenon is further compounded [4].
Given the ways in which obesity complicates the risk
factors for a variety of lifestyle diseases, such as heart
disease, Type II diabetes and digestion system conditions
[5], the financial, social, and personal costs of health
awareness and health management can be very high. The
financial costs of obesity-related healthcare are often
reported. Thorpe et al. (2004) argue that obesity as a meta

Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.
DESVIG 2013
68
factor is over-represented, accounting for a 27% increase in
health related cost annually in the United States. Other
reports claim that the direct costs of obesity as a percentage
of 2008 national health care expenditures were 5.7% for the
US [1].
However, measured as social and personal effects, some
studies suggest that obesitys indirect costs can be higher
[12, 36]. In one study, researchers measured the amount of
employment discrimination faced by individuals labelled
obese based on their BMI index [2]. They found that
pictures of a female that were altered to make her appear
obese were significantly less likely to get expressions of
interest from university graduate school programs, or from
potential employers. They suggest that this is due to a
favourability disadvantage noted in other research [20, 29,
30]. Research also discloses that the percentages of obesity
are higher for children and youth in lower income or
ethnically disadvantaged group [18].
PHYSICAL ACTIVITY AS PREVENTION
Given the reported direct and indirect costs of obesity, and
considering the strong tendency for overweight children
and youth to become overweight adults [4], prevention has
become a major focus for healthcare awareness in schools.
The US Department of Health has issued guidelines
directing programs to encourage children and youth to
undertake one hour of moderate to vigorous physical
activity daily [26]. This guideline follows the medical
research that suggests a direct causal link between
nutritional intake and energy expenditure [31]. While the
US guidelines are focused towards physical education (PE)
programs in schools [24], due to budget and time
constraints at most American schools, children and youth
are not getting access to or participating in daily PE. When
considering additional socioeconomic factors, such as the
reduction of schools which creates an increase in bussing in
students, or the racial or urban/suburban/rural divide is
accounted for [18], most American children and youth do
not attain this daily suggested level [24].
Exergames for health
How can this decrease in physical activity be alleviated in a
way that breeds positive habits in children and youth? One
suggested method is the fusion of positive health practice
ideas and digital games. As a branch of serious games, or
games with a positive social change agenda, exercise-
oriented games (exergames) are a fast-growing branch of
game development. Exergames are often created as a
partnership between pediatric specialists, educational
technology specialists, human factors researchers and social
scientists. The goal of exergames is to encourage active
behaviour by leveraging children and youths existing
fascination with digital games [9], and by playing to that
fascination through gaming titles that nudge youth off their
couches and into exertion-oriented activity. In one study,
researchers suggest that converting passive screen time,
such as time spent watching television without moving, into
active screen time, considered to be the focused
engagement in active digital gaming, can cause a direct
doubled increase in energy expenditure [19].
Learning from games
In scientific research circles, it is now acknowledged that
games can make for excellent learning environments [11,
14, 34]. Game environments encourage, reinforce and teach
a number of positive habits, from learning to read cultural
cues, to figuring out how to collaborate in unstructured ad-
hoc teams. Games can help youth learn leadership skills,
and can give them a sense of mastery over a cultural world.
There are several commercial titles available that could be
deemed exergames, ranging from the Wii Fit and Dance
Dance Revolution on the Wii console platform, to Dance
Party 2 and Tiger Woods PGA Golf Tour 2, both for the
Xbox Kinect platform. There is a growing body of
academic work looking at the possible connections between
the engagement with these commercial titles and their
potential health effects on children and youth (e.g. [10, 13,
15, 16, 32]. However, much of the academic work to date
outside of these console games have focused on decidedly
non-ludic devices, such as pedometers, yet many still
ascribe the exergame label (e.g.: [6, 23] to such
technologically-enabled activities.
While a labelling issue may seem to be an irrelevant or
pedantic policing of genre boundaries to some, such
labelling can affect the successful adoption of a game by
children and youth. For that population, a game involves an
immersive environment accessible from a computer or
television-attached game console, and it is geared at having
fun. A game must be played after all, and a common failing
of most serious games is that they are not playful and
therefore are rarely fun. Most serious games amount to little
more than interactive encyclopaedias with video and noise
added in. In contrast, a exergame which presents an
immersive environment, and which offers escalating
challenges tailored to expertise, and which rewards
competition and successful completion of tasks would be a
title that would be worthy of the idea of play.
DESIGNING ENGAGING EXERGAMES: OXYMORON?
The orientation towards immersion and fun is what enables
a sense of engagement and what the game studies work
calls flow; a concept adopted from Nakamura &
Csikszentmihalyi [27]. Flow is considered to be the highly
temporal subjective sense of optimized, intensely
pleasurable experience. It is expressed colloquially by
players as the ability to lose ones self in the moment, to
suspend disbelief and run with the experience of being an
elf, a golf pro or a bowling ball. For an exergame to be
successful, with success rated both according to feelings of
fun, as well as adequate physical exertion, Sinclair,
Hingston & Masek [33] suggest that it must consider
engagement factors, and Tam, Distler, and Kreit [25]
suggest that flow is one way to conceptualize this kind of
engagement.
DESVIG 2013
69
Regardless of the label used for the genre of game, or for
youths increasing fascination for digital games, exergames
alone may not be sufficient to encourage children and youth
to participate in physical activity, whether with or without
a gaming component [17]. Children and youth need the
social element; that strong support and encouragement of
family and peer groups in order to stick with physical
activity as a regular daily habit [7]. Eyler & Giles [7]
further suggest that social psychological maturity found in
indicators of a child or youths capacity for engagement,
social maturity, resilience and curiosity are also required for
adequate learning takeaways, particularly among females.
Additionally, many commercial games that can be
mobilized as exergames were originally designed as turn-
taking party games. The social- environmental factors must
be taken into account when designing exergame
applications or activity programs that use exergames
towards encouraging positive physical activity for health.
What might an exergame look like, if it is designed with a
conscious sense of flow, a specific social agenda and set of
criteria that encourage social responsibility, individual
reflexivity and a sense of physical mastery over the
exergames challenges? Tam, Distler & Kreit [35] are
instructional here. The researchers note that this sense of
engaged exergaming might arise from a subjective sense of
successfully adapting ones activity to game-presented
challenges in tasks, designed according to an understanding
of the interests and abilities of the intended players.
Through a discussion of various types of games over the 30
year history of exertion gaming, Tam, Distler & Kreit note
that age factors complicate exergame design. The kind of
tag and release games that are popular and considered fun
by 6-11 year olds will be viewed as engaging and rewarding
by 15-22 year olds. Macvean and Robertson [6] concur with
this, suggesting that the two different games they tested
with 12 to 15 year olds had differing rates of enjoyment and
positive reactions from each of the chronological age
groupings in their participant pool. The design
consideration that emerges from this is the need to
understand the social, physical and psychological
tendencies of different age ranges of player.
Tam, Distler & Kreit [35] argue that there is a temporal
design imperative for exergaming: the need to encourage
specific levels of physical activity that endure over a
specific period of time, adjusted according to desirable
exertion intensity parameters. As one research group
demonstrates [21,22] through a study of the intensity and
health payoffs of dance games, the typical design of
commercial games are not geared at sustained physical
activity over a set and prolonged period of time. Most
players of early dance games like Dance Dance Revolution
would participate in one or two dances before ceding their
spot on the dance mat to another player. In doing so, they
aborted the possible health benefits of playing the game for
a prolonged period. The challenge for commercial
exergames, then, is how to design for the diverging values
of social fun, physical exertion and maximized flow. A
further challenge for serious games designers is how to
adopt the wide age-range appeal commercial social party
games, while also designing into their hybrid exergames the
social values of resilience, competency and commitment
related to physical health, that are required for the healthy
current and future lifestyles of children and youth.
Dance Central 2
The most recent generation of commercial physical activity
social games appears to have become aware of their
shortcoming for health. Now, enabled by the Microsoft
Xbox Kinect body-awareness system, titles such as the
Dance Central 2 game from Microsoft are explicitly tagged
commercially as being part of the exercise genre. Dance
Central 2 includes a specific way to play the game that is
akin to the dance-oriented workouts, such as Zumba or
jazzercise, found at most community gyms. When players
log into the Dance Central 2 game interface, they are asked
to choose an avatar, to choose whether to play solo or in
tandem with another, and also to choose a game approach
mode.
One of the new features of Dance Central 2 is the ability to
select a fitness mode. Choosing this mode allows a player
to select a series of songs to which they wish to do a game-
directed, time-specific, dance-oriented workout, and also to
select a difficulty level. Based on play experimentation
done by this papers author on this title, the medium
difficulty level would be the most advantageous level for
balancing exertion and fun.
Given this new fitness mode, it would appear that
commercial games like Dance Central 2 are aware of the
design challenges and previous critiques around
Exergames, and are now explicitly attempting to both
enable a sense of flow and provide a prolonged possible
activity level, geared at maximizing fitness outcomes from
game engagement. However, they still have a ways to go. I
am curious why this do not yet allow for players to play
against one another over the network, rather than restricting
cooperative or competitive play to co-situated presence?
And why not allow the posting of successes in the game to
social media sites such as Facebook, perhaps along with a
relevant photo or screen capture if posting to a site such as
Tumblr?
CONCLUSION
It is acknowledged that the medical, social and design field
aspects are too large to do justice to in a short paper and
there are a number of aspects of exergaming that were not
covered herein. As one example, this paper did not engage
with the specifics of exergame design as they relate to the
health success measures of exergames. These measures are
most frequently tied directly to bodily responses to energy
expenditure, such as oxygen uptake (VO2), heart rate and
participants subjective evaluation of perceived exertion
(RPE) [3]. This is because there is an emerging sense in the
non-medical technologic design literature that, given the
DESVIG 2013
70
need to create positive lifestyle changes among children and
youth, other factors might be more suitable to holistic
technologically-enabled approaches to wellness via
exergames [35, 25]. There are also debates within the
serious games community on whether or not to OEM
current commercial exercise games like Wii Fit or Dance
Party 2 and wrap an extra layer of data and direction around
the game, pursuant to activity level goals for health.
What is severely missing in health measures, however, are
guidelines and evaluation of the sociability of exergames.
Given the work reviewed here that has explicitly linked
exergaming to sociality, should metrics for game evaluation
go beyond the mere health and take into account the role of
peer networks and social support systems?
To date, the research shows clearly that games designed by
academics and researchers specifically for exercise have
rarely been fun or properly social, and games designed
commercially for explicit play purposes have rarely been
able to encourage healthy long-term behaviour, If
exergames could be designed according to best practices
from both the technology side of exergame research, such
as that of Sinclair, Hingston, and Masek [33] as well as
from the positive psychology and sociology side of
academic work, such as that of Masten and Reed [25], the
work that emerges from the combined field of the health
informatics world might finally succeed at creating suites of
exergames that are healthy, impactful, social and fun.
REFERENCES
1. Barrett, A. et al. 2008. Burden of Obesity: 10-Year
Review of Published Literature on Direct and Indirect
Costs in Nine Countires. Value in Health. 11, 3 (May.
2008), A160.
2. Benson, P.L. et al. 1980. THE SOCIAL COSTS OF
OBESITY: A NON-REACTIVE FIELD STUDY.
Social Behavior and Personality: an international
journal. 8, 1 (1980), 9196.
3. Bonetti, A.J . et al. 2010. Comparison of Acute Exercise
Responses Between Conventional Video Gaming and
Isometric Resistance Exergaming. J ournal of Strength
and Conditioning Research. 24, 7 (Jul. 2010), 1799
1803.
4. Cameron, N. et al. 2006. Childhood Obesity:
Contemporary Issues. CRC Press.
5. Colditz, G.A. 1992. Economic costs of obesity. The
American J ournal of Clinical Nutrition. 55, 2 (Feb.
1992), 503S507S.
6. Consolvo, S. et al. 2006. Design requirements for
technologies that encourage physical activity.
Proceedings of the SIGCHI Conference on Human
Factors in Computing Systems (New York, NY, USA,
2006), 457466.
7. Eyler, J . and Giles, D.E. 1999. Wheres the Learning in
Service-Learning? J ossey-Bass Higher and Adult
Education Series. Jossey-Bass, Inc., 350 Sansome St.,
San Francisco, CA 94104; Tel: 888-378-2537 (Toll
Free); Fax: 800-605-2665 (Toll Free); Web site:
http://www.josseybass.com ($28.95).
8. Flegal KM, C.M. 2010. Prevalence and trends in obesity
among us adults, 1999-2008. J AMA: The J ournal of the
American Medical Association. 303, 3 (J an. 2010), 235
241.
9. Foster, L. and Page, A. 2006. Self-perceptions and
physical activity behavior of obese young people.
Childhood Obesity: Contemporary Issues. N. Cameron
et al., eds. CRC Press. 5164.
10. Garn, A.C. et al. 2012. What are the benefits of a
commercial exergaming platform for college students?
Examining physical activity, enjoyment, and future
intentions. J ournal of physical activity & health. 9, 2
(Feb. 2012), 311318.
11. Gee, J .P. 2003. What video games have to teach us
about learning and literacy. Comput. Entertain. 1, 1
(Oct. 2003), 2020.
12. Gorstein, J . and Grosse, R.N. 1994. The indirect costs of
obesity to society. Pharmacoeconomics. 5, 1 (1994), 58
61.
13. Graves, L. et al. 2008. The contribution of upper limb
and total body movement to adolescents energy
expenditure whilst playing Nintendo Wii. European
J ournal of Applied Physiology. 104, 4 (2008), 617623.
14. Ito, M. et al. 2009. Hanging Out, Messing Around, and
Geeking Out: Kids Living and Learning with New
Media. MIT Press.
15. J in, S.-A.A. 2010. I Feel More Connected to the
Physically Ideal Mini Me than the Mirror-Image Mini
Me : Theoretical Implications of the Malleable Self
for Speculations on the Effects of Avatar Creation on
AvatarSelf Connection in Wii. Cyberpsychology,
Behavior, and Social Networking. 13, 5 (Oct. 2010),
567570.
16. Kaye, L.K. et al. Submitted. I need a Wii: Motivations
and experiences of playing videogames. Leisure Studies.
(Submitted).
17. Klein, M.J . and Simmers, C.S. 2009. Exergaming:
virtual inspiration, real perspiration. Young Consumers:
Insight and Ideas for Responsible Marketers. 10, 1 (Mar.
2009), 3545.
18. Kumanyika, S.K. and Grier, S. 2006. Targeting
Interventions for Ethnic Minority and Low-Income
Populations. The Future of Children. 16, 1 (2006), 187
207.
19. Lanningham-Foster, L. et al. 2006. Energy Expenditure
of Sedentary Screen Time Compared With Active
Screen Time for Children. Pediatrics. 118, 6 (Dec.
2006), e1831e1835.
DESVIG 2013
71
20. Lerner, R.M. and Korn, S.J . 1972. The Development of
Body-Build Stereotypes in Males. Child Development.
43, 3 (Sep. 1972), 908.
21. Lieberman, D. 2006. Game enhanced music manuscript.
Proceedings - GRAPHITE 2006: 4th International
Conference on Computer Graphics and Interactive
Techniques in Australasia and Southeast Asia. (2006),
245498.
22. Lieberman, D.A. and Brown, S.J . 1995. Designing
interactive video games for childrens health education.
Interactive technology and the new paradigm for
healthcare. K. Morgan, R. M. Satava, H. B. Sieburg, R.
Mattheus and J .P. Christensen, eds. IOS Press. 201210.
23. Macvean, A. and Robertson, J . 2012. iFitQuest: a school
based study of a mobile location-aware exergame for
adolescents. Proceedings of the 14th international
conference on Human-computer interaction with mobile
devices and services (New York, NY, USA, 2012), 359
368.
24. Mary, P.D.S. et al. 2006. The Role of Schools in
Obesity Prevention. The Future of Children. 16, 1
(2006), 109142.
25. Masten, A.S. and Reed, M.-G.J . 2002. Resilience in
development. Handbook of positive psychology. C.R.
Snyder, ed. Oxford University Press. 89105.
26. Nader PR, B.R. 2008. MOderate-to-vigorous physical
activity from ages 9 to 15 years. J AMA: The J ournal of
the American Medical Association. 300, 3 (J ul. 2008),
295305.
27. Nakamura, J . and Csikszentmihalyi, M. 2002. The
concept of flow. Handbook of positive psychology. C.R.
Snyder, ed. Oxford University Press. 89105.
28. Ogden CL, C.M. 2010. PRevalence of high body mass
index in us children and adolescents, 2007-2008.
J AMA: The J ournal of the American Medical
Association. 303, 3 (J an. 2010), 242249.
29. Richardson, S.A. et al. 1961. Cultural Uniformity in
Reaction to Physical Disabilities. American Sociological
Review. 26, 2 (Apr. 1961), 241.
30. Robert, J . 1967. A study of social stereotype of body
image in children. J ournal of Personality and Social
Psychology. 7, 1, Pt.1 (1967), 101104.
31. Roberts, C.K. and Barnard, R.J . 2005. Effects of
exercise and diet on chronic disease. J ournal of Applied
Physiology. 98, 1 (J an. 2005), 330.
32. Robinson, J . et al. 2012. The effects of Wii FitTM and
IREXTM exergaming platforms on postural stability,
user-acceptance and flow: a randomised trial. Sheffield
Hallam University 30th May 2012 (Sheffield Hallam
University, May. 2012).
33. Sinclair, J. et al. 2007. Considerations for the design of
exergames. Proceedings of the 5th international
conference on Computer graphics and interactive
techniques in Australia and Southeast Asia (New York,
NY, USA, 2007), 289295.
34. Steinkuehler, C. and Duncan, S. 2008. Scientific Habits
of Mind in Virtual Worlds. Journal of Science Education
and Technology. 17, 6 (2008), 530543.
35. Tam, A. et al. 2013. The Gaming Edge. Innovation with
Information Technologies in Healthcare. L. Berkowitz
and C. McCarthy, eds. Springer London. 259282.
36. Trogdon, J .G. et al. 2008. Indirect costs of obesity: a
review of the current literature. Obesity Reviews. 9, 5
(2008), 489500.

DESVIG 2013
72
Democracy has arri ved!
Promoting a reflexi ve Approach to Power in Virtual Worlds
through Interface Modifications
Patrick Prax
Uppsala University
Box 513, 751 20 Uppsala
Patrick.prax@im.uu.se
+46760427398

ABSTRACT
In digital online games and virtual worlds, like in other
digital media, the structure of the medium, its code, its
procedural rhetoric, influences the emerging interaction and
culture of the space. As these games become a more wide
spread and influential form of mass media questions about
the social impact become more pressing and present an
ethical concern for game designers. However,
institutionalized game designers are not the only group that
can influence the structure of games. A way to modify the
code of a virtual world and thus to co-create its design is by
extending its interface with an add-on that interacts with the
application programming interface of the game. A
deliberate modification of this code to influence decision
making in the game might thus lead to a more reflexive
approach to power in the game and even help players to
understand and question real-world power structures.
Together with my partner I have elsewhere presented the
design vision and theoretical framework of a digital tool for
ethical decision making that will be implemented in the
virtual world World of Warcraft [15]. Its purpose will be
to supply players with means to modify the power structure
built into the code of this virtual world and to support more
ethical and democratic decision making in the game. In this
workshop I want to present this project as a case of
designing games for certain social structures using the
concept of not only procedural rhetoric and code but also
procedural ethics and to explore the potential of games as
an alternative political medium. For this purpose I am very
interested in developing and using the framework for the
design, development, and evaluation of the social structures
of online video games that is to be created in the work
shop.

Author Keywords
Game design, social impact, procedural ethics, procedural
rhetoric, code, co-creation, Add-on, User interface
modification,
ACM Classification Keywords
Human Factors; Design; Theory
INTRODUCTION
Interaction in and with digital media is governed by their
code. This means that communication with others over
digital media is only possible when the code permits it and
that organizations, power structures and decision-making
processes have to follow the a priori plan of the creator of
the digital medium [12]. This is particularly visible in
virtual worlds (hereafter VWs) in which groups are bound
to follow the social architecture implemented into the code
of the game. From a perspective of rhetoric the structure
and game mechanics of digital games can be seen as a form
of procedural rhetoric that enforces itself during play and
that has a political message that can be analyzed and
critiqued [2] Williams (2006, p.666) argues that []
issues of code and interface are paramount, as it is software
that enables, and often prevents, certain kinds of
interactions between certain kinds of participants. In a VW
for instance, the restrictions limit the ways that
users/players can organize a group (clan, guild, party)
within it. He suggests studying online game architectures in
order to understand the impact of different kinds of games
and to be able to generalize findings from the analyses of
individual titles. He links social online games to civic
activism and points out that future research might allow
conclusions about what attributes of games can lead to
which civic and social outcomes (Williams 2006, p.668).
The structure of leadership and power that is coded into a
game influences the emerging culture and patterns of
socializing in the game. Consequently, this would change if
the existing model were replaced.
World of Warcraft1 [3] (hereafter WoW) is one example of
a VW. It features an open Application Programming
1
WoW is a good example here for several reasons. First of all it is the VW
and MMORPG that has been most successful VW in the western world. It
Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.

DESVIG 2013
73
Interface (hereafter API), which means that third-party
software, scripts coded in the language LUA, can access
and manipulate information in the game to visualize it in
sophisticated ways or offer additional functionality to
players. These pieces of software are called add-ons and
their use is widespread in WoW. Many of the interface
elements that are now a part of the standard interface of
WoW actually originated as add-ons that got popular
because of their utility. Blizzard, the company behind
WoW, has continuously been adding functionality from
add-ons to new releases of the basic interface, thus showing
that they are both welcoming and taking advantage of user
participation in their game development [14]. Following
Williams suggestion for further research, this paper will
present the design vision and theoretical framework of a
digital tool for ethical decision making that will be
implemented into the interface of WoW. Its purpose will be
to supply players with means to modify the power structure
built into the code of the VW and support more ethical and
democratic decision making. Games, especially VWs,
function as educational tools, where not only cognitive
skills like problem solving, situation awareness and
scientific thinking can be developed but also important
social skills [22; 23; 18; 17]. This, we argue, will give
players an opportunity to reflect on issues of power not only
in the game but also in broader societal contexts.
THEORETICAL FRAMEWORK
The theoretical framework of the design document for the
interface modification uses the Lessigs [12]notion of code
and Bogosts [2]procedural rhetoric where the call for the
workshop uses the term social structures.
Code as Law
To show the importance of the infrastructure, or
architecture, of digital media for its social effects, it is
useful to compare it to other structures in our society that
are aiming at influencing emerging culture and behavior. A
particularly fruitful comparison here is made by Lessig
[12], comparing digital code to law. However, there is
another central difference between code and law. Lessig
explains:
the architecture of the Netor its
codeitself becomes a regulator. In
this context, the rule applied to an
individual does not find its force from the
threat of consequences enforced by the
lawfines, jail, or even shame. Instead,
the rule is applied to an individual
peaked at about 13 million players and is now serving about 11 million
paying subscribers worldwide. Second, WoW has made the genre of
MMORPGs popular and successful in the western market and, in a way,
created the standard against which other similar games have to be
measured.

through a kind of physics. A locked
door is not a command do not enter
backed up with the threat of punishment
by the state. A locked door is a physical
constraint on the liberty of someone to
enter some space.(Lessig, 2006, p. 81-
82, emphasis added)
In this way code is thus very different from law as it
automatically enforces itself instead of motivating
compliant behavior with threats of punishment. However,
the difference goes even further. Code is the material with
which digital media are built. That means that it is not
something external to the medium that enforces limitations
to a perceived freedom. Instead it is what dictates all
possibilities for action and interaction.
Virtual world structure and effects
The structure of the digital medium influences the emerging
interaction of the users and the design of the game promotes
certain kinds of play. The interface and the rules of the
game are part of its structure. This structure then influences
the practices of the players which come to define the social
and cultural parameters of the worlds they inhabit. [22]
Williams [23] gives two examples for the structure of
games:
some games allow for chatting between
all players, whereas others encourage it
by virtue of the goals they set for
players some games are designed for
both solo and group players, whereas
others require group coordination for
success. (Williams, 2006, p.666)
The typical MMORPG (Massively Multi-Player Online
Game) class setup with tanks, healers and damage dealers,
all of whom need each other in order to play successfully, is
another example, as is the requirement of about five players
for a group and at least ten players for a raid group. These
structures force people to socialize if they want to play
high-level content. The interaction of the structure of a
game and the practice of the players in constituting the
culture and society of the game is explained by Thomas and
Brown [22]:
The idea of the game as an institution can
help us understand how it functions in a
broader social context. Institutions
provide structure and meaning to the
game world and set the parameters for
what is possible in the space. To that end
institutions include things like the rules of
the game (both structured by the game
dynamics and mechanics and created
and enforced normatively by players)
and the challenges, quests, and spaces
provided by developers, (Thomas &
Brown, 2009, p. 40, emphasis added)
DESVIG 2013
74
The social structure of the game emerges from the game
dynamics and mechanics both of which are automatically
enforced as code in digital media and in contemporary
digital games not immediately open to be examined and
reflected on by the player. This is where the comparison of
code and game mechanics becomes useful for
understanding the relation of the player/user to both better
and it also the point where Lessigs example of a locked
door stops working. Code does not lock a door that is there
and that one could walk through. Code is not the lock. Code
is the building that the door is or is not a part of. This way
of regulation is more persuasive and harder to reflect upon
and criticize. Instead of asking for that door over there to be
opened, a critique would have to analyze and understand
the structure of the entire building to see the underlying
rationale and then demand this structure or architecture to
be changed. Here is a need for methods and guidelines for
evaluating social structures in online games that the
workshop will be working on in general and for this
particular project.
Add-ons as Social Actors
Add-ons are a major source of innovation in the interface
design of World of Warcraft [14] and they have been
described as social actors, influencing play and social
interaction in the game. The most notable accounts are
Taylors analysis of damage meters as tools for co-
surveillance [19], Chens description of add-ons as
necessary for raiding [4], and Kow and Nardis papers on
the add-on community and the way it influences play[9;
13]. Taylor explains in a very accessible way why add-ons
such as a damage meter are social actors that change the
culture in the game [20; 21]. The community that has
emerged around the production of add-ons for WoW has
created a number of websites dedicated to the topic. Most
notable here are Wowinterface.com and Wowace.com [
To give an example of the importance of add-ons for game
design and for the difficulty of raiding, Golub [6] mentions
the threat meter add-on Omen, which is mandatory for
raiders in his guild. He explains how add-ons become a
central aspect of play and how they make the game more
manageable and reduce effort, make visible invisible parts
of the game, aid players in coordinating with one another,
and capture important aspects of a players history of play
[13]. In this environment, where the structure of the
medium influences social outcomes and where add-ons can
serve as an access point to alter the way the game is being
played, the API of WoW offers the possibility to the players
and third parties like us to influence the game as a medium
and by that to become a social actor in the game.
The Current Structure of Leadership in WoW
Right now WoW, like many other MMORPGs, is using a
model with one all-powerful leader for many in-game
groups. The roles that exist are currently guild leader, group
leader, and raid leader. This also means that, for better or
worse, all decisions that have to do with for example
removal and invitation of group members and sometimes
also loot distribution have to be taken by the group leader.
It is possible to delegate some of these tasks and appoint
lower-level leaders but the main responsibility for decisions
lies with the leader. However, having all this responsibility
as a leader, with the possible stress and conflicts that it may
cause, is only one side of the problem. It also means that a
player is enjoying fairly unlimited power, purely with the
legitimation of being the person who started the group.
There is no democratic element in the decision making
other than perhaps a mutiny, where the players leave the
group and start a new one without the former leader, or at
least with a different all-powerful leader. As mentioned
above, the structure of the game is not determining play.
Players can attempt to find ways out of this structure.
Guilds can, for example, establish a council model where
there are a number of players at the second-highest level of
power who reign over and organize it together, while the
formal one-person guild leadership is residing within an
unused character. The existence of such alternative
practices shows that some players reflect on and renegotiate
the power structures in the game. It also shows that the
power structure that is coded into the game is not
determining the social outcome. Players can subvert the
existing power structures. However, establishing something
like this demands a conscious decision and requires
hacking, or circumventing, even subverting, the game
design. It requires the leader of the guild to freely share her
power. She could still betray the group and switch back to
the unused formal guild leader and take back the full
control by for example kicking the council members from
the guild. The power structure does have an influence on
culture by making one way more difficult to set up and
maintain than another one. Establishing a different structure
requires effort, mutual trust, and a deliberate decision by
the players. In this situation our add-on will offer an easier
way for players to modify the structure and support
practices not explicitly intended by the game designers. By
offering an alternative, it also points out that there is a
possibility to make a decision here. The add-on can thus
create opportunities for a more democratic dialogue.
An Alternative Model for Democratic Decision Making
For reasons of space I will at this point not go into the full
details of the design of our add-on but say that it relates to
this subversive practice in that it is intended to empower
players to break loose from the prescribed power system
that is coded into the game, thus intentionally altering the
social setup in the game world. The add-on is an adaptation
of an online collaborative decision-making tool called
ColLab
2
. ColLab has been developed as a platform for
collaborative and democratic decision making. [11] The
central idea of ColLab is that a tool for decision making
2
You can find and use ColLab here: http://interact.it.uu.se/collab
DESVIG 2013
75
should confront the users with the influences of their
decision on all the stakeholders and their respective
interests [8]. It requires the user to first gather data about
the situation, i.e. all the stakeholders and their interests and
connections to each other. All the stakeholders in a conflict
have various interests that are affecting the dynamics of it.
Furthermore, they form relationships to each other that may
create additional conflicts or increase the severity of the
existing one. There can be tensions, rivalry, competition or
any other shared conflict history influencing the current
situation. Any decision in a conflict means great risk of
hurting the interests of a part of the stakeholders. However,
by viewing a possible decision from the point of view of
every other group and with their, often very relatable and
understandable interests in mind, the user will get help to
counter her cognitive biases [10]. By gaining an
understanding of the social implications from ones own
decisions, the user will be in a better position to fulfill
Immanuel Kants maxim to act so that the decisions one
makes can be considered right, regardless of who makes
them and who is subjected to them [7]. The tool thus gives
support in determining the decision that is most consistent
with ones own morality and is also developing awareness
of moral problems and skills to deal with these.
If players choose to use it, the add-on will enable a new
practice where authority is distributed between players
instead of being concentrated in one arbitrarily assigned
leader. Our add-on will change the structure of the game by
increasing the visibility of the moral choice for both the
player and co-players. It will become apparent to players if
their leader acted against the will of the majority of the
group. Another effect will be that it will make the default
structure that is coded into WoW visible by creating an
alternative to it. The add-on emphasizes the social
consequences for behavior within the multiplayer aspect of
the game. Players will be empowered to apply social
pressure on the leader. The add-on will not enforce the
groups will automatically but if the existing leadership
structure is only an unconscious reproduction of the frozen
discourse of MMORPGs, then presenting a working
alternative could change game design practice and lead to a
democratic decision making system that can actually
enforce a majority vote.
Here we want to use three examples to illustrate how the
add-on is going to work. The first example is a discussion
in a guild about its raid schedule. If there are parents among
the members, they may want the raids to start later to be
sure that the kids are already in bed while others who have
to get up early for their work might want to start earlier.
Here the add-on can serve to provide an overview over
what different guild members want and how any decision
would impact them. The members would have the
possibility to propose compromises and ultimately to vote
on a solution. For the second example we can think of a
guild in which a number of members are in an emotional
conflict with each other, something that is putting strain on
the guild atmosphere. The add-on would here structure the
conflict and help the arguing parties to understand each
others positions. It would also provide a way for other
guild members that are disturbed by the conflict to push for
and participate in a solution. If the fighting parties cannot
find a solution, the add-on could make it possible to decide
in a fair way about a solution, e.g. to kick one side of the
conflict from the guild.
All parties would have the opportunity to contribute to the
discussion and the leaders of the guild could see how their
members are feeling about the conflict in order to avoid a
situation where they kick a player out and realize that others
are following her, thus tearing the guild apart in the process.
The third example is a dispute over loot in an ad-hoc raid
group. Here players are not part of a durable group and are
only banded together for this particular level. The add-on
can be used to limit the power of the raid leader to abuse
her power for loot distribution. For example, if the leader
wants to give a dropped item to a guild member of his, who
is also part of the raid, even though another player would
also have a claim on the item, that player could open a
conflict in the add-on, choose the pre-made Loot Drama
conflict setup, quickly adapt it to the situation, and
broadcast it. The other players could then modify it
collaboratively (much like a Google document), add
stakeholders and solutions, and finally vote for a solution. If
this vote results in support for the player who did not get
the item, then the raid leader has to choose to either follow
the verdict of the group or to act against it openly, with the
risk for players leaving the group. The add-on will make it
possible to discuss how decisions should be made and how
power structures should be implemented into the game. It
might not enable picking the apple from the tree, to return
to the earlier used Garden of Eden example, but it will at
the very least make it possible to see the tree and to then
ask questions about why the apple is untouchable and if that
should not be different.
Another advantage of an add-on like this is that it can be
used for recording player behavior in conflict situations
over a long period of time. Such a research direction has
ethical and privacy implications that need to be sorted out
first. It would require the participating players consent for
aggregating the information, but there is great potential for
understanding not only the use of the particular add-on but
also game play generally.
DESVIG 2013
76

Figure 1: The structure of the add-on interface
CONCLUSION AND FUTURE STUDY
There are a number of possibilities for further research
resulting directly from this project. Besides offering and
testing a new framework for decision-making in virtual
environments, which could be applicable in wider context
than just VWs and online games, this project holds the
possibility to collect data on what players in these spaces
are having discussions and arguments about and how they
are having them. Quantitative studies with users of the tool
will make it possible to draw conclusions about what kind
of effects power structures in games have on players. To be
able to do this the framework for the design, development,
and evaluation of the social structures of online video
games that is to be created in the workshop will be useful.
In order to be able to optimize the design of the add-on and
to evaluate the reaction of the player, as well as the effect of
the decision-making process, we will test the add-on with
different and divergent groups of players. The users will fill
in questionnaires about the add-on, focused on the way that
the add-on has changed conflict resolution in the game. The
authors will conduct in-depth interviews with a number of
diverse players representing different play styles and
approaches to get the most possible points of view on the
add-on, and given that we get players consent, we can
collect data about the use of the add-on, focused on the
number of conflicts resolved, the time each conflict took to
resolve and the default conflicts that were used by the
players.
Digital games are the medium in which many people make
meaningful experiences. Virtual worlds are here especially
important as a space to experience sociability and culture.
With this project we aim to pave a way to change the design
of these spaces in order to make the emerging societies in
virtual spaces more democratic, aware of ethical problems
in their decision making, and reflective of the architecture
DESVIG 2013
77
of power in digital media. Players could potentially take
these lessons with them and apply them to politics and
society in the physical world, thus increasing their
capability to take responsibility for matters of society and to
demand changes, participation, and power. Players might
learn from their experience in the game that power
structures are something that has a political agenda and that
can be modified according to the peoples moral values. We
hope that instead of learning to accept the setup of society
as unchangeable, players will learn to reflect on it, to doubt
it, and to expect legitimization for it, should they find it
undesirable. In times of global economic and political
struggle, we hope that virtual worlds can be more than just
an escape to a different space but instead an inspiration to
societal change. This is our aim.
REFERENCES.
1. Aristotle. Nichomachean Ethics. Retrieved from The
Internet Classics Archive (350 B.C.E):
http://classics.mit.edu/Aristotle/nicomachaen.html
2. Bogost, I. Persuasive Games: The Expressive Power of
Videogames, MIT Press. 2007.
3. Blizzard Entertainment. World of Warcraft. Irvine: CA:
Blizzard Entertainment. (2004-2011)
4. Chen, M. Communication, Coordination and
Camaraderie in World of Warcraft. Games and Culture,
4(1), 47-73, Sage Publications. (2009).
5. Gee, J . What video games have to teach us about
learning and literacy. New York: Palgrave Macmillan.
(2003).
6. Golub, A. Being in the World (of Warcraft): Raiding,
Realism, and Knowledge Production in a Massively
Multiplayer Online Game, Anthropological Quarterly
(2010), 83(1), 17-45,
7. Kant, I. Fundamental Principles of the Metaphysic of
Morals. (1785). Retrieved from Project Gutenberg:
http://www.gutenberg.org/etext/5682
8. Kavathatzopoulos, I. The use of information and
communication technology in the training for ethical
competence in business. J ournal of Business Ethics
(2003). 48(1), 43-51, Kluwer Academic Publishers.
9. Kow, Y. & Nardi, B. (2010). Who owns the mods?
First Monday 15(5).
10. Laaksoharju M. Let Us Be Philosophers! Computerized
Support for Ethical Decision Making. [Thesis]. Uppsala:
Department of Information Technology, Uppsala
University: IT licentiate theses / Uppsala University,
Department of Information Technology, (2010).
11. Laaksoharju In support of democratic dialogue. In
Critique, Democracy and Philosophy in 21st Century
Information Society - Towards Critical Theories of
Social Media, The Fourth ICTs and Society-Conference,
Collection of Abstracts (2012). Retrieved from The ICT
and Society Network: http://www.icts-and-
society.net/wp-content/uploads/Abstracts.pdf
12. Lessig, L. Code. Basic Books, New York, 2006.
13. Nardi, B., & Kallinikos, J . Technology, Agency, and
Community: The Case of Modding in World of
Warcraft, In Holmstrm, J ., Wiberg, M., & Lund, A.
(Eds.), Industrial Informatics Design, Use and
Innovation: Perspectives and Services (2010). 174-186.
14. Prax, P. Co-creative interface development in
MMORPGs - the case of World of Warcraft add-ons,
J ournal of Gaming and Virtual Worlds (2012). 4 :1, pp.
324.
15. Prax, P. and Laaksoharju, M. Democracy has arrived! A
Model for Ethical Decision Making of Players in
MMOs, Procedings of Meaningful Play 2012, (2012)
http://meaningfulplay.msu.edu/proceedings2012/mp201
2_submission_16.pdf
16. Sherlock, L. Genre, Activity, and Collaborative Work
and Play in World of Warcraft: Places and Problems of
Open Systems in Online Gaming, J ournal of Business
and Technical Communication (2009). 23(3), 263-29.
17. Squire, K., & Steinkuehler, C. Generating
cyberculture/s: The case of Star Wars galaxies. In
Cyberlines 2.0: Languages and cultures of the Internet,
ed. D. Gibbs and K. Krause (2006). 177198. Albert
Park, Australia: J ames Nicholas Publishers.
18. Steinkuehler, C. & Duncan, S. Scientific Habits of Mind
in Virtual Worlds. J ournal of Science Education and
Technology (2008). 17(6), 530-543.
19. Taylor, T.L. Does WoW Change Everything? How a
PvP Server, Multinational Player Base, and Surveillance
Mod Scene Caused Me Pause, Games and Culture
(2006a). 1(4), 318-337.
20. Taylor, T.L. Play between Worlds: Exploring Online
Game Culture. Cambridge: MIT Press. 2006b.
21. Taylor, T.L. The Assemblage of Play, Game and
Culture (2009), 4(4), 331-339, Sage Publications.
22. Thomas, D., & Brown, J . Why virtual worlds can
matter. International J ournal of Media and Learning
(2009), 1(1), 37-49
23. Williams, D. Groups and Goblins: The Social and Civic
Impact of an Online Game, J ournal of Broadcasting &
Electronic Media (2006). 50(4), 651-670

DESVIG 2013
78
Encouraging Co-located Players Social Interaction
Sofia Reis, Nuno Correia
CITI, DI, Faculdade de
Cincias e Tecnologia
Universidade Nova de Lisboa
2829-516 Caparica Portugal
se.reis@campus.fct.unl.pt, nmc@fct.unl.pt

ABSTRACT
This position paper presents our ongoing work for
stimulating co-located social engagement among players
with a game that resorts to face recognition and another
game where players are accompanied in their daily lives by
virtual projected imaginary friends. We suggest interaction
between imaginary friends of different players that will
work as a bridge for interaction among their respective
human companions. In the context of the games and of the
initial results obtained a framework for encouraging social
interactions in large groups of players is outlined.
Author Keywords
Games; Social interactions; Co-located.
ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g., HCI):
Miscellaneous; K.8.0 [Games].
1. INTRODUCTION
For centuries, social experiences have been part of games
[14]. Social contact can be remote or co-located, but co-
located interaction contributes more to player enjoyment
than remote interaction [1]. Social contact in games may
involve players or even strangers to the game [8].
Here we present our proposals to stimulate social
engagement among players that are in each others physical
presence. This interaction may be among players who were
acquainted before the game or the game may act as a
platform for strengthening or building new social links.
The next section addresses previous related work by other
authors. Section 3 describes how we encouraged social
interaction. Section 4 outlines a framework for stimulating
social engagement among co-located players. Finally, the
conclusions and future work are presented in Section 5.
2. RELATED WORK
Other authors have already presented interesting results in
what concerns motivating players for social co-located
engagement. In MobiComics teams of users in a common
space cooperate to create comic strips, in their mobile
phones, that are afterwards shown in public displays.
MobiComics managed to encourage social interaction, in
the form of discussion and laughter, especially when users
vote for the comic strips [7]. However this strategy is aimed
at small groups of players. We intend to create a strategy
that may be used by large groups.
Other authors present a Pong like game where two players
sit face to face, across a table, and use mobile phones as a
shared display. They reached the conclusion that when
sitting in front of each other, oral communication was
significantly higher than when sitting side by side. However
this game is designed for groups of only two people [4].
Schminky is a musical game designed to be played at a cafe
or bar environment. Users listen to a melody and four tracks
and have to identify which track is not playing. As people
who were playing the game were easily identifiable by the
headphones and portable computer that resulted as a
motivator for people to interact with strangers sitting at
other tables [11]. In our proposal we intend to take this side
effect, related to the visibility of the players, one step
further and turn it into the core of our proposal, instead of
being just a consequence.
Another work connected two night clubs with two
interactive tables where people could draw and use pre-
made symbols and actions. Even though some results are
presented about people who drew locally together, the main
emphasis was on the joy of communicating with people in
another location [17].
Capital Music is an application that allows users to share
songs with co-located strangers. Even though this
application may create a sense of a link with other people
nearby, all social user interaction is anonymous [15].
3. ENCOURAGING SOCIABILITY
Our objective is to promote social co-located interaction in
online videogames. We intend players to be in each others
physical presence and not competing or collaborating
remotely as often happens in MMORPGs. Furthermore, we
are aiming for that social interaction does not occur only
during the game, but also after the game, so that social
connections among players may be strengthened or




Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.

DESVIG 2013
79
eventually the game may be an instigator to form new
social connections.
The Enchanted Moor browser game [13] was one of our
attempts for promoting social engagement among players.
Enchanted Moors are mythical characters that have been
trapped by spells for centuries in the Iberian Peninsula.
According to legend brave heroes who manage to break the
spell and free the enchanted moor will be rewarded with
vast treasures [2,10,16].
In our browser game the player wanders through a forest
until she finds an enchanted moor princess. The princess
will then tell the player that she needs two people to look at
her magic mirror so that the spell is broken and she is set
free. We resorted to J avascript, HTML5, and to an Haar
Cascade Classifier [5,6] coded in AS3 to detect how many
persons are looking at the camera. Figure 1 (a) shows the
games video capture with the players faces that are
highlighted, in real time, by the outline of a green box.
After the two saviors join their efforts the princess is
released and offers the player a safe (Figure 1 (b)) filled
with gold coins and a magic potion (Figure 1 (c)). This
particular game is incorporated in a larger organized frame
of games where we stimulated remote social contact via
publication of stories authored by players, forums, creation
of in-game objects by players and player union via
brotherhoods. However, here we focus only on co-located
interaction.

(a)

(b)

(c)
Figure 1. I n the Enchanted Moor game (a) after two players
join their efforts to save a princess they are rewarded with a
(b) safe filled with (c) gold coins and a magic potion.
With our game we ensured that there would be social
contact because the player has to convince someone else to
assist her so that the camera from the device where the
game is running can detect two players. Mutual physical
presence of the player and the helper in front of the same
camera is therefore essential to win.
To recruit players we resorted to mailing lists and social
networks. Players tested the game in their own computers
without interference from the researchers. 49 players
answered our online survey about the game. From the 49
answers, 6 were discarded because we found out, through
those players answers to the survey that they had not
allowed the game access to their camera. Without the
camera it is not possible to play and consequently these
respondents were not in a position to evaluate the game.
The characterization of the players who turned in valid
survey responses is presented in Table 1. Answers from
respondents who denied access to the game were
considered invalid.
Enchanted Moor Survey
Number of valid survey answers 43
Gender 34.9%
62.8%
No answer 2.3%
Age Average 20
Range 10-35
Standard deviation 6.9
Countries Brazil 46.5%
Portugal 41.9%
Other countries 9.2%
Did not answer 2.3%

Who helped
the player free
the Enchanted
Moor
Friend 37.8%
Mother 20.0%
Father 8.9%
Sibling 8.9%
Work colleague 8.9%
Others 15.6%
Where the
player was
Home 74.5%
Friends house 10.6%
Work 8.5%
School 4.3%
Relatives house 2.1%
Table 1. Enchanted Moor game survey respondents
characterization and results.
By resorting to face recognition it was our intention to
motivate users for social contact. However, three players
confessed us that they had cheated by showing an image to
the camera. Still, these answers were considered in our
analysis because these respondents, contrary to the ones
who denied access to their camera, actually played the
game, though not in the way we expected. This also shows
that even if a game forces players to socialize, still some of
them may find a way to bypass that requirement and still be
able to win.
DESVIG 2013
80
The Enchanted Moor game encouraged interaction mainly
at home and with friends (Table 1). However we intend to
stimulate wider forms of interaction. For example, how
could interaction be encouraged, not only between friends
and family, but also among people who are hardly familiar?
How it is possible to more easily trigger social contact for
shy people in social events?
In one of our projects we developed an Imaginary Friend
that is projected on the floor via a pico projector. This
virtual companion shares a link with the user and can sense
her emotions. As the user walks around, in her everyday
life, the Imaginary Friend follows and collects the emotion
cookies that the user leaves behind. Previous emotion
cookies can be consulted inside a jar or in a map. The
Imaginary Friend helped players to think about their past
emotions and take decisions about their lives [12].
In Figure 2 a user is pictured with her Imaginary Friend.
The pico projector is positioned on the users left shoulder.
The user is also wearing an electrodermal activity sensor on
her right wrist that is connected to a PLUX device [18]. The
Imaginary Friend resorts to variations on the electrodermal
activity to know when to question the human companion
about what she is feeling.








Figure 2. User with the I maginary Friend.
The game was not initially designed to test sociability but
the Imaginary Friend may serve as a social trigger for timid
people because other people, who are looking at the
projection, may feel the impulse to talk with the user.
Eventually, in social events, the projected Imaginary
Friends may even pull together their human companions
based on the human companions similar interests or
personal characteristics. The Imaginary Friend may even
assist by suggesting the initial conversation topics. In
Figure 3 a party where the users each have an Imaginary
Friend is represented. Eventually, in parties people may
tend to talk only with the people they already know.
However, in the situation depicted in Figure 3 one of the
Imaginary Friends discovers that another person at the party
also enjoys the city of Paris. The Imaginary Friend then
tries to pull the human companion for a conversation with
this still unknown person about the city of Paris. Besides
parties, the virtual companion can also be helpful for
example, with children, at the beginning of the school year,
so that students may more easily know others based on
similar hobbies or preferred activities. In professional
events, such as conferences, Imaginary Friends could pull
together people who work on similar areas.
For this strategy to work, it is necessary that people,
gathered at an event, have pico projectors. There are
already several commercially available pico projectors, as
standalone devices or embedded in mobile phones, so pico
projectors may become quite common in the near future.

Figure 3. I maginary Friends suggesting possible social
connections.
4. SOCIABILITY FRAMEWORK
We intend to expand our efforts and to create a sociability
framework (Figure 4) that can be applied to encourage
social contacts resorting to games.
This framework will, firstly, focus on joining the players.
To assure a common place it is possible to appoint a
latitude and a longitude. Currently smartphones are usually
equipped with GPS and location can also be determined via
the Geolocation API Specification, thus enabling their
owners to head themselves towards a determined position.
After assembling the players, step 2 is to trigger social
interaction for the game. It is not enough for players to
simply be at the same place, but unaware of each other and
to bypass the chances for social connection. GPS suffers
inaccuracies and location via the Internet is even less
accurate so players may be several meters apart. In the
Enchanted Moor we resorted to face detection to assure that
players are effectively side by side and looking at the
camera. However, interaction was mostly with people
familiar to the player. We intend to stimulate this
engagement even among strangers or people who scarcely
know each other. Flash mobs have already demonstrated
that it is possible to join crowds just for the sake of
entertainment. Projected multimedia via pico projectors
may also function as a way for players to spot one another.
Pico projector
Imaginary Friend
EDA sensor
DESVIG 2013
81
After the initial inertia is overcome, players enter the magic
circle [3,9] and actual gameplay begins.
In step 3 it will be important to evaluate how immersed the
players were in the game, how engaging and fun was the
social interaction and how the game affected social
relationships between people. Our objective here is that the
game functions not only as a way to join people and pass
the time but also to check how an entertaining activity can
influence how close people feel to one another and even
help forge new social links.

Figure 4. Outline of a sociability framework for co-located
interaction in games.
5. CONCLUSIONS AND ONGOING WORK
In this paper we presented our results for a game that
encourages social co-located interaction among players
resorting to face detection. We intend to extend this
approach to large groups. A strategy for creating or
strengthening social connections via Imaginary Friends that
act as a bridge between their human companions was also
proposed. Virtual companions are projected on the floor
and walk by the side of their users. The virtual companions
will rely on the users characteristics and will propose
engagement between them based on common similarities. A
sociability framework was outlined that addresses the steps
necessary for joining players at the same location,
triggering the social interaction and finally engaging in
play.
6. ACKNOWLEDGEMENTS
This work was partly funded by FCT/MCTES, through
grant SFRH/BD/61085/2009, and by Centro de Informtica
e Tecnologias da Informao (CITI/FCT/UNL) - 2011-2012
through grant PEst-OE/EEI/UI0527/2011.
7. REFERENCES
1. Gajadhar, B.J ., Kort, Y.A., and Ijsselsteijn, W.A.
Shared Fun Is Doubled Fun: Player Enjoyment as a
Function of Social Setting. In P. Markopoulos, B.
Ruyter, W. IJ sselsteijn and D. Rowland, eds.,
Proceedings of the 2nd International Conference on
Fun and Games. Springer Berlin Heidelberg, 2008,
106-117.
2. Goberna, B., Dacosta, T.O., and Mosquera, J .M.L.
Myths, legends and beliefs on granite caves. Cadernos
do Laboratorio Xeolxico de Laxe 33, (2008), 19-34.
3. Huizinga, J . Homo Ludens. Beacon Press, 1971.
4. Kauko, J . and Hkkil, J . Shared-screen social gaming
with portable devices. Proceedings of the 12th
international conference on Human computer
interaction with mobile devices and services, ACM
(2010), 317-326.
5. Klingemann, M. Quasimondo - Mario Klingemanns
Flash Blog: Optimizing Flash Based Face Detection.
2009.
http://www.quasimondo.com/archives/000687.php.
6. Libspark. mash/Marilena - Spark project.
http://www.libspark.org/wiki/mash/Marilena.
7. Lucero, A., Holopainen, J ., and J okela, T.
MobiComics: collaborative use of mobile phones and
large displays for public expression. Proceedings of the
14th international conference on Human-computer
interaction with mobile devices and services, ACM
(2012), 383-392.
8. Montola, M. and Waern, A. Participant roles in socially
expanded games. Third International Workshop on
Pervasive Gaming Applications, Pervasive
Conference., (2006), 165-173.
9. Montola, M., Stenros, J ., and Waern, A. Pervasive
Games: Theory and Design. Morgan Kaufmann, 2009.
10. Parafita, A. A Mitologia dos Mouros. Gailivro, 2006.
11. Reid, J ., Hyams, J ., Shaw, K., and Lipson, M. Fancy a
Schmink?: a novel networked game in a cafe. Comput.
Entertain. 2, 3 (2004), 11.
12. Reis, S. and Correia, N. An imaginary friend that
connects with the users emotions. Proceedings of the
8th International Conference on Advances in Computer
Entertainment Technology (ACE 11), Lisbon,
Portugal, ACM (2011).
13. Reis, S. and Correia, N. Enchanted Moor.
http://schoolofmagic.net/florestamoura.asp.
14. Salen, K. and Zimmerman, E. Rules of Play: Game
Design Fundamentals. MIT Press, 2003.
15. Seeburger, J ., Foth, M., and Tjondronegoro, D. The
sound of music: sharing song selections between
collocated strangers in public urban places.
Proceedings of the 11th International Conference on
Mobile and Ubiquitous Multimedia, ACM (2012),
34:1--34:10.
16. Vasconcelos, J .L. Volume V Etnologia (Parte I).
Imprensa Nacional, Lisboa, 1938.
17. kerman, P., Puikkonen, A., Huuskonen, P.,
Virolainen, A., and Hkkil, J . Sketching with
strangers: in the wild study of ad hoc social
communication by drawing. Proceedings of the 12th
ACM international conference on Ubiquitous
computing, ACM (2010), 193-202.
18. edaPlux | PLUX. http://plux.info/EDA.


Step 1.
J OIN
Step 2.
TRIGGER
Step 3.
PLAY
Co-located Social
Interaction
DESVIG 2013
82
Designing Games That Foster Equity and Inclusion:
Encouraging Equitable Social Experiences Across Gender
and Ethnicity in Online Games
Gabriela T. Richard
New York University
82 Washington Sq. East, 6
th
Fl.
New York, NY 10003
gabriela@nyu.edu


ABSTRACT
Emerging research and current media events are beginning
to highlight gender and ethnic inequities in online game
culture. Many aspects of game culture continue to exclude
participation by females and ethnic minorities, particularly
through environments of bias and harassment, which can
hinder their sociability. This paper highlights emerging
research on gender and ethnic inequity in gaming,
particularly making links between representations of gender
and ethnicity, harassment, and social exclusion. The paper
also highlights case studies of gamers, which help to
underscore those links. Finally, this paper offers design
principles, grounded in the research and case studies that
can help foster equity and social inclusion in games.
Author Keywords
Computer games; Interactive games; Game culture; Gender;
Race; Ethnicity; Online gaming; Equity; Inclusion; Social
interaction; Inclusive game design; Design principles
ACM Classification Keywords
Human-centered computing; Human computer interaction
(HCI); HCI design and evaluation methods; User studies;
Field studies; Collaborative and social computing;
Collaborative and social computing theory, concepts and
paradigms; User characteristics; Race and ethnicity;
Gender; Cultural characteristics; Computer games;
Interactive games; Massively multiplayer online games
General Terms
Human Factors; Design; Theory.
INTRODUCTION
This paper proposes the concept of designing games for
inclusion and equitable social experiences. For the past 30
years, researchers have pondered the link between gaming
experiences and gender, though more recently researchers
are also investigating racial and ethnic experiences.
However, recent events have highlighted the propensity for
gaming culture to be negative and even hostile to females
and racial or ethnic minorities. I propose that this
encourages the exclusionary nature of social experiences in
online games. I also propose that, based on related research
on ways to reduce differential environmental support, as
well as case studies provided of gamers across gender and
ethnicity, there are design principles we can incorporate to
increase equitable social experiences with online games.
GENDER AND GAMING
Research on gendered experience with digital games has
spanned the past 30 years, with most of the early work
exploring sex and gender differences in desire to play, play
styles, preferences and ability [for example, see 1, 2], and
later work exploring more contextual and intersecting
constructs, like ethnicity and sexuality [3]. Some work
continues to explore gender or sex differences. For
example, these studies have found females exhibited less
confidence to play and engage with games and computer
technology [4], played significantly less than males, had
significantly less spatial ability than males, and had a lack
of desire for competitive play [5, 6, 7]. However, it should
be noted that these studies did not make a distinction
between gamers and non-gamers, as well as didnt
explicitly explore what other factors could be contributing
to gender differences.
Other research is positing that what was initially thought of
as gender differences actually may have more to do with
contextual factors, like age [8], available leisure time [9],
and overall differential access, experience, and support [10]
often entangled with one or more of these factors. For
example, in one complex quantitative study with a large
sample of gamers across gender [11], researchers found no
gender differences amongst female and male gamers who
played core games (typically associated with male-
oriented themes, competition and high time investment),
but did find differences between females who played more
casual games and everyone else (including casual male
gamers). Furthermore, researchers found that females were
required and encouraged to participate in more
responsibilities than males, leaving them less time to play
[9]. Other researchers found that females are often given
differential social support in gaming spaces, including
public spaces and the home, where they are encouraged to
take on supportive passive (and gendered) roles, while
males are, conversely, encouraged to dominate and show
their skills [12, 13]. In particular, research is finding that
Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.
DESVIG 2013
83
when females are given equal access to play and train,
gender differences decline [10]; specifically, one study
found that playing action games for a relatively small
amount of time resulted in similar spatial skills across
gender, with females experiencing significant gains [14].
This corresponds with similar research on social context
and spatial skills in engineering where research found that
gender differences in spatial skills (once thought to be
mostly biological) were leveled once females had the
chance to train [15].
In other words, this research shows support for the idea that
ones gender was more likely to influence the kinds of
experiences one was encouraged to engage in. Thus,
females are less likely to be encouraged to play games in
the same ways, and are thus less likely to have the same
access to experiences that would put them on the same level
as their male counterparts. However, social support and
training appears to lessen many gender differences.
ETHNICITY AND GAMING
Most research on differential experiences with games and
game culture has almost exclusively addressed gender, with
research on players of different races and ethnicities only
recently gaining exposure in the literature [16]. Several
notable studies have started to explore the gaming and
technology experiences of ethnic minorities [for example,
17, 18, 19]. Research is finding, for example, that ethnic
minorities dont have the same level of access to high-tech
computer equipment for leisure as Whites [18], and, in
particular, ethnic minorities cite experiences of ethnic and
gender harassment in online gaming [17, 19]. Specifically,
this research demonstrates that players are profiled based on
the appearance of their avatars, and the way they speak
(linguistic profiling) [19].
GAME CULTURE, INEQUITY AND EXCLUSION
There are several contextual factors that may be influencing
differential encouragement, support and play experiences in
game culture along gender and ethnic lines, which can, in
turn, result in differential sociability, particularly for
females and ethnic minorities.
The way that games are marketed can play a big role in
excluding females and ethnic minorities, something I
termed recently as entry points that influence access to
the narrative [20]. In other words, certain players may miss
out on games they may like because they are marketed in a
way that makes them seem like they are unappealing across
gender or pandering to gender roles, even if the designed
experience and narrative within the game doesnt. For
example, newer games that allow you to play as a female or
male character, and the ethnicity and sexuality of your
choice are marketed as if they have a main White male
heterosexual playable character, even though they dont
[see 20, for examples]. Further, researchers have critiqued
game marketing for primarily being featured in male-
oriented media (like male interest themed magazines and
television) [21]. In a recent study, researchers found that
when individuals felt it wasnt easy to find an appealing
game, they were less likely to play digital games, across
gender [9], though the amount of male non-gamers was
significantly smaller than female non-gamers. While that
study didnt explicitly discuss marketing, other researchers
have highlighted that the lack of visibility of females and
ethnic minorities in gaming magazines and popular media
can reinforce a context of exclusion whereby they may feel
less inclined to play [22].
Further, newer research shows that the lack of female and
minority playable characters in games may have an effect
on female and minority game players own sense of
belonging [23], as well as other individuals sense of
whether they belong [24], in online gaming and virtual
worlds. In particular, content analysis of the most popular
commercial computer and video games were found to
contain mostly male (over 85%) and mostly White (over
85%) primary playable characters; furthermore, females
and ethnic minority characters were highly stereotyped or
highly sexualized [23]. Research has further found that
negative stereotypes of ethnic minorities affected White
players sense of who belonged, and they were less likely to
support ethnic minorities [24]. Furthermore, the lack of
ethnic minority characters negatively affected minorities
sense of belonging in virtual worlds and online games,
whereas the inverse was true when more diversity present
[25]. Finally, when playing with sexualized female
characters, females were shown to feel more negatively
about themselves and their abilities, and females and males
were less likely to positively appraise female ability,
regardless of the characters skills [26].
While online spaces were initially considered places where
marginalized gamers, particularly females, could go to
escape gendered expectations, like they were subjected to in
public gaming spaces [13, 27], newer high profile events in
online game culture demonstrate that this may no longer be
Figure 1. Game magazines and marketing for recent games
continue to market to males or reinforce stereotypes.
Figure 2. Recent high-profile events of harassment. Left: a game
made to beat up Anita Sarkeesian after she got funding to study
female tropes in games. Right: gamer Miranda Pakozdi growing
agitated, as she is sexually harassed during a tournament.
DESVIG 2013
84
the case, especially as our real identities are more visible
online; this has largely changed as the kinds of social
interactions built in games has evolved. Recent high-
profile media events in online gaming culture have shown
that females are subjected to harassment, victimization and
silencing [28, 29, 30]. Some gamers, particularly female
gamers, have started documenting gender harassment on
their own [30], through newly established websites created
to collect and document online gaming harassment.
Researchers suggest that gaming spaces are increasingly
being considered male bonding spaces that encourage this
kind of exclusivity [31]. Researchers are also theorizing
that sports and gaming, which tend to be associated with
masculinity, create gender-biased environments [32].
Because sports and gaming have been so tied to
masculinity, the playing fields have become uneven such
that a female cannot beat a male without his masculinity
being called into question, and a male cannot beat a female
without being accused of unfair aggression [32]. This
unevenness encourages a special kind of gender-based
hostility and bias.
METHODOLOGY
In an effort to understand how games could be designed to
be more equitable, particularly for learning, I conducted an
ethnography of game culture, over 3 years, where I was
both a participant and an observer, as part of my (ongoing)
dissertation. In addition to experiencing, observing and
participating in game culture, virtually and in person, I also
conducted interviews and surveys of gamers, across gender,
ethnicity, and sexuality. For the purpose of underscoring
themes related to inequity in gaming and social experiences
here, I am presenting relevant, salient descriptive case
studies of gamers
1
and events derived from the ethnography
and interviews. Descriptive case studies have been shown
to help illustrate distinctive features of a phenomenon,
particularly one that has been underexplored [33].
FINDINGS
One theme that came up in the interviews of gamers was
harassment as salient to females and ethnic minorities;
specifically, minorities were more likely to cite being
victims of ethnic-based harassment, whereas females,
across ethnicity, were more likely to cite gender-
harassment. In addition to participants discussing the
negativity of being harassed, they also discussed how that
harassment lead to their silencing, and limited participation.
This was particularly true for females. While ethnic
minority males were less likely to cite muting themselves
due to harassment, there were a few who did, demonstrating
that some of them also felt the need to silence themselves
and limit their sociability in online gaming. As most game
players who play online know, with many competitive
games, particularly on consoles as opposed to computers,

1
Names have been changed to protect identity.
voice chat is a necessary part of competitive gaming
experience. Limiting your social interaction online puts
you at a disadvantage in competition, particularly in
strategic games that require teamwork, and in some games,
not interacting on voice chat will get you booted from game
play, or result in other negative reactions from fellow
teammates.
Case 1: Cindy, female, ethnic minority, age 29
Cindy had been playing video games for the past 24 years,
having been encouraged to play by her late mother, who
loved classic role-playing games. She became particularly
adept at competitive first-person shooter games, like Call of
Duty 4, and even entered a tournament where she and a
male friend took first place. However, she typically mutes
herself during play, and tends to keep to herself, which she
describes as mostly due to harassment:
[Guys online] would send me pictures of things I didnt
want to see, or they would harass me, or if I were good,
because I was great at Call of Duty 4, theyd say I was a
guy playing under a girls name I dont talk on the mic, I
just play I just stopped talking cuz theyd be like, oh
thats a girl, lets harass her or ask for her number or
something.
Case 2: Jason, male, ethnic minority, age 33
J ason is a lifelong gamer who has played online
competitive games since the late-1990s, and was at one
time considered one of the top competitive gamers in
SOCOM, a popular competitive game (he has since changed
his gamer tag). However, most of his competitive gaming
was during the days before the prevalence of voice-based
communication and strategy. Specifically, in the early days,
only text-based chat was available; later limited voice chat
that had to be manually triggered was added on and was
often limited to only in-game interaction. As a result, he
explained that most trash-talking and voice-based deliberate
harassment was limited. Nowadays, while he still plays
competitively online, he doesnt compete in tournaments
and ladders like he used to, and he doesnt voice chat
except when he feels comfortable with players in the room,
or with friends hes used to playing with.
I guess they can tell Im Spanish from my voice I just
ignore and keep away from that. It happens a lot Most of
the time, I play with a mic on but I keep myself muted. I
want to hear just in case they want to strategize, but [the
harassment] gets annoying.
He also described an episode where he was cyber-stalked
by a fellow gamer, who discovered he was Hispanic and
later harassed him in an online match for his ethnicity.
Case 3: Paul, male, ethnic minority, age 25
Not all of the ethnic minority male players interviewed felt
silenced by the kinds of harassment they received, though
they did express being frustrated by attempts to exclude and
limit their play, particularly through voice profiling (or
sounding like a minority). One example is Paul, who has
DESVIG 2013
85
played video games for 21 years. He plays all genres of
games, spending, on average, over 40 hours a week playing,
the majority of which he devotes to online competitive
games that require voice-based strategy.
Most of my friends online are [ethnic minorities] Black,
Puerto Rican, Dominican Every time I game with them,
if they even sounded remotely with a deep voice, they were
automatically Black, they were the N word and told to go
back to Africa. Sometimes, theyd call me all kinds of
spic [and say] oh, youre Mexican, go back to Mexico.
He also has described several episodes of being ridiculed
for being an ethnic minority and told that as a minority he
shouldnt play certain online competitive games.
The guy called me an immigrant, a Mexican, and said that
we [me and my cousin] shouldnt be playing his sport
[hockey]. I told him I wasnt Mexican, Im Puerto Rican
not an immigrant to which he replied, theyre all the
same.
Case 4: Tina, female, ethnic minority, age 25
Some females, in particular, were frustrated by the
representation of female characters in games, though not all
shared that frustration. Ethnic minority males, however,
were less likely to be sensitive to ethnic minority
representation, or lack thereof.
Tina has been playing games since she was a child, having
been introduced to gaming by her uncle and older sister.
She primarily plays online music games, like Rockband,
and also team-based cooperative strategic first-person
shooter games.
Most games when you see women in them, they have
[unrealistic proportions] thats not really how women are.
In the same game, they have a realistic male and an
unrealistic female. Its sexist in my opinion. Tina,
female, ethnic minority, 25
Case 5: The Halo 4 moderation debate
However, while many female gamers expressed concerns
about being harassed, there were diverging views on how to
moderate these behaviors. In late 2012, top Halo 4
developers, two of whom are female, announced stricter
regulations against harassment, particularly sexual
harassment [34]. They further described how they made
careful decisions about female representation. However,
what stood out the most was what came to be seen as
permanent bans or zero tolerance policies for sexist talk.
While the developers never spelled out the details of their
strict enforcement against online sexist behavior, this
announcement was met with strong reactions, particularly
by female gamers in my ethnographic study. In particular,
while there was complexity to the debate, it boiled down to
one side thinking permanent bans for sexism were too
harsh, being particularly irked by how it represented female
gamers (this side felt strongly that it made female gamers
look weak and put them on the spot), and the other side
applauding the increase in moderation and protection for
females from cyber-bullying and gender harassment.
However, the lack of clarity in how exactly users would be
moderated, reviewed or regulated mostly seemed to
frustrate the debate, with one side thinking a permanent ban
would happen after one minor event, and the other feeling it
would be weighed justly by frequency and severity.
DESIGN PRINCIPLES FOR EQUITY AND INCLUSION
Differential experience by gender and ethnicity, as cited in
the literature and highlighted in the case studies, appears to
be increasing vulnerability to exclusionary practices. Based
on the research and my own data, I have derived design
principles that will assist in fostering inclusion and equity
in social experiences in and around online games.
Increase frequency and variety of ethnic minorities and
females in games
As suggested by Dill and Burgess [24] increasing the
frequency and variety of representations of ethnic
minorities is important to reducing negative stereotypes
Whites have toward minorities. Their research showed that
negative stereotypes of minorities in computer and video
games tended to make Whites feel less inclined to vote for
or support professional Black men in other contexts. In
other words, the kinds of stereotypes we have in the media
can have effects in the real world. Based on the studies by
Miller and Summers [22] and Williams, et. al. [23], females
and ethnic minorities are more likely to be less visible, less
important, and highly stereotyped in games. Based on the
Dill and Burgess study, this can have a direct relationship
on the kinds of interactions people experience in the real
world. Further support for how this can affect gamers
treatment of other gamers is evidenced in the ethnic and
gender harassment directed at females and ethnic
minorities: in some cases this harassment lead to their
silencing and lack of participation. Studies further show
support that increasing equitable representations across
ethnicity can make ethnic minorities feel more like they
belong [25]. Increasing the frequency and variety of female
representation would likely do the same.
Design Inclusive Social Networks Around Games
Emerging research is showing that supportive communities
can reduce players vulnerability to negative online gaming
experiences, particularly across gender [35]. Similar
findings have been found with reducing gender bias and
exclusion in male-dominated curriculum areas through
same sex schools [36]. Specifically, one set of hypotheses
for why females in same sex schools were more likely to be
confident in their abilities in traditionally male-dominated
curriculum areas than their counterparts in co-ed schools
stated it could be due to supportive environments with
access to female role models in those areas. Results from
the study on gender supportive communities [35] show that
females and males in female-supportive communities,
across ethnicity (majority and minority groups), show
statistically significantly higher levels of self-confidence
DESVIG 2013
86
and identity in gaming than members of a variety of other
gaming communities. Since the gender equitable online
community in the study (PMS Clan) has strict moderation
around harassment (of all forms), cheating, and trolling,
they may actually be creating an environment that most
people enjoy, regardless of gender or ethnicity. Many
games have forums and social communities built around
them. Developers should make conscious efforts to design
inclusive environments around their games, in addition to
increasing the variety of characters in their games.

Figure 3. COD Black Ops I I s reporting systems ease of use.
Adopting and Maintaining Strict and Transparent
Moderation Thats Easy to Use
Some gaming consoles (like Microsofts Xbox 360) and
online games (like the Halo and Call of Duty series) have
incorporated systems so that players can report harassment
and offensive content, amongst other things like cheating.
However, it isnt always clear how in-game voice
harassment (which cant be as easily recorded as a text or
picture message) is monitored or policed. Furthermore,
most game players arent sure of the extent that these
reporting systems are monitored or enforced, as was
evidenced in the case study about the Halo 4 strict
moderation announcement. During the course of my
ethnographic inquiry, which included going to conferences
and conventions where developers and representatives
discussed these systems, it was difficult to determine how
they were being enforced, and very little details were
offered. Also, gamers lamented that sometimes their reports
went unanswered, or they themselves became victims of
enforcement at the hand of the person they tried to enforce
(for example, it appears that, in certain systems, multiple
complaints about a person can put them under review, so
this can possibly be exploited for victimization). Some cite
that female gamers often dont report because they are
discouraged by the lack of clarity in enforcement [30].
Moreover, some users who have sent extremely derogatory
messages on Xbox Live, for example, still have open
accounts [30]. Further, many of these reporting systems are
a several step process which makes it difficult to report
players while engaged in online gaming (i.e., opening
multiple slow screens with lack of easy accessibility to the
game player list requiring one to have to remember one
name out of dozens during live matches). Transparency is
an important part of effective and strict moderation, since
players should be aware of how they will be protected, and
violators should know that these systems (and the games
social environment) will not tolerate negative behavior.
Finally, ease of use is incredibly important in tandem with
transparency, so that players are not further discouraged to
report. One example of how this could be implemented is
Call of Duty Black Ops IIs system, see figure 3, which
allows you to click directly on a players name to report
with clearly marked and easy to select choices derived from
common complaints, resulting in a quick and easy process.
CONCLUSION AND FUTURE DIRECTIONS
This paper presents a set of design principles that can help
encourage equitable play and social experiences in play,
based on research that shows that play experiences are
disproportionately inequitable for females and ethnic
minorities. Future research is needed to understand how this
model ultimately supports play.
ACKNOWLEDGMENTS
I would like to thank my colleagues at the dolcelab and my
dissertation committee for their support. This work was
funded by the National Science Foundation (SES-1028637)
and a writing grant from AAUW. Any opinions, findings,
or recommendations expressed in this material are those of
the author and do not necessarily reflect their views.
REFERENCES
1. Subrahmanyam, K. and Greenfield, P.M. Effect of video
game practice on spatial skills in girls and boys. J ournal
of Applied Developmental Psychology, 15 (1994),13-32.
2. Okagaki, L. and Frensch, P. A. Effects of video game
playing on measures of spatial performance: Gender
effects in late adolescence. J ournal of Applied
Developmental Psychology, 15, 1 (1994), 33-58.
3. Richard, G.T. Gender and Game Play: Research and
Future Directions. In B. Bigl & S. Stoppe (Eds.):
Playing with Virtuality, Theories and Methods of
Computer Game Studies, Peter Lang, 2013, 269-284.
4. Bonanno, P. and Kommers, P. A. M. Exploring the
influence of gender and gaming competence on attitudes
towards using instructional games. British J ournal of
Educational Technology, 39, 1 (2007), 97-109.
5. Boyle, E. and Conolly, T. Games for learning: Does
gender make a difference. In Proceedings of the 2nd
European Conference on Games Based Learning,
(Barcelona, Spain, 2008), Academic Publishing, 69-75.
6. Hartmann, T. and Klimmt, C. Gender and computer
games: Exploring females' dislikes. J ournal of Computer-
Mediated Communication, 11, 4 (2006), 910-931.
7. Lucas, K. and Sherry, J . L. Sex differences in video
game play: A communication-based explanation.
Communication Research, 31, 5 (2004), 499-523.
8. Yee, N. (2008). Maps of digital desires: Exploring the
topography of gender and play in online games. In Y. B.
Kafai, C. Heeter, J . Denner & J .Y. Sun (Eds.) Beyond
Barbie and Mortal Kombat: Perspectives on Gender
and Gaming, Cambridge, MA: MIT Press, 83-96.
DESVIG 2013
87
9. Winn, J . and Heeter, C. Gaming, Gender, and Time:
Who makes time to play? Sex roles, 61, 1 (2009), 1-13.
10. J ensen, J. & de Castell, S. Girls@Play: An ethnographic
study of gender and digital gameplay. Feminist Media
Studies, 11, 2 (2011), 167-179.
11. Vermeulen, L., Loy, J.V., De Grove, F. & Courtois, C.
You Are What You Play? A Quantitative Study into
Game Design Preferences across Gender and their
Interaction with Gaming Habits. In Proceedings of the
5th International Digital Games Research Association
Conference (Utrecht, The Neatherlands, 2011), DiGRA.
12. Kiesler, S., Sproull, L. and Eccles, J .S. Pool halls, chips,
and war games: Women in the culture of computing.
Psychology of Women Quarterly, 9, 4 (1985), 451-462.
13. Bryce, J . and Rutter, J . Gender dynamics and the social
and spatial organization of computer gaming. Leisure
Studies, 22, 1 (2003), 1-15.
14. Feng, J ., Spence, I., and Pratt, J . Playing an action video
game reduces gender differences in spatial cognition.
Psychological Science, 18, 10 (2007), 850-855.
15. Sorby, S.A. Educational research in developing 3-D
spatial skills for engineering students. International
J ournal of Science Education, 31, 3 (2009), 45980.
16. Kafai, Y. B., Cook, M. S. and Fields, D. A. Blacks
Deserve Bodies Too!: Design and discussion about
diversity and race in a Tween virtual world. Games and
Culture, 5, 1 (2010), 43-63.
17. Nakamura, L. Don't hate the player, hate the game: The
racialization of labor in World of Warcraft. Critical
Studies in Media Communication, 26, 2 (2009),128-144.
18. DiSalvo, B. and Bruckman, A. Race and gender in play
practices: young African American males. In
Proceedings of the Fifth International Conference on
the Foundations of Digital Games (California, USA,
2010), ACM, 56-63.
19. Gray, K. L. Intersecting oppressions and online
communities. Information, Communication & Society,
15, 3 (2012), 411-428
20. Richard, G.T. Playing as a woman as a woman as if a
man. Well Played, 1, 3 (2012), 70-93.
21. Fron, J ., Fullerton, T., Morie, J ., & Pearce, C. The
hegemony of play. Proceedings of the 3
rd
International
Digital Games Research Association Conference:
Situated Play (Tokyo, J apan, 2007), DiGRA.
22. Miller, M.K. and Summers, A. Gender differences in
video game characters roles, appearances, and attire as
portrayed in video game magazines. Sex Roles, 57, 9/10
(2007), 733-742
23. Williams, D., Martins, N., Consalvo, M. and Ivory, J .D.
The virtual census: Representations of gender, race, and
age in video games. New Media and Society, 11, 5
(2009), 815-834.
24. Dill, K.E. and Burgess, M.C.R. Influence of Black
masculinity game exemplars on social judgment.
Simulation & Gaming, 2012, first published online at:
http://sag.sagepub.com/content/early/2012/07/24/10468
78112449958.abstract?rss=1
25. Lee, J . R. and Park, S. G. Whose Second Life Is This?
How Avatar-Based Racial Cues Shape Ethno-Racial
Minorities Perception of Virtual Worlds.
Cyberpsychology, Behavior & Social Networking, 14,
11 (2011), 637-642.
26. Behm-Morawitz, E. and Mastro, D. The Effects of the
Sexualization of Female Video Game Characters on
Gender Stereotyping and Female Self-Concept. Sex
Roles, 61 (2009), 808823.
27. Taylor, T.L. Multiple Pleasures Women and Online Gaming.
Convergence: The International J ournal of Research into
New Media Technologies, 9, 1 (2003), 21-46.
28. A. Sarkeesian, 2012. Image Based Harassment and
Visual Misogyny. Retrieved J uly 20, 2012 from
http://www.feministfrequency.com/
29. OLeary, A. In Virtual Play, Sex Harassment Is All Too
Real. The New York Times, (2012, August 1), A1.
http://www.nytimes.com/2012/08/02/us/sexual-
harassment-in-online-gaming-stirs-anger.html
30. Fletcher, J. Sexual harassment in the world of video
gaming. BBC News Magazine, (2012, J une 3).
http://www.bbc.co.uk/news/magazine-18280000
31. Kimmel, M. Guyland: The Perilous World Where Boys
Become Men. Harper Collins, NY, USA, 2008.
32. Bertozzi, E. You Play Like a Girl!: Cross-Gender
Competition and the Uneven Playing Field.
Convergence: The International J ournal of Research
into New Media Technologies, 14, 4 (2008), 473487.
33. Yin, R. K. Case Study Research: Design and Methods.
Sage, Thousand Oaks, CA, USA, 2009.
34. Gauntlett, A. Halo 4 Sexism Earns You a Lifetime Ban.
The Escapist (2012, Oct 31).
http://www.escapistmagazine.com/news/view/120412-
Halo-4-Sexism-Earns-You-a-Lifetime-Ban
35. Richard, G. T. and Hoadley, C. M. Investigating
Supportive Online Gaming Communities as a Means of
Reducing Stereotype Threat Vulnerability Across
Gender in Online Game Culture. Manuscript submitted
for publication, 2013.
36. Picho, K. and Stephens, J .M. Culture, context and
stereotype threat: A comparative analysis of young
Ugandan women in Coed and Single Sex schools. The
J ournal of Educational Research, 105, (2012), 52-63.

DESVIG 2013
88
Supporting social interaction for older users in game-like
3D virtual worlds
Panote Siriaraya
School of Engineering and
Digital Arts, University of Kent
CT2 7NT, Canterbury
spanote@gmail.com
Panayiotis Zaphiris
Department of Multimedia and
Graphic Arts
Cyprus University of
Technology
3036 Limassol
panayiotis.zaphiris@cut.ac.cy
Chee Siang Ang
School of Engineering and
Digital Arts, University of Kent
CT2 7NT, Canterbury
csa8@kent.ac.uk


ABSTRACT

There has been ongoing research into how 3D virtual
worlds can be designed to support everyday life activities
such as meeting friends, shopping or obtaining health
information. In this paper, we report a study with 38
older participants engaging with 3D and non-3D virtual
grocery stores in order to identify key factors which
affected their experience and satisfaction in social
engagement within the virtual environment. A mixed
method of questionnaire and contextual interview were
used for data collection and analysis. We present some
preliminary finding within the scope of this position paper,
focusing on various aspects of virtual world such as avatars
and non-verbal communication cues.
Author Keywords
3D Virtual worlds, older people, social interaction, design
ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g., HCI):
Miscellaneous.

General Terms
Human Factors; Design;
INTRODUCTION
There has been ongoing research into how 3D virtual
worlds can be designed to support everyday life activities
such as meeting friends, shopping or obtaining health
information. One key factor which greatly contributes to the
experience and satisfaction of such activities is the social
engagement as the result of the shared presence of other
users carrying out these activities virtually. Although there
have been a variety of research studies on this topic, most
of them focused on mainstream computer users who are
generally familiar with technology.

Research looking into the use of 3D virtual worlds for older
people has been relatively limited. Older people may have
different needs as well as physical and cognitive abilities
compared to their younger peers. This could lead to an
unfortunate missed opportunity, as such 3D technology
would potentially exclude the group of users who could
truly benefit from them.

Thus, in this study we aimed to investigate how 3D virtual
worlds and avatars affect communication and social
interaction for older users. A better understanding about
these factors would lead to a more inclusive design of 3D
virtual environments (One which is better able to
correspond to the needs and preferences of older users as
well) and promote better use of these tools for social
interaction. More specifically, the objectives of the study
are to investigate:

1. how certain factors of the virtual world affect the
quality of social interaction for older people. For
instance, factors such as the avatar and the 3D spatial
environment would be investigated.
2. how certain factors of older users affect their use of
virtual worlds for social interaction. For instance,
factors such as previous experience with computers and
older users perception towards computers would be
investigated.
3. how virtual worlds could be designed to better support
older peoples social interaction
Even though this study focuses on virtual worlds, we
believe the results would also be applicable to those looking
to improve sociability of computer games for older people
as well. Many online games, especially Massively
Multiplayer Online role-playing games share similar
elements as Virtual worlds [1]. For instance, the
embodiment of user interactions through avatars and the use
of a 3D spatial environment to provide an immersive
experience for users are common in both platforms.
Therefore, understanding how these elements effect the
social experience for older people in virtual worlds would
also be beneficial in improving the design of social

Copyright is held by the author/owner(s).
CHI13, April 27 May 2, 2013, Paris, France.

DESVIG 2013
89
interaction in online games for users of this age group as
well.

METHODS
We employed a mix of quantitative questionnaires, and
qualitative semi-structure interviews. 38 older people were
recruited to engage in a collaborative activity in a 3D
virtual world, in context of this study, a virtual grocery
store. We chose to focus on shopping activity because it
was an activity older people are familiar with in their daily
life [2] After having piloted the study, we found that
participants had difficulties evaluating and describing their
experiences in the 3D virtual store, as they had minimal
past experience in using the Internet (particularly online
shopping). In order to provide a context that framed the
discussion, a non-3D virtual store resembling a
conventional online shopping website was developed. In the
study, each pair of participants would visit both 3D and
non-3D virtual stores.
Identifying older people who are willing to use 3D
technology was challenging. For practical reasons,
participants were recruited by snowball sampling [3].
Initially, older people from local church groups and
volunteer organisations were contacted. Then, these
participants were asked to introduce other people who
might be interested in participating.
In this study, older people refer to those who are 55 years or
older. Participants in this study included those in the pre-
senior (55-64), Young old (65-74) and middle-old (75-84)
age cohorts (See [4] for a description). The average age of
the participants was 66.8 Years (SD=6.5) with the oldest
being 82 Years old. We believe that it was important to
recruit participants from the pre-senior age range as well as
such participants would be the main group likely to engage
with and benefit from this technology in the future, due to
the cohort effect of aging.
We recruited participants with all levels of computer and
gaming experience. Participants were given a shopping
voucher worth 10 for their participation. The study was
reviewed and approved by the Universitys research ethics
committee.


Figure 1: the 3D grocery store

Figure 2: the non-3D grocery store

A pair of older users participated in each session. First, a
brief was given to explain the purpose of the study. Then,
they logged on to the registration page containing the
virtual dressing room and selected an avatar to represent
themselves. Participants were given a choice of five avatars
(Older male, Older female, Younger male, Younger female
and the non human). These avatars were the same as those
used in the study by [5]. In the 3D store, the avatars were
represented using a 3D model (As can be seen in Figure 1).
In the 2D store, a static image was taken from the 3D model
and used as the avatar (As can be seen in Figure 2).
Each pair participated in both the 3D and non-3D store (see
Figure 1 and 2). After the participants have been given a
brief tutorial session lasting about 5-10 minutes, they were
asked to proceed to two separate locations. They were given
approximately 15-30 minutes to carry out two tasks (one in
3D and another in non-3D store). Both tasks required that
they completed a shopping activity with their partner.

A mixed method was used. First, questionnaires were
administered to examine the factors related to social
interaction experience of the participants. Semi-structured
interviews were later carried out to help provide context
and illustration to the questionnaire findings [6]. The
questionnaires used were adapted from the study by [5].

PRELIMINARY RESULTS
We planned to carry out regression analysis on the
questionnaire data to identify predictor factors affecting
social interaction experience. Within the scope of this
positional paper, we will present some preliminary finding
of the interview data.
Use of avatars
As shown in Table 1, the male participants overwhelmingly
preferred older male avatars. For female participants
however, the younger female character was the most
popular, followed by the older female avatar.

Table 1: The types of avatars selected in the study
DESVIG 2013
90
Avatar Gender
Male Female
Older Male 12 (70.58%) 1 (4.76%)
Older Female 0(0%) 5 (23.8%)
Younger Male 2 (11.76%) 2 (9.52%)
Younger Female 1 (5.88%) 9 (42.85%)
Non-Human 2 (11.76%) 4 (19.05%)

Participants were asked to describe the reason they chose
their avatar in the interviews. The most popular reason
given by male participants (n=9) was that they decided to
select avatars which best resembled their physical world
appearances. The female participants (n=5) who selected
the old female avatar also reported a similar reason for. One
reason given for this was that participants did not want the
avatars to misrepresent their real age and gender.
Well I didn't want to be a robot [referring to the non
human character], I didn't want to be male and much as
I'd love to be young and slim, I [amnot] so I just felt that...
you know, I suppose that I'm probably affected by the
stories you read of people online pretending to be
something they absolutely are not. (Participant 21,
Female, 66, selected the older female avatar)
Perhaps the most surprising reason of choosing a younger
avatar or one with a different gender was that some
participants did not select the avatar to represent
themselves, but to represent a family member or a shopping
companion (n=3).
I chose for someone who looked like my daughter and so I
choose my daughters name (Participant 29, Female, 67,
selected the young female avatar)
Use of avatars Difficulty in associating a visual avatar
with physical people
One commonly reported theme was the difficulty of older
users in associating the avatars in the virtual world with real
people. Participants (n=16) reported some kind of difficulty
in associating themselves or their partners with the avatars
when communicating in the 3D virtual world.
Well, to put it this way, he[the avatar] was a nice looking
chap, but it wasn't him, you know, I couldn't think of himas
being [Participant 2], he wasn't a man I was going
shopping with, so it didn't make it real in a way,[It seems I
was] shopping with a stranger (Participant 1, Female,
64).
Another reason why participants felt it was difficult to
associate the avatars with their partners seems to be due to
the visual appearance of the avatar not matching the true
appearance of their partner. This was mentioned by seven
participants.

It was odd to see [Participant 12] in a different guise. It's
a sort of disguise shall we say. But it was good, slightly
weird because she didn't look like [Participant 12], quite.
And you think "Who's this lady I'mtalking to" That does
come to the back of your mind (Participant 11, Male, 70).

A previous study related to older peoples perception of
avatars also reported a similar result. Cheong et al [7]
found that older people reported a low level of Homophily
when perceiving the avatars which suggested that they
could have difficulty relating and identifying themselves to
the avatars. Similarly, in this study, older people seemed to
find it difficult to not only embody themselves with the
avatars, but also found it difficult to perceive the avatars as
representatives of their partners as well.
Lack of non-verbal communication
One frequently mentioned limitation in regards to avatar-
mediated communication was the lack of non-verbal
communication. Participants felt that the inability of avatars
to display facial expressions and other forms of body
language (such as hand gestures) had a significant impact to
their social experience in this platform. Participants also
mentioned that they wanted to be able to control the avatars
to carry out non-verbal communications (such as to point to
products in the store) while they are communicating (n=17).
In a sense that the avatars were...more or less
grotesquely inhuman. I mean my character the man, one
could see that he was especially designed to be a man, but
you wouldn't necessary be able to pick up body language in
the way which he appear, and you certainly wouldn't picked
that fromthe one that my wife choose, who was more of the
monster figure [non human avatar]... Amusing and you've
got that geographical sense of being closer or further apart,
but that means of communication which is dependent on
body language was missing (Participant 2, Male, 62).

Indeed, the importance of non-verbal gestures in avatar
mediated communication has been reported in past studies
(see [8]). In addition, one participant reported that the
inability of avatars to display facial gestures made her feel
they were emotionless and made her perceive the avatars as
less human. Two participants even reported that their
inability to display emotions and their awkward movements
made them feel more like robots. This could be one reason
older people had difficulty associating avatars with actual
people. Other participants felt that the lack of such cues
limited the efficiency of their communication and made it
more difficult to trust their partners as they felt that users
could hide their emotions and body language during
communication.
Difficulty in suspending disbelief
Most of the participants reported a sense of artificiality
when engaging with the 3D store and felt that their
experience was not real enough. A variety of factors were
DESVIG 2013
91
frequently cited which reduced the participants sense of
realism in the 3D virtual store. Seven participants
mentioned the inability of computers to transmit other
senses such as touch and smell as a key disadvantage.
Other participants (n=9) cited factors such as a lack of other
actors in the stores such as other customers and store clerks
that made their experience feel less realistic.

I didn't feel like I was in the store at all... I didn't get the
feeling of going shopping.... Umm... Because there aren't
any real people there.... you know, in the supermarket, with
me, I tend to talk to people...but here it is only one-to-
one.... (Participant 26, Female, 62)
Other factors include the lack of details/information related
to the product (n=8) and the lack of variety or choices of the
product (n=17). These discrepancies between the virtual
store and the physical one have been reported also by [2].
Participants (n=14) often reported that they felt their
experience in the 3D store was more similar to a video
game experience. For participants, the perception of
controlling a character to move around a computer
generated virtual store was similar to the kind of games
they had played earlier or had seen their younger family
members play.
It's just like a game that my grandchildren play. Making
figures do things...hitting each other and shooting each
other or jump off something. It's the same really.
(Participant 32, Female, 66)

DISCUSSIONS AND CONCLUSIONS
The results highlighted a number of factors that influence
the experience of older users in virtual worlds. Physical
presence was found to be related to many of the measures
for the quality of social interaction. These results are in line
with previous studies. A study looking into social-oriented
virtual worlds found that presence (both physical and
social) affected user satisfaction [9], performance and
enjoyment (see [10]). The interviews showed that older
users felt their experience was artificial and thought it was
more similar to a game. This perception could have
negative implications if virtual worlds were to be used in
serious contexts as older people could hold a stereotype of
virtual worlds being just games and useful only for
entertainment purposes.
This study also suggested that older people found 3D
avatars to be of limited use in social interaction. It appeared
that older people were able to perceive the avatars as being
a representative, a puppet that they control, but do not
perceive them as a representation or an extension of
themselves in the virtual world [11]. Although further
studies would still need to be carried out to confirm why
this is the case, our study showed that older people found
the lack of non-verbal behaviour and facial expression to be
a key limiting factor of avatar. Not only did this reduced the
level of realism, some even felt that this made it difficult to
trust their partners as many of the non-verbal cues which
could serve as indicators for deception (such as facial
expression and body language) were not available [12].

Although some of these findings have been previously
reported in past studies on 3D virtual world, very few of
these past studies focused on older people. When older
people were involved, most past studies focused on only
specific aspects of virtual world. As for as we know, this is
among the first studies examining how older people used a
fully functioning virtual world for social interaction. This
study calls for more future research in this direction.
REFERENCES


1. Castronova, E. (2005). Synthetic worlds. Chicago: The
University of Chicago Press
2. Sjolinder, M., Hook, K., Nilsson, L.-G., & Andersson,
G. (2005). Age differences and the acquisition of
spatial knowledge in a three-dimensional environment:
Evaluating the use of an overview map as a navigation
aid. International J ournal of Human-Computer
Studies , 537-564.
3. Heiman, W. G. (2001). Understanding research
methods and statistics: An integrated introduction for
psychology Second Edition. Boston: Houghton Mifflin
Company.
4. Getzel, G. S., & Mellor, M. J . (1985). Understanding
normatve Growth and Development in Aging: Working
with Strengths. In Gerontological social work practice
in the community (pp. 37-54). Hew York: The
Haworth Press Inc.
5. Siriaraya, P., & Ang, C. S. (2012). Age differences in
the perception of social presence in the use of 3D
virtual world for social interaction. Special Issue on
presence and interaction, Interacting with Computers
24 , 280-291.
6. Bryman, A. (2012). Mixed methods research:
combining quantitative and qualitative research. In A.
Bryman, Social research methods: 4th edition (pp.
627-652). Oxford, UK: Oxford University Press.
7. Cheong, W. L., J ung, Y., & Theng, Y.-L. (2011).
Avatar: A virtual face for the elderly. VRCAI, (pp.
491-496). Hong Kong, China, December 11-12,2011.
8. Bente, G., & Kramer, N. C. (2011). Virtual gestures:
embodiment and nonverbal behavior in computer-
mediated communication. In A. Kappas, & N. C.
Krmer, Face-to-Face Communication over the
Internet: Emotions in a Web of Culture, Language, and
Technology (pp. 176-210). Cambridge University
Press.
9. J ung, Y. (2011). Understanding the Role of the Sense
of Presence and Percieved autonomy in User's
Continued Use of social virtual worlds. J ournal of
Computer-Mediated Communication , 492-510.
DESVIG 2013
92
10. Lombard, M., & Ditton, T. (1997). At the heart of it all:
The concept of presence. Journal of computer mediated
communication 3,(2) .
11. Bartle, R. A. (2004). Designing virtual worlds.
Indianapolis, IN: New Riders.
12. Zuckerman, M., & al, e. (1981). Verbal and nonverbal
communication of deception. In L. Berkowitz, ADV
experimental social psychology 14 (pp. 2-60). New
York: Academic Press.





DESVIG 2013
93
DESVIG 2013 Author Index
Author Index
Al Aamri, Fatma 1
Alha, Kati 56
Ang, Chee Siang 89
Balci, Koray 7
Christou, Georgios 12
Consalvo, Mia 17
Correia, Nuno 79
Cox, Anna L. 29
Das, Sauvik 20
Diephuis, Jeremiah 23
Geerts, David 12
Gold, Margaret 29
Good, Judith 52
Greuter, Stefan 1
Gross, Shad 62
Harrison, Chuck 20
Hochleitner, Christina 23
Hochleitner, Wolfgang 23
Jennett, Charlene 29
Kirman, Ben 33
Klauser, Matthias 39
Kloetzer, Laure 29
Korhonen, Hannu 56
Lankes, Michael 23
Law, Ee Lai-Chong 45
Martin, Eleanor 52
Mayra, Frans 56
Nacke, Lennart 39
Nagappan, Nachiappan 20
Paavilainen, Janne 56
Pace, Tyler 62
Peyton, Tamara 68
Phillips, Bruce 20
1
DESVIG 2013 Author Index
Prax, Patrick 73
Prescod, Paul 39
Reis, Soa 79
Richard, Gabriela 83
Salah, Albert Ali 7
Siriaraya, Panote 89
Springett, Mark 45
Stenros, Jaakko 56
Walz, Steen 1
Zaphiris, Panayiotis 89
Zimmermann, Thomas 20
2
DESVIG 2013 Keyword Index
Keyword Index
3D Virtual worlds 89
abusive behavior analysis 7
Activity theory 45
Add-on 73
Aect 45
Aective practice 45
Bayes Point Machines 7
Behaviour Change 39
Body 45
Citizen Science 29
co-creation 73
Co-located 79
co-location 23
code 73
Communities 29
Contagion 45
Cultures of Participation 62
cyberbullying 7
design 17, 89
Design 29
digital games 83
Digital games 45
Discourse 45
Emotion 45
equity 83
ethnicity 83
Evaluation 12
Facebook 56
Facebook Games 12, 39
family 17
friends 17
Game Analytics 39
game culture 83
Game design 52, 73
game design 33
Game Design 39
Game Metrics 39
1
DESVIG 2013 Keyword Index
games 23
Games 68, 79
Games User Research 39
Gamication 68
gaming concepts 7
gender 83
Health 68
Health Games 39
inclusion 83
Machinima 62
MMORPG 12
older people 89
Online Games 12, 39
online gaming 83
online multiplayer 20
platform 17
player motivations 17
player proling 7
player roles 23
procedural ethics 73
procedural rhetoric 73
public display 23
race 83
Reading for Pleasure 1
Reading motivation 1
Serious Games 29
Sociability 12, 29, 39, 56
sociability 23, 33
Sociable reading application 1
social conduct 7
Social games 33
social games 7
Social Games 12, 39, 56
social impact 73
Social interaction 52
social interaction 83, 89
Social interactions 79
Social Networks 56
social play 20, 33
sociality 17
Sociality 45, 68
2
DESVIG 2013 Keyword Index
strangers 17
Teamwork 52
tribal play 33
tribalism 33
User interface modication 73
Video Game 12
Video Games 56
video games 20
Video games 83
Workshop 68
World of Warcraft 62
Youth 68
3

























ISBN: 978-9963-2904-0-6