Exp Brain Res (1996) 110:62-66 RESEARCH ARTICLE

9 Springer-Verlag 1996

J a n e t L. T a y l o r . D.I.

McCloskey

Selection of motor responses on the basis of unperceived stimuli

Received: 30 January 1995 / Accepted: 17 January 1996

A b s t r a c t In a previous study, the sensory phenomenon of "backward masking" was used to demonstrate that subjects can preprogram a single stereotyped voluntary movement or movement-sequence and that such a movement can be triggered in response to a stimulus that is not perceived (that is, a stimulus of which the subject is unaware). In the present study, visual stimuli were presented at random in one of two different locations to normal human subjects in a choice reaction-time (RT) task. When the stimulus appeared in one of the locations, subjects made a motor response. When the stimulus appeared in the other location, subjects made a different motor response. Large and small stimuli were presented in either location. In some trials, the small stimulus was followed 50 ms later by the large stimulus. The small stimulus was then "masked" by the large stimulus and could not be perceived on forced-choice testing. Despite not perceiving the test stimulus in either of its randomly selected locations, subjects were able to select and execute the motor response appropriate for each location. The RTs for responses to the masked stimulus and to the same stimulus presented without masking (and so, easily perceived) were the same. This result implies that appropriate programs for two separate movements can be simultaneously held ready for use, and that either one can be executed when triggered by specific stimuli without subjective awareness of such stimuli and so without further voluntary elaboration in response to such awareness. Key words Motor program 9Triggered movement 9 Reaction time - Masking 9Human

Introduction
Although cogent physiological evidence for it is sparse, it is generally assumed that for many simple and cornJ.L. Taylor ( ~ ) 9D.I. McCloskey Prince of Wales Medical Research Institute, High Street, Randwick, N.S.W. 2031, Australia; Fax: +61 2 382 2722; e-mail: jl.taylor@unsw.edu.au

plex motor performances there exists, before the initiation of movement, an internal representation of the parameters of the movement. This representation is then used for the execution of that movement. The representation, or instruction set, or program, may have been acquired as a general skill months or years before, or may have been developed and learnt for the particular motor performance. In a recent study (Taylor and McCloskey 1990), we showed that intended simple and more complex movements could be triggered in normal subjects in response to sensory stimuli of which the subjects were unaware. This implied that an appropriate, preselected motor program could be held ready for execution and then triggered in an automatic way, without further voluntary or cognitive intervention. In the present study, we show that one rather than another motor response can be triggered by one rather than another sensory stimulus of which the subject is unaware. In terms of a simple model of reaction-time processes in which stimulus detection is followed by response selection and then response execution, the findings implies that both selection and triggering of the response can occur in an automatic way. As an experimental tool in these studies we have used "masking" to deliver sensory stimuli of which the experimental subjects are not aware. Masking is a well known sensory phenomenon in which the close spatial and temporal proximity of two stimuli of unequal intensities results in the smaller stimulus being more difficult to detect than if it had been presented alone. In visual backward masking, a dim test light is presented shortly before a bright masking light so that the test light becomes more difficult to detect. If the difference in stimulus intensities and the interval between the stimuli are chosen appropriately, then the test light may not be perceived at all (Alpern 1953; Breitmeyer 1984; Raab 1963; Schiller and Smith 1966). However, several studies have shown that in a simple reaction-time task, despite being unable to detect the test light, subjects are able to make a voluntary movement in response to it (Fehrer and Biederman 1962; Fehrer and Raab 1962; Harrison and Fox 1966; Schiller and Smith 1966).

63 In our p r e v i o u s study (Taylor and M c C l o s k e y 1990), w e c o n f i r m e d the ability o f subjects to m a k e s i m p l e and m o r e c o m p l e x v o l u n t a r y m o v e m e n t s in r e s p o n s e to a m a s k e d visual stimulus and, with f o r c e d - c h o i c e testing, s h o w e d that subjects c o u l d not p e r c e i v e the stimulus. Because subjects did not p e r c e i v e the triggering stimulus, the p a r a m e t e r s o f the m o v e m e n t r e s p o n s e c o u l d not have b e e n d e t e r m i n e d at, or after, r e c e i p t o f the stimulus at the time w h e n the e x e c u t i o n o f the m o v e m e n t was initiated, but m u s t have b e e n a l r e a d y d e t e r m i n e d . That e x p e r i m e n t gave e v i d e n c e that a m o t o r p r o g r a m m u s t have existed, and s h o w e d that such a p r o g r a m , once preselected, c o u l d then be released, or triggered, in an a u t o m a t i c way. F u r t h e r e v i d e n c e that supports the p r o p o s a l that single m o t o r r e s p o n s e s can be p r e p r o g r a m m e d c o m e s f r o m a study b y D a y et al. (1989). If a cortical shock is given during a v o l u n t a r y r e a c t i o n time, the m o v e m e n t is del a y e d but is then p e r f o r m e d s u c c e s s f u l l y after the delay. This i m p l i e s that the m o t o r c o m m a n d for the m o v e m e n t was f o r m u l a t e d as a w h o l e and c o u l d be stored for at least the t i m e o f the delay. Examination o f cortical potentials and reflex responses to muscle stretch during the period before a reaction-time response also indicate that, if the parameters of the response are fully known 1.5 s before the stimulus is presented, then the response can be p r e p r o g r a m m e d and is ready to be triggered by the stimulus ( M a c K a y and Bonnet 1990). The p r e s e n t study extends these studies. It e x a m i n e s the ability o f subjects to h o l d two m o t o r r e s p o n s e s in r e a d i n e s s at the one time and then to select b e t w e e n t h e m and p e r f o r m t h e m individually, contingent on specific triggers. S u b j e c t s d e m o n s t r a t e an ability to select a m o tor r e s p o n s e on the basis o f the l o c a t i o n o f a m a s k e d , unp e r c e i v e d stimulus.
LED display 1.5 ~ visual arc
A l l

fixation point B

test light alone

B

A simultaneous test and mask B A

B

--11,

# I , I t // # I
50 100 150

separated test and mask

__n
__1 0 start

//
500

msecs

timing

Fig. 1A, B Two arrays of five ultra-bright light-emitting diodes (LEDs) were set to each side of a pin-point red fixation light. Three different stimuli could be presented either to the left or to the right fo the fixation point. Test light alone: a dim, 5-ms flash of the central LED (A). Simultaneous test and mask: test light with a simultaneous-onset, bright, 50-ms flash of the four surrounding LEDs (B). Separated test and mask: test light followed at 50 ms by the masking light (bright, 50-ms flash of the four surrounding LEDs). Timing of all responses started at time zero, as shown. Response time to simultaneous test and mask therefore includes a 50-ms delay that makes it directly comparable with any reactions to the masking light in the combined stimulus Efficacy of masking

Materials and methods

Apparatus and stimuli Subjects were seated in darkness 3 m from a visual stimulus display. The stimulus display consisted of a red, pinpoint, light-emitting diode (LED) and two groups of five, ultrabright, yellow LEDs of 7 mm diameter (see Fig. 1). These were positioned 1.5 ~ of visual arc to the left and to the right of the red LED, which was used as a fixation point and remained on at all times. A circle of diameter 30 cm framed the display and was lit very dimly. Subjects were required to make reaction-time responses to three different stimuli. These stimuli could be presented either to the left or to the right of the fixation light by using one of the two sets of five LEDs (see Fig. 1). Each set consisted of a middle and four surrounding LEDs:
1. Test light - a dim 5-ms flash of the middle LED at time zero followed at 500 ms by a bright, 50-ms flash of the four surrounding LEDs. The 500-ms delay was long in comparison with reaction time so this combination tested reaction times to the test light. 2. Simultaneous test and masking lights - a dim, 5-ms flash of the middle LED together with a bright, 50-ms flash of the four surrounding LEDs, all with a simultaneous onset 50 ms after the beginning of timing. 3. Separated test and masking lights - the dim, test light at time zero followed, after a 50-ms delay, by the bright flash of the surrounding LEDs.

Each subject performed a forced-choice test, attempting to differentiate between the simultaneous test and masking lights and the separated test and masking lights. Subjects initially were presented several times with the two stimuli, identified each time as light 1 or light 2. The lights were then presented randomly and the subject asked to identify them. One set of 20 forced choices was made before any reaction-time trials were performed and one set after the reaction-time trials were complete. For one set of differentiations, the stimuli were presented to the left of the fixation point and, for the other, to the right. Any subject who identified correctly more than 60% of the trials was discarded from the study. Of the 18 subjects who took part in the experiments, 2 were discarded on this basis.

Experimental design Twelve subjects who were unable to discriminate between the simultaneous and the separated test and masking lights took part in experiment 1. After performing a set of forced-choice differentiation trials, they were asked to perform a choice reaction task. For this task, a stimulus was presented either to the right or to the left of the central fixation point. In response to a light appearing on the left, subjects flexed the left elbow and, in response to a light on the right side, they flexed the right elbow. Trials were performed

64 every 15-20 s, with the stimulus light appearing 1-3 s after a warning tone. Reaction times were measured to the commencement of biceps electromyographic activity (EMG), which was recorded through surface electrodes. Three groups of 24 trials were performed. For two of these sets, either the simultaneous-lights stimulus or the separated-lights stimulus was presented in a random order. In the other set, the stimulus was the small test light alone. Reaction times of less than 100 ms or more then 450 ms were discarded, as were trials in which errors were made (wrong arm, or both arms, moved). These trials were repeated at the end of the group. For analysis, the largest and smallest reaction times for each stimulus (e.g. left, combined stimulus) were discarded and means and standard deviations found for the remaining 10 trials. Following completion of the reaction-time trials, the second set of forced-choice differentiations were performed. In a second experiment, subjects were asked to make different movements with the left hand and right arm in response to the leftand right-side stimuli. Six subjects, two of whom had also taken part in experiment 1, performed the experiment. In response to the appearance of a right-side light, they again flexed the right elbow. However, in response to the left light, subjects were instructed to open the left hand. Reaction times on the right were measured to the commencement of biceps EMG and on the left to the commencement of EMG recorded from surface electrodes placed over the extensor digitorum. In all other details experiments 1 and 2 were identical. m s f r o m the onset o f t i m i n g (see M a t e r i a l s and m e t h ods). The m e a n reaction times o f the 12 subjects w h o m a d e the s a m e m o v e m e n t , f l e x i n g either arm, are shown in Fig. 2. A n analysis o f variance ( A N O V A ) on the g r o u p e d data, f o l l o w e d b y a S t u d e n t - N e w m a n - K e u l s m u l t i p l e c o m p a r i s o n , s h o w e d no significant difference b e t w e e n reaction times to test alone (239.7-+38.7 ms) and to the s e p a r a t e d test and m a s k stimulus (246.5_+38.1 ms), but s h o w e d that r e a c t i o n times to s i m u l t a n e o u s test and m a s k (284.3_+38.7 ms, w h i c h i n c l u d e the 50-ms delay; see Fig. 1) were significantly l o n g e r than reaction times to the s e p a r a t e d c o m b i n a t i o n (P<0.01). That is, subjects reacted to the test light, although it was c o m p l e t e l y masked. T h e r e was no difference b e t w e e n the left and right arm responses. F i g u r e 2 also shows the m e a n results for the six subjects w h o m a d e different m o v e m e n t s with the two sides o f the b o d y in r e s p o n s e to the l a t e r a l i z e d stimuli. A g a i n an A N O V A and m u l t i p l e c o m p a r i s o n shows that subjects r e a c t e d to the s e p a r a t e d - l i g h t s stimulus in times (247.4_+19.5) c o m p a r a b l e with their r e a c t i o n times to the test light alone (250.4+_22.1), and that these times were shorter than their r e a c t i o n times to the s i m u l t a n e o u s test and m a s k with an a d d i t i o n a l 50-ms d e l a y (284.2+19.7; P<0.01). T h e left h a n d and right arm r e s p o n s e times were not different.

Results
If subjects react to the test light in the s e p a r a t e d - l i g h t s stimulus they should r e a c t in a t i m e s i m i l a r to their res p o n s e time to the test light alone. I f subjects react to the m a s k i n g light in this c o m b i n a t i o n they s h o u l d res p o n d in a r e a c t i o n t i m e s i m i l a r to their r e a c t i o n t i m e to the s i m u l t a n e o u s - l i g h t s s t i m u l u s plus an a d d i t i o n a l 50 ms, w h i c h c o r r e s p o n d s to the d e l a y e d onset o f the b r i g h t light as the s e c o n d part o f the s e p a r a t e d c o m b i n a t i o n . A s an e x p e r i m e n t a l convention, to a l l o w direct c o m p a r i son b e t w e e n r e a c t i o n t i m e to the s e p a r a t e d stimulus and r e a c t i o n t i m e s to e a c h o f the c o n t r o l stimuli (test a l o n e and s i m u l t a n e o u s stimulus), w e p r e s e n t e d the s i m u l t a n e o u s stimulus, not at t i m e zero, but with a d e l a y o f 50

Discussion
Subjects d e m o n s t r a t e d an ability to select and execute m o t o r r e s p o n s e s on the basis o f the l o c a t i o n o f stimuli that they d i d not perceive. In each e x p e r i m e n t , both o f the p o s s i b l e m o t o r responses were m a d e in r e a c t i o n to the small test light w h e n it a p p e a r e d in the s e p a r a t e d test and m a s k stimulus, d e s p i t e s u b j e c t s ' i n a b i l i t y to detect the p r e s e n c e o f this light on f o r c e d - c h o i c e testing. F o r each response, subjects' r e a c t i o n times to the s e p a r a t e d B
T

Fig. 2 A Mean (+SD) reaction times for experiment 1, in which subjects bent the left (empty circles) or the right ~lled circles) arm in response to left- or rightsided stimuli. Reaction times to the separated test and mask stimulus are the same as to the test light but are shorter than to the simultaneous stimulus when it was delayed by 50 ms. The right and left arm responses are not different. The small and large boxes at the bottom symbolise the component lights of each stimulus. B Mean (-+SD) reaction times for experiment 2, in which subjects opened the left hand (boxes) or bent the right arm (circles) in response to left- or right-sided stimuli. As in experiment 1, reaction times to the separated stimulus match those to the test light

0 left arm 9 (ight arm 300

3oo
9 right arm 250 200

[] left hand

reaction time (ms)

250 200

masking light ~ test light --~

50 0
CE3

50
0 separated simultaneous test separated simultaneous

test

65 lights stimulus were too short to be in response to the large masking light, but were comparable with reaction times to the test light by itself. The response movements involved in these experiments were made because the subjects decided prior to presentation of the stimuli to perform them: to move the left side in a particular way to a stimulus on the left, or the right side in another particular way to a stimulus presented on the right. The subjects could have decided at any time (until presumably very close to the presentation of the stimulus) not to execute the movements on presentation of the stimuli. On this basis (to do, or not to do, according to an intention that was preformed and was able to be stated), the movements were voluntary movements. However, although these response movements were voluntary, they could not have been formulated in response to the presentation of the stimulus. Subjects were unaware of the stimulus to which they reacted and, thus, could make no conscious or volitional decisions on movement parameters at the time of its presentation. An internal description of the response movement must already have existed in such a form that no further volition was required for execution of the movement. That is, there was a preformulated motor program. This confirms the findings and analysis of our earlier study (Taylor and McCloskey 1990). For the subject to be able to select the correct response, both of the possible movements must have existed simultaneously in this form, with the execution of each contingent on its specific trigger. There were two preformulated motor programs existing simultaneously, and the appropriate selection was made without further voluntary intervention. Stimulus location, in contrast to other stimulus features, may influence response selection in reaction-time tasks even when it is irrelevant to the choice of response (Simon and Rudell 1967). It has been suggested that location may be processed automatically (Kornblum et al. 1990; cf. Hasbroucq and Guiard 1991), and that when stimulus-response compatibility is high, in tasks such as pointing or reaching to the location of a stimulus, the entire response may occur without volition. In our study, stimulus-response compatibility was high in that a stimulus required a response from the same side of the body. However, subjects in our study were equally able to bend the arm or open the hand in response to a left-sided stimulus. Thus, it is unlikely that bending the arm or opening the hand existed as an automatic response to the stimulus before subjects received the appropriate instruction. Rather, the responses were specific voluntary movements that were formulated and associated with the stimulus only because subjects were so instructed. Although subjects were unable to determine the presence or absence of the small test stimulus on forcedchoice testing, it is possible that this may not indicate a failure to perceive the small stimulus when it is masked, but rather may mean that subjects perceive the separated test and mask stimulus as blurred into one light which cannot be differentiated from the simultaneous test and mask. However, a previously reported control experiment (Taylor and McCtoskey 1990) shows that subjects who can differentiate between a long, 100-ms light and a short, 50-ms light identify the separated test-and-mask stimulus as a short (50 ms) light. They do not perceive it as commencing at the start of the test light and continuing to the finish of the masking light. Furthermore, the concordance of reaction times to the separated test-andmask stimulus with those to the test stimulus alone shows that subjects are reacting to only the small light of the stimulus and not to some larger, brighter perceived combination light. It is well known that reaction times decrease with increased intensity of the stimulus and such a reduction does not occur here. Increasing the complexity of movements has been shown to increase choice reaction times (Rosenbaum 1980; Sheridan 1981). Some studies have found that these increases have not been mirrored by increases in simple reaction times (where subjects have no choice of response; Ktapp et al. 1974, 1981). This has been used as evidence in support of the preprogramming of these simple reaction-time responses. The increases in choice reaction times are thought to reflect the longer motor processing time necessitated by a more complex movement (Klapp 1981). In these studies preprogramming of a movement is used to signify that all motor processing up to execution of the movement is complete before the stimulus is presented. Therefore, the start of execution is not delayed by the additional processing of a more complex response. In our experiments, we have used preformulation of a motor program to mean that the movement can be executed without further volition. This does not imply that all motor processing has been completed. Taylor and McCloskey (1990) showed that when complex motor responses were triggered without volition in a simple reaction-time task, reaction time was longer than for less complex responses. However, although motor processing continued during reaction time, all the parameters of the movement must have been defined before that time so that both simple and more complex movements could be considered to be preformulated. Similarly, in the present study subjects demonstrated an ability to select and perform two distinct movements in response to two specific stimuli, neither of which could be consciously perceived. Each of the response movements was preformulated and could be individually triggered without further volition. Thus, two movements were held in readiness simultaneously. For experiment 1, in which the same reaction-time movement was made with each arm, it could be argued that only a single motor program was prepared and that the masked stimulus triggered selection of the appropriate arm. However, in experiment 2, although the motor responses remained simple, different muscle groups and different actions were required for each movement. Subjects continued to react to the unperceived stimulus, preformulating two response movements and then selecting and executing one of them. The further differences between motor responses did not cause increases in reaction time for either response.

66

In these experiments, we have demonstrated that subjects can preformulate and, without volition, select motor responses. However, subjects here chose between only two alternative responses and these required movements of different sides of the body. The findings do not necessarily apply to pairs of responses that differ in other parameters. F u r t h e r m o r e , they give no i n d i c a t i o n as to
w h e t h e r subjects m i g h t be able to p r e f o r m u l a t e and h o l d in readiness m o r e than two m o v e m e n t s . Acknowledgements This work was supported by the National Health and Medical Research Council of Australia.

References
Alpern M (1953) Metacontrast. J Opt Soc Am 43:648-657 Breitmeyer BG (1984) Visual masking: an integrative approach. Oxford University Press, New York Day BL, Rothwell JC, Thompson PD, Maertens-de-Noordhout A, Nakashima K, Shannon K, Marsden CD (1989) Delay in the execution of voluntary movement by electrical or magnetic brain stimulation in intact man. Brain 112:649-663 Fehrer E, Biederman I (1962) A comparison of reaction time and verbal report in the detection of masked stimuli. J Exp Psychol 64:126-130 Fehrer E, Raab D (1962) Reaction time to stimuli masked by metacontrast. J Exp Psychol 63:143-147 Harrison K, Fox R (1966) Replication of reaction time to stimuli masked by metacontrast. J Exp Psychol 71:162-163

Hasbroucq T, Guiard Y (1991) Stimulus-response compatibility and the Simon effect: toward a conceptual clarification. J Exp Psychol Hum Percept Perform 17:246-266 Klapp ST (1981) Motor programming is not the only process which can influence RT: some thoughts on the Marteniuk and MacKenzie analysis. J Mot Behav 13:320-328 Klapp ST, Wyatt ER Lingo WM (1974) Response programming in simple and choice reactions. J Mot Behav 6:263-271 Klapp ST, Greim DM, Marshburn EA (1981) Buffer storage of programmed articulation and articulatory loop: two names for the same mechanism or two distinct components of short-term memory? In: Lang J, Baddeley A (eds) Attention and performance IX. Erlbaum, Hillsdale, NJ Kornblum S, Hasbroucq T, Osman A (1990) Dimensional overlap: cognitive basis for stimulus-response compatibility - a model and taxonomy. Psychol Rev 97:253-270 MacKay WA, Bonnet M (1990) CNV, stretch reflex and reaction time correlates of preparation for movement direction and force. Electroencephalogr Clin Neurophysiol 76:47-62 Raab DH (1963) Backward masking. Psychol Bull 60:118-129 Rosenbaum DA (1980) Human movement initiation: specification of arm, direction and extent. J Exp Psychol Gen 109:444474 Schiller PH, Smith MC (1966) Detection in metacontrast. J Exp Psychol 71:32-39 Sheridan MR (1981) Response programming and reaction time. J Mot Behav 13:161-176 Simon JR, Rudell AP (1967) Auditory S-R compatibility: the effect of an irrelevant directional cue on information processing. J Appl Psychol 51:300-304 Taylor JL, McCloskey DI (1990) Triggering of preprogrammed movements as reactions to masked stimuli. J Neurophysiol 63:439-445