Neuropsychologm, Vol. 26, No. 3, pp. 435444, Printed in Great Bntain.
1988.
W28-3932188 $3.00+0.00 Pergamon Press plc
ATTENTIONAL FACTORS IN THE OCCURRENCE OF STIMULUS-RESPONSE COMPATIBILITY EFFECTS* MIEKE VERFAELLIE,? DAWN BOWERS and KENNETH M. HEILMAN~ Department of Neurology, College of Medicine, University of Florida, and the Veterans Administration Medical Center, Gainesville, FL 32610, U.S.A.; and tMemory Disorders Research Center, Boston Veterans Administration
Medical
Center,
and Boston
University
School of Medicine
(Received 7 October 1986; accepted 24 September
1987)
Abstract-Two major hypotheses have been advanced to account for stimulusPresponse compatibility effects in the situation in which the location of the target is irrelevant for choosing the correct response. According to the attentional hypothesis, compatibility effects reflect a response bias, favoring the effector on the same side as the stimulus. According to the coding hypothesis, compatibility effects result from a correspondence between the spatial codes of the stimulus and effector. In the present study, two components of attention-selective attention and intention-were independently manipulated by providing selective preparatory information before onset of a target stimulus. Attentional information indicated where the target stimulus would occur; intentional information indicated which hand would have to be used to respond. Compatibility effects were observed only in the condition in which intentional information, but no attentional information, was provided. These findings support the attentional hypothesis and indicate that a specific aspect of attention, namely a selective readiness to respond to the stimulus, is a necessary condition for compatibility effects to occur.
INTRODUCTION THE CONCEPT of stimulus-response
(S-R) compatibility, as introduced by FITTS and SEECER [S], refers to the fact that the time needed to respond to a certain stimulus depends not only on the characteristics of the stimulus and the characteristics of the response but also on the relation between the two. The more efficient S-R combinations are termed compatible and the less efficient ones, incompatible. Although compatibility effects occur in a variety of situations, spatial compatibility has undoubtedly been studied most often. Here, the factor determining the speed of response is the relation between the spatial arrangement of stimulus and response. For example, when a flash of light is presented to the left or right of a central fixation point, and manual responses are made to the left and right of the body midline, responses are reliably faster when the task requires that the left key be pressed in response to the left light and the right key be pressed in response to the right light, as opposed to the opposite pairings [ 1, 61. A similar phenomenon occurs when the location of the stimulus is in itself irrelevant to the task. For example, the *This work was supported in part by the Medical Research Service of the Veterans Administration. SCorrespondence to be addressed to Dr Heilman, Department of Neurology, Box J-236. College of Medicine, University of Florida, Gainesville, FL 32610, U.S.A. 435
436
MIEK-LVEKPAELLIE, DAWNBowr~s
and KENNETH M. HEILMAN
color of a stimulus presented on the left or the right side of a display may indicate whether a left or right hand response is required [ 18, 21,241. Typically, responses are faster to the color indicating a left hand response when the stimulus occurs on the left than when it occurs on the right, and faster to the color indicating a right hand response when the stimulus occurs on the right than when it occurs on the left. Although this phenomenon, for which HEDGEand MARSH [9] introduced the term “Simon effect”, can be thought of as a variation of spatial compatibility, a distinction between these two types of spatial compatibility is important, because different mechanisms might underlie their occurrence [ 15,223. Traditionally, spatial compatibility effects have been explained by means of an anatomic pathway transmission model: the reaction time is longer when the hand opposite the stimulated field responds, because in this situation information must be transmitted between the hemispheres and hence information processing takes more time than when information can remain in the same hemisphere. Several investigators [2,24], however, have ruled out a purely anatomical explanation by having subjects cross their hands, so that the left hand presses the right key and the right hand presses the left key. When the hands are crossed in a choice reaction time task, faster responses are made by the hand positioned in the same side of space as the stimulus, even though this information must cross the corpus callosum. Several hypotheses have been advanced to account for the faster reaction times in the more compatible S-R pairings. SIMON et al. [18, 201 suggested that there is a natural tendency to respond toward the source of stimulation. This tendency is conceived of as a stereotype, comparable to the orienting reaction. When the position of the stimulus does not match the position of the response, responses are relatively slow, because an inappropriate tendency to respond toward the source of the stimulus must be inhibited. Conversely, when the position of the stimulus and the position of the response correspond, the tendency to respond toward the source of stimulation is consistent with that commanded by the stimulus, hence, the shorter reaction times. Alternatively, WALLACE [24,25] proposed an explanation in terms of the correspondence between the coding of the stimulus position and the coding of the position of the effector. If the stimulus and response share the same spatial code, responses will be faster than when the positions are coded in different ways. The assumption here is that when a stimulus is presented, a spatial code is formed and compared with the spatial code of the response. When both are compatible, the time for S-R translation is less than when they are incompatible. Finally, a third interpretation has been offered by HEILMAN et al. [ 10, 111. They proposed that compatibility effects may reflect an attentional bias favoring the response on the same side as the stimulus. According to these authors, each hemisphere is responsible not only for receiving stimuli in contralateral space and for controlling the contralateral limbs, but also for mediating selective attention and intention (i.e., readiness to respond) in contralateral hemispace, independent of which hand is used. When a stimulus is presented in one side of space, this activates the contralateral hemisphere and directs attention to the corresponding hemispace. Because the activated hemisphere also mediates responses in contralateral hemispace, the hand located in that hemispace will be favored, irrespective of whether it is controlled by the same or opposite hemisphere. Both the hypotheses advanced by SIMON [18] and HEILMAN and VALENSTEIN [lo] are attentional in nature, in that they assume that a lateralized stimulus creates an asymmetry in one or more aspects of attention, i.e. selective attention or response readiness, or both. Whereas Simon’s hypothesis does not address the nature of this asymmetry, HEILMAN and VALENSTEIN[lo] propose an explanation in terms of underlying hemispheric organization.
ATTENTIONAL
FACTORS
IN THE OCCURRENCE
OF STIMULUS-RESPONSE
COMPATIBILITY
EFFECTS
437
Each hemispace represents a perceptual field in which stimuli are attended to and processed by the contralateral hemisphere as well as a behavioral field in which stimuli are selectively acted upon by the contralateral hemisphere. If stimulus processing and response preparation are mediated by the same hemisphere, responses are faster than when they are mediated by a different hemisphere. The explanation proposed by WALLACE[24,25], on the other hand, can be referred to as a coding hypothesis [12, 151, because S-R compatibility effects reflect the results of a comparison between the codes for the stimuli and the responses. Recently, NICOLETTIet al. have provided support for the coding hypothesis, both in situations in which the position ofthe stimulus is the relevant cue for the subsequent response [15] and in situations in which the position ofthe stimulus is irrelevant [22]. In these studies, the two stimuli and the two responses were presented on the same side of the body midline, thus eliminating differential hemispatial orientation. Contrary to predictions of the attentional theory, compatibility effects were still observed. Because for both stimuli and response keys one member of the set could be classified as left and the other as right, these findings were interpreted as support for the spatial coding hypothesis. In the present report, we describe the effects of attentional manipulations on the occurrence of SR compatibility effects in a situation in which the location of the stimulus is irrelevant to the task (i.e. the Simon effect). This study evolved from a series of experiments designed to study two components of attention, referred to as selective attention and intention [4,23]. Selective attention refers to the mechanisms by which one decides what to process and how far to process the information. Intention is defined in terms of a selective readiness to respond or preparation for action and refers to the mechanisms by which one decides whether or not to act upon a stimulus. The effects ofselective attention and intention were studied in a choice reaction time task in which the presence of attentional and intentional cues was manipulated independently. Attentional cues consisted of telling the subjects where in space the target light would occur; intentional cues consisted of telling the subjects which hand they would have to use to respond. Subjects were required to respond as quickly as possible, with the left or right hand, to one of two lateralized lights, one presented in left hemispace and the other one presented in right hemispace. The left hand was always positioned on the left side of the body and the right hand on the right side. The stimulus property relevant for choosing the response was the brightness of the target light. Thus, a dim light indicated a left (right) hand response and a bright light indicated a right (left) hand response, irrespective of whether the target stimulus was presented in left or right hemispace. We found that both attentional and intentional cues reduce the latency of responses and that the right hemisphere is dominant for intention, but not for selective attention. These findings are presented in greater detail elsewhere [23]. This study also has important implications for explanations of the Simon effect. As previously described, subjects responded with the left or right hand to the brightness of stimuli presented to the left or right of midline. In this situation, both stimulus and response positions can be coded according to the same left/right code. According to the spatial coding hypothesis, responses are faster when the stimulus and response share the same spatial code than when they are coded in different ways. If this hypothesis is correct, compatibility effects should be obtained in the present task independent of the nature of the attentional manipulations, since the spatial codes remain unaltered by the attentional manipulations. Alternatively, the attentional hypothesis states that compatibility effects reflect an attentional bias favoring responses mediated by the same hemisphere that processes the stimulus. According to this hypothesis, then, one would predict that in the present study,
43x
MIEKE VERFAELLIE,DAWN BOWERS and KENNETH M. HEILMAN
compatibility effects may vary as a function of the attentional and intentional cues, because these may alter the subject’s implicit information processing and response strategies as well as the underlying pattern of hemispheric activation.
METHODS Twenty-four male college students (aged 18 32 yr) from introductory were right-handed according to self-report. Apparutus
psychology
courses served as subjects.
All
and stimuli
Subjects were given a choice reaction time task in which they manually responded to a lateralized target presented to the left or right of fixation. The fixation point consisted of a white dot with a 0.4 cm dia, which was visible at all times. Before onset of the target stimulus, two midline warning stimuli were presented that might indicate where in space the target would occur (attentional warning) and which hand the subjects would have to use for responding (intentional warning). Wurninq stirnuli.The warning stimuli that provided the attentional and intentional cues consisted of two sets of three vertically aligned lights (luminance=0.30 mscp) with white lenses (see Fig. 1). One set was positioned 2.4,4.8 and 7.2’ above the fixation point, and the other set at similar angles below the fixation point. One set conveyed attentional (or no) information, and the other set conveyed intentional (or no) information, with the position of the attentional and intentional warning stimulus set (above or below the fixation point) counterbalanced across subjects.
0
El cl El
0
0
IDI q El FIG. 1. Schematic
representation
of the stimulus
display.
Superimposed on the top and bottom light of the attrntiorml warnin~g stimulus set was an arrow pointing to the left or right. indicating where in space the target would occur. Thus, illumination of a warning light with an arrow pointing to the left indicated that the target light would occur, on the majority of trials, to the left of fixation The opposite held true for the arrow pointing to the right. The middle light was a neutral light. indicating that the next trial would occur, but giving no selective attentional information. These warning stimuli were defined as the attentional warning stimulus set because the information they provided. if any, enabled subjects to shift their attention correctly to the probable spatial location of the subsequent target stimulus. Superimposed on the top and bottom light of the inrmtioncrl warniny stimulus set was an index finger pointing to
ATTEKTIONAL
FACTOKS
IN ‘IHE WXL’KKENCE
OF STIMULUS-KESPONSk
TOMPATIBILITY
EFFECTS
439
the left or right. Illumination of the index finger pointing to the left indicated the target stimulus that followed would require a left hand response most of the time. The opposite held true for the finger pointing to the right. Again, the middle light was a neutral light indicating that the next trial would occur, but giving no selective preparatory information. These warning stimuli were defined as the intentional warning stimulus set because the information they provided, if any, concerned the hand to be used for responding to the target stimulus. Target stimuli.The target stimuli consisted of two incandescent lights with white lenses (luminance=0.30 mscp) located 7” to the left and right ofthe fixation point. Each light could be either bright or dim. A dim light was obtained by adding 430 R resistance. The response board consisted of two horizontally aligned microswitches mounted 16 cm to the left and right of body midline. The right index finger was positioned on the right microswitch, and the left index finger was positioned on the left microswitch. Interval timing was controlled by digital logic (BRS/LVE, Laurel, MD), and response times were measured by a digital timer (Gerbrands G1270, Gerbrands Corp., Arlington, MA) that was activated by onset of the target stimulus and stopped by the manual release of the appropriate response key. Procedure Each trial began with the simultaneous presentation of two central warning stimuli, one of each set, for 500 msec. After variable interstimulus intervals (2500,2800,3100, and 3400 msec), one of the target stimuli was presented for 500 msec as well. There were four warning conditions, consisting of every possible combination of attentional and intentional information being either present or absent: (I) both attentional and intentional information was given; (2) only attentional information was given, together with the neutral light of the intentional set; (3) only intentional information was given, together with the neutral light of the attentional set; and (4) no selective information was given, but two neutral warning stimuli were presented. To offset anticipatory responses, 20% of the trials were invalid. On invalid trials, the information provided by the attentional or intentional cue was incorrect with respect to the subsequent target stimulus. On invalid trials with both attentional and intentional information, both warning stimuli gave incorrect information. Subjects sat in a dark room in front of the panel displaying the stimulus lights, so that the distance between their eyes and a centrally placed fixation point was approximately 60 cm. Subjects were instructed to fixate on the fixation point* and to respond as accurately and quickly as possible with the hand indicated by the target light, while using the preliminary information as much as possible to facilitate responding. They were told that on the majority of trials, the information provided by the warning stimuli would be correct, but that on a small number of trials, this information might be incorrect. Subjects attended two experimental sessions, given a week apart. At the beginning ofeach session, the subject was given practice in discriminating between a required left and right hand response. For this purpose, he was given trials in which a target stimulus only was presented, until a criterion of 90% accuracy was reached. Following this, the subject was given an additional 50 practice trials in which warning stimuli were presented as well. This was followed by two blocks of experimental trials, with a 5 min rest interval between blocks. Within each block, the order of the four warning conditions was randomized, but occurred equally often. Likewise, hemispace in which the target stimulus was presented and hand of response were randomized but occurred equally often. A total of 640 experimental trials were administered, giving 40 trials for each combination of warning, hemispace, and hand of response. During both practice and experimental trials, the subject received feedback about the accuracy of his responses. An incorrect response was followed by a tone, and a response correction was required before the next trial was administered.
RESULTS Preliminary inspection of the data showed no systematic differences between the results of sessions 1 and 2. Thus, these data were combined in subsequent analyses. Mean reaction times and accuracy data were obtained for each subject as a function of the presence of attentional and intentional information, hemispace in which the target was presented, and hand of response. The means for the conditions in which valid (or no) information was provided are presented in Table 1. *Eye position was not monitored since previous work has shown that the control of attention is largely independent of the eye movement system 114, 161. Furthermore, it appears that when instructed to do so, subjects can accurately maintain fixation [13]. In particular in luminance detection tests, subjects spontaneously suppress eye movements even if they are allowed to make them, because eye movement actually slows down overall reaction time 1171.
440
MIEKE VEKFAELLIE,DAWN BOWERS and KENICTH M. HEILMAN
Table
I. Percentage
of correct
responses
and mean reaction times (msec) for the conditions no) information was provided
Attentional-Valid Left hemispace Right hemispace Intentional-Valid Left hand Right hand Intentional-No Left hand Right hand
in which valid (or
Attentional-No Right hemispace Left hemispace
98% 402 msec 98% 413 msec
91% 415 msec 98% 401 msec
91% 419 msec 91% 468 msec
96% 454 msec 98% 424 msec
91% 520 msec 89% 529 msec
89% 532 msec 89% 530 msec
90% 558 msec 88% 552 msec
89% 560 msec 90% 552 msec
The results of the analysis of the response accuracy data are reported in detail elsewhere [23]. For the purpose of the present discussion, it is important, however, to point out that none of the findings could be interpreted in terms of a speed-accuracy “trade-off”. Correct mean reaction times were submitted to a weighted repeated measures analysis of variance (ANOVA) in which the factors were Attentional Warning (yes, no), Intentional Warning (yes, no), Hemispace of Target (left, right), and Hand of Response (left, right). The results of the analysis disclosed main effects for Attentional Warning [F (1, 23)= 74.02, P
DISCUSSION In the experiment described here. subjects responded with the left or right hand to stimuli presented to the left or right of body midline. Even though the spatial position of the stimulus was irrelevant to the task, significant compatibility effects were observed, consistent with
ATTENTIONAL
FACTORS
550
2
pYzii+
550 500
450 I
450
4oo
I.
k
&___
--------8
400 RHS
LHS
____-_ i’
OF STIMULUS-RESPONSE
500
“,
E
IN THE OCCURRENCE
COMPATIBILITY
441
EFFECTS
l
piGiGiq
LHS
RHS
_____________________
LHS
FIG. 2. Meah reaction
function
LHS
RHS Lelt
Hand ---
Right
RHS
Hand -
time for the left and right hand to target stimuli in left and right hemispace of the presence of valid attentional and intentional information.
as a
previous reports of the Simon effect [lS, 201. In contrast to previous studies, however, our findings indicate that the spatial arrangement of stimuli and responses is not the only factor mediating compatibility effects. Across all warning conditions of the present experiment, the spatial arrangement of stimuli and responses was the same. Compatibility effects, however, were present only in the condition in which intentional warning, but no attentional warning, was provided. The absence of compatibility effects in the condition in which no warning was provided requires some discussion in that the Simon effect was originally demonstrated in a paradigm in which no cues were used. This apparent contradiction, however, may be resolved when considering differences in demand characteristics between the present task and the paradigm originally developed by Simon. Specifically, in the present study no warning trials were randomly intermixed with trials in which preparatory information was given, and as such, the intentional set varied from trial to trial. Indeed, by varying preparatory intentional information, we created explicit response biases. We suggest that on trials without intentional information, this manipulation resulted in the absence of an intentional set. In contrast, we suspect that in the absence of intentional cues in general, there is an implicit, non-specific readiness to respond throughout the experiment. Consequently, the no warning condition in the present experiment should not be seen as a replication of the standard experiment. Notwithstanding, these findings present important implications for both coding and attentional explanations of the Simon effect. According to the coding hypothesis, compatibility effects reflect the results of a comparison between the spatial codes for the stimuli and the responses. Consequently, compatibility effects should be present across all warning conditions of the present study because the spatial codes of stimuli and responses remain unaltered by the attentional manipulations that were introduced. Alternatively, according to the attentional hypothesis, compatibility effects may vary as a function of the attentional manipulations, because these may alter the pattern of hemispheric activation. The present findings indicate that compatibility effects occurred solely under conditions of
442
MIEKEVERFAELLIE, DAWN BOWERSand KENNETHM. HEILMAN
intentional warning only. As such, these findings do not support the coding hypothesis. They do fit, however, in the context of an attentional theory as postulated by Heilman et a/. [lo, 111. According to this theory, when a stimulus is processed by the contralateral hemisphere, this hemisphere is also primed to mediate a response in contralateral hemispace. Responses in contralateral hemispace, therefore, will be faster than responses in ipsilateral hemispace, which are mediated by the opposite hemisphere. Thus, processing of a stimulus creates an implicit response bias favoring responses on the same side. In this study, response bias was created explicitly by providing preliminary intentional information. Compatibility effects were then observed only when intentional information was given. They disappeared when a selective readiness to respond was no longer present, as was the case in the conditions in which no intentional cues were provided. Thus, activation of the hemisphere mediating the response, either implicitly or explicitly, appears to be critical for compatibility effects to occur. It is not entirely clear why compatibility effects were absent in the condition in which both attentional and intentional information are provided, because in this condition, as in the condition with intentional warning only, the hemisphere mediating the response was activated by the intentional cue. One explanation is that these conditions differ with respect to the amount of information they provide. In the condition in which both attentional and intentional information are provided, the only degree of uncertainty resulted from the fact that the information provided was invalid on 20% of the trials. On the majority oftrials, the information was correct, and as such, this condition resembled a simple reaction time task. That is, most trials did not require a choice between different responses or a decision about the stimulus. Under conditions of intentional information only, on the other hand, there was additional uncertainty as to where the target stimulus would occur. We suspect this greater degree of uncertainty resulted in the longer reaction times found in this condition. The presence of compatibility effects in the condition with intentional information only, but not in the condition with both.attentional and intentional information, we suggest, reflects the fact that there was relatively little uncertainty in the latter task. This finding appears consistent with existing evidence that compatibility effects are absent in simple reaction time tasks [l, 21. Further evidence for the importance of attentional factors in the occurrence of compatibility effects comes from a study by BOWERS et al. [S]. In that study, subjects responded to lateralized lights, while deviating their eyes to a fixation point in left hemispace. midline, or right hemispace. The authors found that when the visual half field in which a stimulus was presented was aligned in a compatible side of hemispace, responses were faster than when visual half field and side of hemispace were not aligned. Based on these findings. they concluded that the relationship between each hemisphere and attentional/intentional processes in contralateral hemispace might be an important determinant of response speed to lateralized stimuli. Finally, an attentional explanation has also been invoked to explain how an irrelevant accoustical stimulus can interfere with responses to a lateralized visual stimulus. As SIMON ef al. 119,201 showed, the tendency to respond to a light with the hand on the same side can be counteracted by simultaneously presenting a competing stimulus to the contralateral ear. This finding cannot easily be explained in the framework of the coding hypothesis because no response was required to the accoustical stimulus. According to the coding hypothesis, only stimuli that require a response are coded so that irrelevant features can bias a response.
ATTENTIONAL
FACTORS
IN
THE OCCURRENCE
OF STIMULLJS+RESPONSECOMPATIBILITY EFFECTS
443
Consistent with an attentional hypothesis, Simon et al. suggested that an abrupt accoustical stimulus can elicit a response tendency toward “the major source of stimulation”. Within the context of a hemispheric model, it can be postulated that the accoustical stimulus activates the corresponding hemisphere more than the visual stimulus does. Consequently, the hemisphere processing the accoustical stimulus will be primed to mediate responses in the corresponding side of space. A question remains as to how to reconcile the findings of UMILTA and NICOLETTI [22] with the evidence discussed here, which favors an attentional interpretation of the Simon effect. As discussed, UMILTA and NICOLETTI [22] obtained compatibility effects in a situation in which both stimuli were presented on the same side of space and, similarly, both response keys were located in the same hemispace. According to the attentional hypothesis, compatibility effects should be absent in this situation, because presumably the same hemisphere processes both stimuli and the same hemisphere mediates both responses. One possibility is that each hemisphere may be specialized for attending to and/or acting on stimuli in contralateral hemispace only in a free-field situation, that is, when there is a high probability that stimuli or responses, or both, will randomly occur in either hemispace. When stimuli or responses repeatedly occur on the same side of space, as is the case in the Umilta and Nicoletti study [22], this may result in both hemispheres realigning attention and intention to that side of space. We suggest that under these conditions the functional midline of space may have shifted. A stronger test of the attentional hypothesis, therefore, might have been a “non-blocked” experimental condition in which stimuli and responses randomly occurred in left and right hemispace. Alternatively, it can be argued that the coding hypothesis and the attentional hypothesis are not mutually exclusive. The present study does not provide a critical test of the coding hypothesis because across all conditions both stimuli and responses could be coded in terms of the same left/right factor. Therefore, our findings should not be taken as evidence against a spatial code factor. Rather, they indicate that an intentional mechanism operates in conjunction with a spatial code factor. In most reaction time studies a readiness to respond is implicitly present. However, by manipulating this factor explicitly, this study points to its importance in the occurrence of the Simon effect. The picture that emerges, then, is a complex one in which a variety of factors may contribute to the presence or absence of compatibility effects. The Simon effect, for instance, has also been observed in situations in which stimuli and responses are vertically arranged [7]. Little is known about how the brain mediates attention in upper vs lower hemispace. In addition to spatial correspondence [15], factors such as stimulus salience [7] and movement preference of different body parts [3] have been invoked to explain this finding. Whichever the most parsimonious explanation may be, it appears that a complete understanding of the many situations in which compatibility effects can be observed will require careful analysis of a variety of information-processing characteristics. The present study points to attentional processes as one of them.
REFERENCES 1.
H. A. and RIZZOLATTI. G. Spatial compatibility and anatomical time. Neuropsychologia 15, 295-302, 1977. 2. BERLUCCHI, G., CKEA, F., DI STEFANO, M. and TASSINARI, G. Influence of spatial stimulus-response compatibility on reaction time of ipsilateral and contralateral hand to lateralized light stimuli. /. cup. Ps@ol.: Hum. Percept. Perform. 3, 505-517, 1977. ANZOLA, G. P.. BERTOLONI, A., BWHTEL,
factors
in simple and choice reaction
444
MIEKE VEKFAELLIE.DAWN BOWERS and KENNETH M. HEILMAN
3. BAUER, D. W. and MILLER, J. Stimulus-response compatibility and the motor system. Q. J. exp. fs)‘chol. 34A, 367-380, 1982. hemispheric activation. Neuropsycholoyia 18, 309-319, 4. BOWERS, D. and HEILMAX, K. M. Material-specific 1980. 5. BOWERS, D., HEILMAN, K. M. and VAN DEN ABELL, T. Hemispaceevisual half field compatibility. Neuropsychologia 19, 757-165, 1981. 6. BREBNEK,J., SHEPARD. M. and CAIRNEY, P. Spatial relationships and the motor system. Acta Psycho/. 36,l~ 15, 1972. I. COTTON, B., TZENG, 0. J. and HARDYCK, C. Role of cerebral hemispheric processing in the visual half field stimulus-response compatibility effect. J. ~xp. Psqachol.: Hum. Percept. Perform. 6, 13 23, 1980. 8. FITTS, P. M. and SEEGEK,C. M. Spatial characteristics of stimulus and response codes. J. cup. Psychol. 46, 193-210, 1953. on two-choice response-time. 9. HEDGE, A. and MARSH, N. W. A. The effect of irrelevant spatial correspondence Actu Psychol. 39,421439, 1975. 10. HEILMAN, K. M., WATSON, R. T. and VALENSTEIN, E. Neglect and related disorders. In Clinical Neuropsychology, K. M. HEILMAN and E. VALENSTE~N(Editors), pp. 243-293. Second Edition, Oxford University Press, New York, 1985. underlying hemispatial neglect. Ann. Neural. 5, 166 170, 11. HEILMAN, K. M. and VALENSTEIN,E. Mechanisms 1979. 12. HEISTER, G. and SCHROEDER-HEISTER,P. S-R compatibility effects or cerebral laterality effect? Comments on a controversy. Neuropsycholoyia 23, 427430, 1985. 13. JONES, B. and SANTI, A. Lateral asymmetries in visual perception with and without eye movements. Cortex 14, 164-168. 1970. 14. KLEIN, R. Does oculomotor readiness mediate cognitive control of visual attention? In Attention und Performance VIII, R. S. NWKEKWN (Editor). Lawrence Erlbaum, Hillsdale, New Jersey, 1979. effects on the 15. NICOLETTI, R., ANZOLA, G. P., LUPPIXO, G., RIZZOLATTI, G. and U~~~LTA,C. Spatial compatibility same side of the body midline. J. exp. Psycho/.: Hum. Percept. Perform. 5, 664673. 1982. 16. POSNER, M. I. Orienting of attention. Q. J. exp. Psycho/. 32, 3-25, 1980. 17. POSNER, M. I.. SNYDER, C. R. and DAVIUSON,B. J. Attention and the detection of signals. J. exp. Psychol.: Gen. 109, 16&174, 1980. J. exp. Psychol. 81, 174-176, 1969. 18. SIMON, J. R. Reactions towards the source of stimulation. 19. SIMW, J. R., CRAFT, J. L. and SMALL, A. M. Reaction toward the apparent source of an auditory stimulus. J. rxp. Psychol. 89, 203 206, 1971. Reaction time as a function ofear20. SIMON, J. R., HINKICHS, J. V. and CRAFT, J. L. Auditory S-R compatibility. hand correspondence and ear-response-location correspondence. J. exp. Psqachol. 86, 97-102, 1970. 21. SIMON, J. R., SLY, P. E. and VILAPAKKAM,S. Effects of compatibility of S-R mapping on reactions toward the stimulus source. Actu P.~ycho/. 47, 63 81, 1981. due to irrelevant spatial cues. 22. U~~~LTA,C. and NICOLLXTI, R. Attention and coding effects in S ORcompatibility In Attention and Performance XI, M. POSNER and 0. S. MAKIN (Editors), pp. 457 471. Lawrence Erlbaum, Hillsdale, New Jersey, 1985. 23. VERFAELLIE.M., BOWERS. D. and HEILMAX, K. M. Hemispheric asymmetries in mediating intention, but not selective attention. Neuropsychologia. In press. 24. WALLACE, R. J. S-R compatibility and the idea of a response code. J. exp. Psychol. 88, 354360, 1971. effects involving kinesthetic cues. J. exp. Psychol. 93, 163-168, 1972. 2s. WALLACE, R. J. Spatial S--R compatibility