- Email: [email protected]
Spatial Stimulus —Resonse Compatibility in Callosotomy Patients and Subjects with Callosal Agenesis
N
B C
P
R Q1
V E
6
ScienceL P 0
@
r
1 r
G $
B +
0
S
S
t
C
piC a S
P a C S A
m oi l wa a
A G T
G A
GAB
L
SE
u g I
Dipartimento di Scienze Neurologiche e della Visione, Sezione di Fisiologiaumana,Stradale Grazie 8, 37134 Verona, Italy
AGLIOT1, S., G. TASSINARI AND G. BERLUCCHI. Spatialstimulw–resporrsecompatibilityin callosotomypatients and subjects with callosal agenesis.NEUROSCI BIOBEHAV REV 20(4)623+29.—Subjectswith partial or complete defects of the corpus callosum, either congenital or acquired, performed a choice reaction time (RT) task involvinga right or left key-press response to a light presented at random in the right or left visual hemifield. Like normal subjects, all of them exhibited two additive effects typical of these tasks: the spatial stimulus–responsecompatibility effect (faster RT for stimuli and responses matched for side), and the hand placement effect (longer RT for responses performed with crossed hands). Two subjects with a complete callosal defect, one acquired and the other congenital, showed a third effect, not present in normal subjects, consisting of a marked advantage for RT of responses with the hand anatomically ipsilateral to the stimulus, independent of both stimulus–response compatibility and hand placement. These findings can be interpreted according to a hierarchical model of information processing assuming that, in the absence of the corpus callosum, the matching of the mental codes for the stimulus and response sets takes place solely in the hemisphere reeeiving the stimuhrs, with a subsequent rapid-intrahemispheric or slow-interhemispheric transmission of the response command to the appropriate motor centers. Copyright O 1996Elsevier Science Ltd. Stimulus–response compatibility Callosotomy Callosal agenesis Choice reaction time
INTRODUCTION
A TYPICAL task for studying spatial stimulusresponse compatibility effects (24) involves making a speeded decision to press a right or left key depending on the position of a light flash presented at random in one or the other visual hemifield. In one condition, called compatible, subjects are instructed to react to right flashes by pressing the right key and to left flashes by pressing the left key; in another condition, called incompatible, they are instructed to use the opposite stimulus–response pairings. A further experimental manipulation involves responding with the hands in an orthodox, anatomical position on half of the trials in each condition, and a crossed position on the remaining trials. When the hands are in the anatomical position, the right key is operated by the right hand and the left key is operated by the left hand; when the hands are in the crossed position, each hand operates the opposite key. In normal subjects, RT is faster when the locus of the flash and the locus of the response are matched for the
side compared to when they are not, regardless of whether the hands are placed in an uncrossed or crossed position. In turn, RT of responses made with crossed hands is longer than RT of responses made with uncrossed hands, independent of the correspondence or non-correspondence between the side of the stimulus and the side of the response. The independent and additive effects of the two factors of spatial stimulus–response compatibility and hand placement most probably arise from the response selection stage of information processing, the selection being made easier by the congruence between the positional codes of flash and key, as well as by the congruence between the positional codes of key and hand (14,24,34).When ordered from the shortest to the longest, RTs form the following systematic pattern: compatible uncrossed < compatible crossed < incompatible uncrossed < incompatible crossed. This RT pattern, first reported by Brebner et al. (8), has been confirmed in several successive studies on normal subjects (2,24),and has also been found to be unmodifiable by extensive practice with the task (11,25).A third factor which may potentially affect
SR
AGLIOTI, TASSINARI
RT in spatial stimulus–response compatibility tasks is the difference in length between the most direct neural pathways assumed to mediate responses with the hand ipsilateral to the stimulus (i.e. right hand and right stimulus, left hand and left stimulus) and the most direct neural pathways assumed to mediate responses with the hand contralateral to the stimulus (i.e. left hand and right stimulus, right hand and left stimulus). Given that each visual hemifield is projected to, and each hand is under the motor control of the opposite hemisphere, ipsilateral responses should be subserved by a relatively short intrahernispheric pathway, while contralateral responses should be subserved by a longer pathway including an interhemispheric component. In normal subjects, ipsilateral responses are indeed slightly but systematically shorter than contralateral responses (by 2-3 ms) in simple RT tasks which involve a single keypress reaction to literalized flashes and therefore do not require a choice between two keys (3,5,15,19).In simple RT tasks, the hand placement has no effect, since the ipsilateral advantage is the same, regardless of whether the hands are in the uncrossed or crossed position (2,4). However, no significant effect from the assumed difference in length between neural pathways for ipsilateral and contralateral responses has ever been found in the above described choice RT tasks, where, as already mentioned, the prevailing effects are those from spatial stimulus–response compatibility and crossed–uncrossed hand placement (24,7). The reasons for the apparent absence of effects from assumed differences in length of neural pathways in spatial compatibility tasks have not been ascertained. Two different hypotheses can be offered to account for it: (1) the neural pathway effect found in simple RT tasks is at least an order of magnitude smaller than the effects from spatial stimulus–response compatibility and hand placement, and therefore it can be easily masked by these latter effects in choice RT tasks (4,7,34); (2) neural pathway effects occur solely when visuomotor responses are promptly emitted in a reflexlike fashion in simple RT tasks, while in choice RT tasks involving spatial compatibility effects the emission of the response must be preceded by a decision process during which there is ample time for visual information to reach both hemispheres whichever the side of entry of the visual stimulus (33). Recently, we have examined the role of the corpus callosum in the interhemispheric transfer of information for fast visuomotor integration in patients with total or partial callosal defects, either acquired or congenital (1,5,32). The corpus callosum is the major pathway for communication between the hemispheres, and since it contains fibers large enough for transferring information between the hemispheres in a few milliseconds (28), it is most likely that it provides the interhemispheric substrate for making fast, reflex-like responses with one hand to contralateral visual stimuli. This possibility is definitely supported by the finding that the ipsilateral–contralateral RT differences in simple RT tasks are much increased in subjects with total callosal agenesis (1,5,1OA,17,18,2O)or patients with drug-resistant forms of epilepsy submitted to a total therapeutic resection of the corpus callosum
AND BERLUCCHI
(1,5,9,29). The acallosal patients are abnormally slow when they respond with one hand to a visual stimulus in the opposite hemifield, such that the ipsilateral-contralateral difference may be as long as about 20 ms in callosal agenesis and more than 60 ms after callosal section. These long ipsilateral-contralateral differences most probably reflect a slow interhemispheric transfer of information along multisynaptic extracallosal brainstem pathways (1,5). As in normal subjects, they are unaffected by the uncrossed or crossed placement of the hands (1,20). In the study reported here, we have assessed whether the abnormal ipsilateral–contralateral differences seen in acallosal subjects, much greater than those seen in normal subjects, can affect the RT pattern in the typical spatial stimulus–response compatibility paradigm described above. Earlier investigations have revealed typical effects of spatial stimulus–response compatibility in subjects with callosal defects (16,30), but with a single exception (1OA)RT assessments have usually been limited to the uncrossed-hands condition, so that it is not known whether acallosals, like normal controls, exhibit a hand placement effect, and whether this effect, if present, adds to or interacts with the compatibility effect and possibly the neural pathway effect. To test these possibilities, we have studied two subjects who had undergone surgical callosal sections, one complete and the other limited to the anterior twothirds of the structure, and two subjects with total callosal agenesis, and have compared their data with those collected in eight normal controls performing the same task. METHODS
Subjects
Male subject M.E., born in 1970,was submitted to a complete section of the corpus callosum at the Neurosurgical Institute of the Catholic University in Rome (Prof. G. F. Rossi). Callosotomy was performed in two stages (Februa~ and June 1989) in an attempt to control a form of post-traumatic epilepsy with complex partial seizures and secondary generalization which had proved totally resistant to pharmacological therapy as well as to a removal of a focus in the right prefrontal cortex. Callosotomy has resulted in a marked favorable change in both severity and frequency of the seizures. Pharmacological treatment with Phenobarbital and Phenytoin has been continued throughout the postoperative period. At the times of testing for RT (April 1990,December 1991), standard clinical examinations revealed a stationary condition with no neurological deficits, except for a severe lefthand ideomotor dyspraxia in verbal command (but not on imitation), a left-hand anemia and alexia in the left hemifield. The completeness of the callosal section and the integrity of the anterior commissure have been confirmed by magnetic resonance imaging (MRI). I.D., also born in 1970,is another epileptic male patient who was submitted to resection of the anterior two-thirds of the corpus callosum in May 1990, also at the Neurosurgical Institute of the Catholic University in
SPATIAL COMPATIBILITY
IN ACALLOSALS
Rome (Prof. G. F. Rossi). Also in this case, the extent and location of the callosal section have been confirmed by MRI. At the time of testing for the present study (April 1991),he showed no apparent sign of interhemispheric disconnection. P.M. and R.B., two male subjects born, respectively, in 1961 and 1976, have been diagnosed by MRI as congenitally lacking the corpus callosum. At the time of testing, they were free from major neurological symptoms, and appeared to have normal intelligence; the only clear sign of interhemispheric disconnection exhibited by these subjects was an increase of simple RT of each hand to flashes in the contralateral visual field. More detailed information about these four subjects with callosal defects can be found in Aglioti et al. (l), Tassinari et al. (32). Control subjects were eight male students or staff members of the University of Verona, ranging in age from 23 to 34 years. Four were experienced with RT tasks, and four were not. Apparatus
Testing was carried out in a partially sound-proofed cubicle where the subject sat facing a horizontal perimeter 57-cm in radius attached to a white screen. The screen and the perimeter were illuminated from above at a luminance of 0.15 cd/m2. Each subject positioned his head in a forehead- and chin-rest located at the centre of curvature of the perimeter. Two solid state miniature bulbs (TIL 222) were fastened to the perimeter, one on the left and the other on the right of a central mark for fixation. From the patient’s viewpoint, the angular distance between the fixation mark and the bulb on each side was 1OO.Each bulb could be lighted individually by a 5-ins square pulse of current, producing a gallium phosphide green flash with the same duration and an intensity of about 1000pcd. The response devices were two button-keys placed at 20 cm from the midline, one on the right and the other on the left, within a comfortable reaching distance from the subject’shands. Each button-key was mounted on the top of a brass cylinder; the subject clutched each cylinder with the hand on the same or the opposite side and positioned the tip of each thumb on the button-key; the required response consisted of pressing one or the other key by flexing the corresponding thumb. During testing, subjects wore a headset which delivered a standard auditory warning signal consisting of a 1000Hz tone with a suprathreshold but comfortable intensity. Procedure
Subjects were first accustomed to the experimental situation by running several practice trials, and then underwent formal testing over two or more days. The two main independent variables, namely the relationship between the stimulus side and the response side and the uncrossed or crossed placement of the hands, were varied orthogonally, giving rise to four combinations: compatible stimulus–response pairings with uncrossed placement of the hands; incompatible stimulus–response pairings with uncrossed placement of the
hands; compatible stimulus–response pairings with crossed placement of the hands; and incompatible stimulus–response pairings with crossed placement of the hands. In each subject, each of these four combinations was tested in a block of 50 trials, during which the light stimulus appeared 25 times on the right and 25 times on the left in a random, unpredictable order. In the two blocks with compatible stimulus–response pairings, subjects had to press the right key in reaction to right flashes and the left key in reaction to left flashes. In one of these blocks, the hands were held uncrossed, so that the right hand pressed the right key and the left hand pressed the left key; in the other block, the hands were held crossed at the elbow, so that the left hand pressed the right key and the right hand pressed the left key. In the remaining two blocks, stimulus–response pairings were incompatible, so that subjects had to press the left key in reaction to right flashes and the right key in reaction to left flashes; the placement of the hands was uncrossed in one block and crossed in the other block. The order of the four different blocks varied from subject to subject: it was counterbalanced across subjects in the normal group, whereas the subjects with callosal defects were each assigned to a sequence chosen at random. Subject M.E. also performed a second, different sequence, so that he yielded twice as many RTs as the other subjects. Each trial began with a warning auditory signal delivered via the headphone set, followed after an interval varying randomly between 1 and 3s by a right or left flash. Subjects were instructed to fixate on the central mark upon hearing the warning signal and to maintain fixation until after pressing the appropriate key. Eye position and maintenance of fixation were monitored by television, and trials with failures to fixate were aborted and repeated. On each trial, the time elapsing between flash onset and key-pressing response was electronically measured to the nearest millisecond.The presentation of the flashes and the recording of the RT data were automatically controlled by a computer which rejected RTs of responses made with the wrong hand or with both hands, as well as RTs shorter than 150ms or longer than 989 ms. Rejected trials were replaced later in the sequence until obtaining acceptable RTs for all trials. RESULTS
Table 1 shows mean RTs for the relevant comparisons between compatible–incompatible field–key combinations, uncrossed–crossed hand–key combinations, and ipsilateral-contralateral hand–key combinations for the control group as well as for the four subjects with callosal defects. The differences between members of each pair were analysed statistically by assessing the significance of the appropriate two-way interactions in separate ANOVAs having Hand, Field and Key as main factors. The ANOVA for the normal group followed a repeated measurement design with subjects considered as a random variable, each subject contributing a mean score for each level of the three main factors; the ANOVAs for each experimental subject were performed on raw data.
AGLIOTI, TASSINARI AND BERLUCCHI TABLE 1 DIFFERENT COMBINATIONS BETWEEN HANDS, KEYS AND HEMIFIELDS FOR THE CONTROL GROUP AND FOR THE DIFFERENT SUBJECTS WITH CALLOSAL DEFECTS Field–Key Compatible Incompatible Controls M.E. I.D. P.M. R.B.
309.3 401.9 92.6 @
Hand–Key Uncrossed Crossed 329.1 382.1 53.0 (p
All subjects, whether normal or acallosal showed the typicaleffects:RT was significantlyshorter with compatible than with incompatible spatial stimulus-response combinations, and with uncrossed than with crossed hand placement. The hand–field interaction was insignificantin all cases, except in the complete callosotomy subject M.E. and the complete agenesis subject R.B., both of whom showed a large and significant advantage for the hand ipsilateral to the stimulus. Figures 1–5 show the ordering of RTs as a function of field–hand compatibility and hand placement for the control group and for each of the experimental subjects. In all cases, except M.E. and R.B., such ordering followed the typical pattern compatible uncrossed < compatible crossed < incompatible uncrossed < incompatible crossed. M.E. and R.B. showed a different pattern in so far as the compatible crossed condition yielded a longer RT compared to the incompatible uncrossed condition. This effect appears to be due to the prevalence in the former condition of the ipsilateral-contralateral difference over the compatible–incompatible difference. The trends shown by M.E. and R.B. were present also in acallosal subject P.M., but were not as pronounced as in the other two subjects with total callosal defects, and the hand–field interaction did not reach significance, although there was a large difference in favor of ipsilateral over contralateral RTs. DISCUSSION
previous keeping “ (2&l,14,21,24,25,2~!~), th~~~sent results in ~%% subjects bear out the separate effects from the two main independent variables: the effect from the compatibility between locus of stimulusand locus of response, and the effect from the hand placement, both of which are best attributed to the ways in which the stimulus and response sets are mentally coded, and to the ease with which a stimuluscode is translated into a response code. The selection of the key ipsilateral to the flash and the selection of the key on the side of the body attached to the responding limb appear to be naturally easier than
Field–Hand Ipsilateral Contralateral 358.3 352.9 -5.4 (n.s.) 625.4 684.5 59.1 (p
the corresponding converse selections. The occurrence of the two effects is attested, respectively,by the significance in the ANOVA of the field–key interaction and the hand–key interaction. The additivity of the two effects is suggested by the gradual increase of RT from the compatible uncrossed condition to the incompatible crossed condition, as shown in Fig. 1. By contrast, as in earlier investigations(2,4,7,8,14,34),the results from our normal group offered no evidence for the occurrence of systematic RT effects from the presumed difference in length between neural pathways for visuomotor integration within and across hemispheres. The hand–field interaction was completely insignificantin the ANOVA and the mean RT of contralateral responses was insignificantlyshorter than the mean RT of ipsilateral responses;in all subjects,right-hand responses were not significantlyfaster for right compared to left flashes,and left-hand responses were not significantlyfaster for left compared to right flashes. A comparable pattern of results was obtained in patient I.D. with an anterior callosal section, confirming that the integrity of the splenium of the corpus callosum sufficesfor ensuring a virtually normal behavior in most, if not all, visuomotor tasks (6,13,32). In at least two of the three subjects with total callosal defects, the pattern of results differed from that of the normal subjects in so far as RT was influenced not only by the field–key and hand-key interaction, but also by the hand–field interaction. In callosotomized subject M.E. and callosal agenetic subject R.B., the latter interaction was highly significant because stimuli presented in each field evoked faster responses from the anatomically ipsilateral than contralateral hand, independent of the spatial stimulus–response compatibility as well as of the uncrossed or crossed placement of the hands. As a result, the ordering of RTs of the four combinations between stimulus–response compatibility and hand placement was different from that of the normal subjects: in both M.E. and R.B., RT of the compatible crossed combination was as long as or longer than the RT of the incompatible uncrossed and incompatible crossed combinations. In the other acallosal subject P.M., the hand–field interaction was
SPATIAL COMPATIBILITY
IN ACALLOSALS P. M.: C8110wlagmds
Cmlrob(n=B) 9--
m
I
C
W
4
4 .
+-
.?
!
I
3
2
—
2
FIG. 1. Normal controls. Means and standard errors of RTs for the four combinations of compatible and incompatible reactions performed with uncrossed a;d crossed hands. RTs of compatible responses are shown as means of responses with the right hand to the right flash and with the left hand to the left flash in the uncrossed-hands condition, and with the left hand to the right flash and with the right hand to the left flash in the crossed-hands condition. RTs of incompatible responses are shown as means of responses with the left hand to the right flash and with the right hand to the left flash in the uncrossed-hands condition, and with the right hand to the right flash and with the left hand to the left flash in the crossed-hands condition.
U
c
C
FIG. 4. Subject P.M. with callosal agenesis. Means and standard errors of RTs for the four combinations of compatible and incompatible reactions performed with uncrossed and crossed hands. See ~egendto Fig. 1 f& explanations.
R.B.:Callo8d a@imb 5
E m 4
u.
C0mpbbeal1080t0my
-. I
C .
T
Subiect . R.B. with callosal agenesis. Means and standard errors of RTs for the four combinati&s of compatible and incompatible reactions performed with uncrossed and crossed hands. See legend to Fig. 1 for explanations. F
1 .
U
— l
FIG. 2. Subiect M.E. with complete callosotomy. Means and standard erro~s of RTs for the four-combinations of compatible and incompatible reactions performed with uncrossed and crossed hands. See legend to Fig. 1 for explanations.
1.D. : AnbrlIxc91hN010MY m
c~
I
T
m
~
r
4
4 4 4
—
FIG. 3. Subject I.D. with anterior callosotomy. Means and standard errors of RTs for the four combinations of compatible and incompatible reactions performed with uncrossed and crossed hands. See legend to Fig. 1 for explanations.
not significant,but there was, nevertheless, a tendency for a clear-cut advantage for RT of ipsilateral over contralateral responses. Why this acallosal subject did not show exactly the same general pattern of results as that of M.E. and R.B. remains to be determined. A somewhat different pattern of results has also been reported by Di Stefano et al. (lOA) who however did not distinguish between agenetic subjects and a single partially callosotomized patient. These findingssuggest that rules for coding the stimulus and response sets in the performance of a typical spatial compatibility task are similar in the normal and acallosal brain; yet, unlike normal subjects, in acallosal subjects there is a “neural pathway” effect which appears to be additive with those of spatial stimulus–response compatibility and hand placement. This neural pathway effect is similarto that found in the same subjectsin simple RT tasks, in which the other two effects are absent because performance does not require the coding of the positionsof either stimulusor response (l). As already mentioned, the simple RT tasks which yield differences between ipsilateral and contralateral manual responses to literalized flashes in normal
AGLIOTI, TASSINARI
subjects are likely to depend on relatively straightforward cortico-cortical connections, akin perhaps to the transcortical connections that are thought to account for reflex-like responses of the motor cortex to inputs from muscle spindles (22). In acallosal subjects, the main interhemispheric cortical link required for contralateral responses is missing and is substituted by less efficient extracallosal connections which account for the increase in contralateral over ipsilateral RT. The absence of a neural pathway effect in spatial choice RT tasks in normal subjects has been attributed to its masking by the much greater effects from spatial stimulus–response compatibility and hand placement. However, at least according to the additive factor logic, the effect should be unmasked by appropriate subtraction procedures (7). Although previous limited analyses and metanalyses of results for spatial compatibility tasks hinted at the possible occurrence of RT differences in favor of ipsilateral over contralateral RT, these differences were never found to be significant (2,4,7,8,14).The finding of a small insignificant difference in favor of contralateral RTs in the normal subjects of the present experiment would seem to argue that the previous contrary, equally insignificant findings were indeed chance results, and that the normal RT pattern in spatial compatibility tasks is unaffected by the simple neural pathway effect. The neural substrates involved in choosing the literality of the response as a function of the literality of the stimulus are probably quite different from the straightforward connections assumed to mediate simple visuomotor RTs of each hand to a literalized visual stimulus; hence it makes sense that the simplest and most direct anatomical connections between visual fields, cerebral hemispheres and hand muscles can influence simple but not choice RT. Why, then, can a neural pathway effect become apparent in subjects lacking the corpus callosum? Is this simply due to the mere size of the difference between ipsilateral and contralateral RTs in acallosals, which in simple tasks is so much greater than in normal subjects? This possibility would be in agreement with the hypothesis that a neural pathway effect cannot be seen in spatial compatibility tasks in normal subjects
AND BERLUCCHI
because it is too small compared to the effects arising from the coding of stimuli and response and the matching of codes. Since, as discussed above, we tend to reject this possibility, we propose the following alternative interpretation. We hypothesize that in normal subjects performing spatial compatibility tasks, both hemispheres participate in the decisive processes leading to a response to a literalized visual stimulus due to rapid interhemispheric communication via the corpus callosum. Once a decision is reached, the speed of activation of the appropriate motor mechanisms is independent of the side of entry of the literalized stimulus. By contrast, in acallosal subjects, the decisive processes would be largely limited to the hemisphere initially receiving the literalized stimulus, so that the subsequent activation of the appropriate motor mechanisms would vary in speed depending on whether the executive motor command must be issued by the decisional or the opposite hemisphere. In the latter case, a time-consuming interhemispheric transfer would be needed, accounting for the occurrence of the neural pathway effect and its additivity with the other effects arising from the earlier decisional stages. Indirect support for our hypothesis comes from the well established evidence that (1) digital and manual responses are known to be controlled exclusively by the contralateral hemisphere (1,5,6,13,31); (2) each hemisphere of the acallosal brain was shown to be cognizant of both sides of extrapersonal space (10,12,23); and (3) while each hemisphere is able to code and direct attention to specific locations in both the visual hemifield, the speed of responses to stimuli in such locations was found to be, as in the present study, much greater for the contralateral than for the ipsilateral hand (26). A
The studies described in this paper have been carried out with financial support from the ConsiglioNazionale delle Ricerche, the Ministero dell’Universit2t e della Ricerca Scientific e Tecnologica, the Human Frontier Science Programme Organization and the Human Capital Programme of the European Community.
R 1. Aglioti, S.; Berlucchi, G.; Pallini, R.; Rossi, G. F.; Tassinari, G. Hemispheric control of unilateral and bilateral responses to literalized light stimuli after callosotomy and in callosal agenesis. Exp. Brain Res. 95:151–165;1993. 2. Anzola, G. P.; Bertoloni, G.; Buchtel, H. A.; Rizzolatti, G. Spatial compatibility and anatomical factors in simple and choice reaction time. Neuropsychologia 15:295–305;1977. 3. Bashore, T. R. Vocal and manual reaction time estimates of interhemispheric transmission time. Psychol. Bull. 89:352–368; 1981. 4. Berlucchi, G.; Crea, F.; Di Stefano, M.; Tassinari, G. Influence of spatial stimulus–response compatibility on reaction time of ipsilateral and contralateral hand to literalized light stimuli. J. Exp. Psychol. Hum. Perc. Perf. 3:505-517;1977. 5. Berlucchi, G.; Aglioti, S.; Marzi, G. A.; Tassinari, G. Corpus callosum and simple visuomotor integration. Neuropsychologia 33:923-936;1995. 6. Bogen, J. E. Physiological consequences of complete or partial commissural section. In: Apuzzo, M. L. J., ed. Surgery
of the Third Ventricle. Baltimore: Williams & Wilkins; 1987:175-194. 7. Bradshaw, J. L.; Umiltil, C. A reaction time paradigm can simultaneously index spatial compatibility and neural pathways effects: a reply to Levy. Neuropsychologia22:99–101;1984. 8. Brebner, J.; Shephard, M.; Cairney, P. Spatial relationships and S-R compatibility. Acta Psychological37:93-106;1972. 9. Clarke, J. M.; Zaidel, E. Simple reaction times to Iateralized light flashes: Varieties of interhemispheric communication routes. Brain 112:849+70;1989. 10. Corballis, M. E. Visual integration in the split brain. Neuropsychologia33:937-959;1995. lOA. Di-Stefano, M.; Sauerwein, H. C.; Lassonde, M. Influence of anatomical factors and spatial compatibility on the stimuhrsresponse relationship in the absence of the corpus callosum. Neuropsychologia30:177-185;1992. 11. Dutta, A.; Proctor, R. W. Persistence of stimulus–response compatibility effects with extended practice. J. Exp. Psychol.: Learn. Mere. Cogn. 18:801-809;1992.
SPATIAL COMPATIBILITY
IN ACALLOSALS
12. Gazzaniga, M. S. Perceptual and attentional processes following callosal section in humans. Neuropsychologia 25:119–133;1987. 13. Gazzaniga, M. S. Principles of human brain organization derived from split-brain studies. Neuron 17:217-228;1995. 14. Heister, G.; Schroeder-Heister, P.; Ehrenstein, W. H. Spatial coding and spatio-anatomical mapping: Evidence for a hierarchical model of spatial stimulus–response compatibility. In: Proctor, R. W.; Reeve, T. G., eds. Stimulus–response compatibility. Amsterdam: Elsevier; 1990:117-143. 15. Hoptman, M. J.; Davidson, R. J. How and why do the two cerebral hemispheres interact? Psychol. Bull. 116:195–219;1994. 16. Jeeves, M. A. Psychologicalstudies of three cases of congenital agenesis of the corpus callosum. In: Ellinger, E. G. Functions of the corpus callosum. London: J. & A. Churchill; 1965:73–94. 17. Jeeves, M. A. A comparison of interhemispheric transmission times in acallosals and normals. Psychon. Sci. 16:245-246;1969. 18. Milner, A. D. Visual integration in callosal agenesis. In: Lassonde, M.; Jeeves, M. A., eds. Callosal agenesis—a natural split-brain? New York: Plenum Press; 1994:171–183. 19. Milner, A. D.; Lines, C. R. Interhemispheric pathways in simple reaction time to literalized light flash. Neuropsychologia 20:171-179;1982. 20. Milner, A. D.; Jeeves, M. A.; Silver, P. H.; Lines, C. R.; Wilson, J. G. Reaction times to literalized visual stimuli in callosal agenesis: stimulus and response factors. Neuropsychologia 23:323-331;1985. 21. Nicoletti, R.; Anzola, G. P.; Luppino, G.; Rizzolatti, G.; Umiltil, C. Spatial compatibility effects on the same side of the body midline. J. Exp. Psychol., Hum. Perc. Perf. 8:664-673; 1982. 22. Phillips, C. G. Motor apparatus of the baboon’s hand. Proc. Roy. S 173:141–174;1969. 23. Plourde, G.; Sperry, R. W. Left hemisphere involvement in left spatial neglect from right-sided lesions. A commissurotomy study. Brain 107:95–106;1984.
24. Proctor, R. W.; Reeve, T. G., eds. Stimulus–response compatibility. Amsterdam: Elsevier; 1990:89-116. 25. Proctor, R. W.; Dutta, A. Do the same stimulus–response relations influence choice reactions initially and after practice? J. Exp. Psycho].,Learn. Mere. Cogn. 19:922-930;1993. 26. Reuter-Lorenz, P. A.; Fendrich, R. Orienting attention across the vertical meridian: Evidence from callosotomy patients. J. Cognit. Neurosci. 2:232-238;1990. 27. Riggio, L.; Gawryszewski, L.; Umilth, C. What is crossed in crossed-hand effects? Acta Psychological62:89–100;1986. 28. Ringo, J. L.; Doty, R. W.; Demeter, S.; Simard, P. Y. Time is of the essence: A conjecture that hemispheric specialization arises from interhemispheric conduction delay. Cerebral Cortex 4:331-343;1994. 29. Sergent, J.; Myers, J. J. Manual, blowing, and verbal simple reactions to literalized flashes of light in commissurotomized patients. Perc. Psychophys.37:571-578;1985. 30. Smith, K. U. Bilateral integrative action of the cerebral cortex in man in verbal association and sensori-motor coordination. J. E Psychol. 37:367-376;1947. 31. Sperry, R. W.; Gazzaniga, M. S.; Bogen, J. E. Interhemispheric relationships: the neocortical commissures: syndromes of hemispheric disconnection. In: Vinken, P. J.; Bruyn, G.W., eds. Handbook of clinical neurology, vol. 4. Amsterdam: NorthHolland; 1969:4:273-290. 32. Tassinari, G.; Morelli, M.; Berlucchi, G. Interhemispheric transmission of information in manual and verbal reaction-time tasks. Hum. Neurobiol. 2:7745; 1983. 33. Tassinari, G.; Aglioti, S.; Pallini, R.; Berlucchi, G.; Rossi, G. F. Interhemispheric integration of simple visuomotor responses in patients with partial callosal defects. Behav. Brain Res. 64:141-149;1994. 34. Umilta, C. A.; Nicoletti, R. Spatial stimulus–response compatibility. In: Proctor, R. W.; Reeve, T. F., eds. Stimulus–response compatibility. Amsterdam: Elsevier; 1990:89-116.
B C
P
R Q1
V E
6
ScienceL P 0
@
r
1 r
G $
B +
0
S
S
t
C
piC a S
P a C S A
m oi l wa a
A G T
G A
GAB
L
SE
u g I
Dipartimento di Scienze Neurologiche e della Visione, Sezione di Fisiologiaumana,Stradale Grazie 8, 37134 Verona, Italy
AGLIOT1, S., G. TASSINARI AND G. BERLUCCHI. Spatialstimulw–resporrsecompatibilityin callosotomypatients and subjects with callosal agenesis.NEUROSCI BIOBEHAV REV 20(4)623+29.—Subjectswith partial or complete defects of the corpus callosum, either congenital or acquired, performed a choice reaction time (RT) task involvinga right or left key-press response to a light presented at random in the right or left visual hemifield. Like normal subjects, all of them exhibited two additive effects typical of these tasks: the spatial stimulus–responsecompatibility effect (faster RT for stimuli and responses matched for side), and the hand placement effect (longer RT for responses performed with crossed hands). Two subjects with a complete callosal defect, one acquired and the other congenital, showed a third effect, not present in normal subjects, consisting of a marked advantage for RT of responses with the hand anatomically ipsilateral to the stimulus, independent of both stimulus–response compatibility and hand placement. These findings can be interpreted according to a hierarchical model of information processing assuming that, in the absence of the corpus callosum, the matching of the mental codes for the stimulus and response sets takes place solely in the hemisphere reeeiving the stimuhrs, with a subsequent rapid-intrahemispheric or slow-interhemispheric transmission of the response command to the appropriate motor centers. Copyright O 1996Elsevier Science Ltd. Stimulus–response compatibility Callosotomy Callosal agenesis Choice reaction time
INTRODUCTION
A TYPICAL task for studying spatial stimulusresponse compatibility effects (24) involves making a speeded decision to press a right or left key depending on the position of a light flash presented at random in one or the other visual hemifield. In one condition, called compatible, subjects are instructed to react to right flashes by pressing the right key and to left flashes by pressing the left key; in another condition, called incompatible, they are instructed to use the opposite stimulus–response pairings. A further experimental manipulation involves responding with the hands in an orthodox, anatomical position on half of the trials in each condition, and a crossed position on the remaining trials. When the hands are in the anatomical position, the right key is operated by the right hand and the left key is operated by the left hand; when the hands are in the crossed position, each hand operates the opposite key. In normal subjects, RT is faster when the locus of the flash and the locus of the response are matched for the
side compared to when they are not, regardless of whether the hands are placed in an uncrossed or crossed position. In turn, RT of responses made with crossed hands is longer than RT of responses made with uncrossed hands, independent of the correspondence or non-correspondence between the side of the stimulus and the side of the response. The independent and additive effects of the two factors of spatial stimulus–response compatibility and hand placement most probably arise from the response selection stage of information processing, the selection being made easier by the congruence between the positional codes of flash and key, as well as by the congruence between the positional codes of key and hand (14,24,34).When ordered from the shortest to the longest, RTs form the following systematic pattern: compatible uncrossed < compatible crossed < incompatible uncrossed < incompatible crossed. This RT pattern, first reported by Brebner et al. (8), has been confirmed in several successive studies on normal subjects (2,24),and has also been found to be unmodifiable by extensive practice with the task (11,25).A third factor which may potentially affect
SR
AGLIOTI, TASSINARI
RT in spatial stimulus–response compatibility tasks is the difference in length between the most direct neural pathways assumed to mediate responses with the hand ipsilateral to the stimulus (i.e. right hand and right stimulus, left hand and left stimulus) and the most direct neural pathways assumed to mediate responses with the hand contralateral to the stimulus (i.e. left hand and right stimulus, right hand and left stimulus). Given that each visual hemifield is projected to, and each hand is under the motor control of the opposite hemisphere, ipsilateral responses should be subserved by a relatively short intrahernispheric pathway, while contralateral responses should be subserved by a longer pathway including an interhemispheric component. In normal subjects, ipsilateral responses are indeed slightly but systematically shorter than contralateral responses (by 2-3 ms) in simple RT tasks which involve a single keypress reaction to literalized flashes and therefore do not require a choice between two keys (3,5,15,19).In simple RT tasks, the hand placement has no effect, since the ipsilateral advantage is the same, regardless of whether the hands are in the uncrossed or crossed position (2,4). However, no significant effect from the assumed difference in length between neural pathways for ipsilateral and contralateral responses has ever been found in the above described choice RT tasks, where, as already mentioned, the prevailing effects are those from spatial stimulus–response compatibility and crossed–uncrossed hand placement (24,7). The reasons for the apparent absence of effects from assumed differences in length of neural pathways in spatial compatibility tasks have not been ascertained. Two different hypotheses can be offered to account for it: (1) the neural pathway effect found in simple RT tasks is at least an order of magnitude smaller than the effects from spatial stimulus–response compatibility and hand placement, and therefore it can be easily masked by these latter effects in choice RT tasks (4,7,34); (2) neural pathway effects occur solely when visuomotor responses are promptly emitted in a reflexlike fashion in simple RT tasks, while in choice RT tasks involving spatial compatibility effects the emission of the response must be preceded by a decision process during which there is ample time for visual information to reach both hemispheres whichever the side of entry of the visual stimulus (33). Recently, we have examined the role of the corpus callosum in the interhemispheric transfer of information for fast visuomotor integration in patients with total or partial callosal defects, either acquired or congenital (1,5,32). The corpus callosum is the major pathway for communication between the hemispheres, and since it contains fibers large enough for transferring information between the hemispheres in a few milliseconds (28), it is most likely that it provides the interhemispheric substrate for making fast, reflex-like responses with one hand to contralateral visual stimuli. This possibility is definitely supported by the finding that the ipsilateral–contralateral RT differences in simple RT tasks are much increased in subjects with total callosal agenesis (1,5,1OA,17,18,2O)or patients with drug-resistant forms of epilepsy submitted to a total therapeutic resection of the corpus callosum
AND BERLUCCHI
(1,5,9,29). The acallosal patients are abnormally slow when they respond with one hand to a visual stimulus in the opposite hemifield, such that the ipsilateral-contralateral difference may be as long as about 20 ms in callosal agenesis and more than 60 ms after callosal section. These long ipsilateral-contralateral differences most probably reflect a slow interhemispheric transfer of information along multisynaptic extracallosal brainstem pathways (1,5). As in normal subjects, they are unaffected by the uncrossed or crossed placement of the hands (1,20). In the study reported here, we have assessed whether the abnormal ipsilateral–contralateral differences seen in acallosal subjects, much greater than those seen in normal subjects, can affect the RT pattern in the typical spatial stimulus–response compatibility paradigm described above. Earlier investigations have revealed typical effects of spatial stimulus–response compatibility in subjects with callosal defects (16,30), but with a single exception (1OA)RT assessments have usually been limited to the uncrossed-hands condition, so that it is not known whether acallosals, like normal controls, exhibit a hand placement effect, and whether this effect, if present, adds to or interacts with the compatibility effect and possibly the neural pathway effect. To test these possibilities, we have studied two subjects who had undergone surgical callosal sections, one complete and the other limited to the anterior twothirds of the structure, and two subjects with total callosal agenesis, and have compared their data with those collected in eight normal controls performing the same task. METHODS
Subjects
Male subject M.E., born in 1970,was submitted to a complete section of the corpus callosum at the Neurosurgical Institute of the Catholic University in Rome (Prof. G. F. Rossi). Callosotomy was performed in two stages (Februa~ and June 1989) in an attempt to control a form of post-traumatic epilepsy with complex partial seizures and secondary generalization which had proved totally resistant to pharmacological therapy as well as to a removal of a focus in the right prefrontal cortex. Callosotomy has resulted in a marked favorable change in both severity and frequency of the seizures. Pharmacological treatment with Phenobarbital and Phenytoin has been continued throughout the postoperative period. At the times of testing for RT (April 1990,December 1991), standard clinical examinations revealed a stationary condition with no neurological deficits, except for a severe lefthand ideomotor dyspraxia in verbal command (but not on imitation), a left-hand anemia and alexia in the left hemifield. The completeness of the callosal section and the integrity of the anterior commissure have been confirmed by magnetic resonance imaging (MRI). I.D., also born in 1970,is another epileptic male patient who was submitted to resection of the anterior two-thirds of the corpus callosum in May 1990, also at the Neurosurgical Institute of the Catholic University in
SPATIAL COMPATIBILITY
IN ACALLOSALS
Rome (Prof. G. F. Rossi). Also in this case, the extent and location of the callosal section have been confirmed by MRI. At the time of testing for the present study (April 1991),he showed no apparent sign of interhemispheric disconnection. P.M. and R.B., two male subjects born, respectively, in 1961 and 1976, have been diagnosed by MRI as congenitally lacking the corpus callosum. At the time of testing, they were free from major neurological symptoms, and appeared to have normal intelligence; the only clear sign of interhemispheric disconnection exhibited by these subjects was an increase of simple RT of each hand to flashes in the contralateral visual field. More detailed information about these four subjects with callosal defects can be found in Aglioti et al. (l), Tassinari et al. (32). Control subjects were eight male students or staff members of the University of Verona, ranging in age from 23 to 34 years. Four were experienced with RT tasks, and four were not. Apparatus
Testing was carried out in a partially sound-proofed cubicle where the subject sat facing a horizontal perimeter 57-cm in radius attached to a white screen. The screen and the perimeter were illuminated from above at a luminance of 0.15 cd/m2. Each subject positioned his head in a forehead- and chin-rest located at the centre of curvature of the perimeter. Two solid state miniature bulbs (TIL 222) were fastened to the perimeter, one on the left and the other on the right of a central mark for fixation. From the patient’s viewpoint, the angular distance between the fixation mark and the bulb on each side was 1OO.Each bulb could be lighted individually by a 5-ins square pulse of current, producing a gallium phosphide green flash with the same duration and an intensity of about 1000pcd. The response devices were two button-keys placed at 20 cm from the midline, one on the right and the other on the left, within a comfortable reaching distance from the subject’shands. Each button-key was mounted on the top of a brass cylinder; the subject clutched each cylinder with the hand on the same or the opposite side and positioned the tip of each thumb on the button-key; the required response consisted of pressing one or the other key by flexing the corresponding thumb. During testing, subjects wore a headset which delivered a standard auditory warning signal consisting of a 1000Hz tone with a suprathreshold but comfortable intensity. Procedure
Subjects were first accustomed to the experimental situation by running several practice trials, and then underwent formal testing over two or more days. The two main independent variables, namely the relationship between the stimulus side and the response side and the uncrossed or crossed placement of the hands, were varied orthogonally, giving rise to four combinations: compatible stimulus–response pairings with uncrossed placement of the hands; incompatible stimulus–response pairings with uncrossed placement of the
hands; compatible stimulus–response pairings with crossed placement of the hands; and incompatible stimulus–response pairings with crossed placement of the hands. In each subject, each of these four combinations was tested in a block of 50 trials, during which the light stimulus appeared 25 times on the right and 25 times on the left in a random, unpredictable order. In the two blocks with compatible stimulus–response pairings, subjects had to press the right key in reaction to right flashes and the left key in reaction to left flashes. In one of these blocks, the hands were held uncrossed, so that the right hand pressed the right key and the left hand pressed the left key; in the other block, the hands were held crossed at the elbow, so that the left hand pressed the right key and the right hand pressed the left key. In the remaining two blocks, stimulus–response pairings were incompatible, so that subjects had to press the left key in reaction to right flashes and the right key in reaction to left flashes; the placement of the hands was uncrossed in one block and crossed in the other block. The order of the four different blocks varied from subject to subject: it was counterbalanced across subjects in the normal group, whereas the subjects with callosal defects were each assigned to a sequence chosen at random. Subject M.E. also performed a second, different sequence, so that he yielded twice as many RTs as the other subjects. Each trial began with a warning auditory signal delivered via the headphone set, followed after an interval varying randomly between 1 and 3s by a right or left flash. Subjects were instructed to fixate on the central mark upon hearing the warning signal and to maintain fixation until after pressing the appropriate key. Eye position and maintenance of fixation were monitored by television, and trials with failures to fixate were aborted and repeated. On each trial, the time elapsing between flash onset and key-pressing response was electronically measured to the nearest millisecond.The presentation of the flashes and the recording of the RT data were automatically controlled by a computer which rejected RTs of responses made with the wrong hand or with both hands, as well as RTs shorter than 150ms or longer than 989 ms. Rejected trials were replaced later in the sequence until obtaining acceptable RTs for all trials. RESULTS
Table 1 shows mean RTs for the relevant comparisons between compatible–incompatible field–key combinations, uncrossed–crossed hand–key combinations, and ipsilateral-contralateral hand–key combinations for the control group as well as for the four subjects with callosal defects. The differences between members of each pair were analysed statistically by assessing the significance of the appropriate two-way interactions in separate ANOVAs having Hand, Field and Key as main factors. The ANOVA for the normal group followed a repeated measurement design with subjects considered as a random variable, each subject contributing a mean score for each level of the three main factors; the ANOVAs for each experimental subject were performed on raw data.
AGLIOTI, TASSINARI AND BERLUCCHI TABLE 1 DIFFERENT COMBINATIONS BETWEEN HANDS, KEYS AND HEMIFIELDS FOR THE CONTROL GROUP AND FOR THE DIFFERENT SUBJECTS WITH CALLOSAL DEFECTS Field–Key Compatible Incompatible Controls M.E. I.D. P.M. R.B.
309.3 401.9 92.6 @
Hand–Key Uncrossed Crossed 329.1 382.1 53.0 (p
All subjects, whether normal or acallosal showed the typicaleffects:RT was significantlyshorter with compatible than with incompatible spatial stimulus-response combinations, and with uncrossed than with crossed hand placement. The hand–field interaction was insignificantin all cases, except in the complete callosotomy subject M.E. and the complete agenesis subject R.B., both of whom showed a large and significant advantage for the hand ipsilateral to the stimulus. Figures 1–5 show the ordering of RTs as a function of field–hand compatibility and hand placement for the control group and for each of the experimental subjects. In all cases, except M.E. and R.B., such ordering followed the typical pattern compatible uncrossed < compatible crossed < incompatible uncrossed < incompatible crossed. M.E. and R.B. showed a different pattern in so far as the compatible crossed condition yielded a longer RT compared to the incompatible uncrossed condition. This effect appears to be due to the prevalence in the former condition of the ipsilateral-contralateral difference over the compatible–incompatible difference. The trends shown by M.E. and R.B. were present also in acallosal subject P.M., but were not as pronounced as in the other two subjects with total callosal defects, and the hand–field interaction did not reach significance, although there was a large difference in favor of ipsilateral over contralateral RTs. DISCUSSION
previous keeping “ (2&l,14,21,24,25,2~!~), th~~~sent results in ~%% subjects bear out the separate effects from the two main independent variables: the effect from the compatibility between locus of stimulusand locus of response, and the effect from the hand placement, both of which are best attributed to the ways in which the stimulus and response sets are mentally coded, and to the ease with which a stimuluscode is translated into a response code. The selection of the key ipsilateral to the flash and the selection of the key on the side of the body attached to the responding limb appear to be naturally easier than
Field–Hand Ipsilateral Contralateral 358.3 352.9 -5.4 (n.s.) 625.4 684.5 59.1 (p
the corresponding converse selections. The occurrence of the two effects is attested, respectively,by the significance in the ANOVA of the field–key interaction and the hand–key interaction. The additivity of the two effects is suggested by the gradual increase of RT from the compatible uncrossed condition to the incompatible crossed condition, as shown in Fig. 1. By contrast, as in earlier investigations(2,4,7,8,14,34),the results from our normal group offered no evidence for the occurrence of systematic RT effects from the presumed difference in length between neural pathways for visuomotor integration within and across hemispheres. The hand–field interaction was completely insignificantin the ANOVA and the mean RT of contralateral responses was insignificantlyshorter than the mean RT of ipsilateral responses;in all subjects,right-hand responses were not significantlyfaster for right compared to left flashes,and left-hand responses were not significantlyfaster for left compared to right flashes. A comparable pattern of results was obtained in patient I.D. with an anterior callosal section, confirming that the integrity of the splenium of the corpus callosum sufficesfor ensuring a virtually normal behavior in most, if not all, visuomotor tasks (6,13,32). In at least two of the three subjects with total callosal defects, the pattern of results differed from that of the normal subjects in so far as RT was influenced not only by the field–key and hand-key interaction, but also by the hand–field interaction. In callosotomized subject M.E. and callosal agenetic subject R.B., the latter interaction was highly significant because stimuli presented in each field evoked faster responses from the anatomically ipsilateral than contralateral hand, independent of the spatial stimulus–response compatibility as well as of the uncrossed or crossed placement of the hands. As a result, the ordering of RTs of the four combinations between stimulus–response compatibility and hand placement was different from that of the normal subjects: in both M.E. and R.B., RT of the compatible crossed combination was as long as or longer than the RT of the incompatible uncrossed and incompatible crossed combinations. In the other acallosal subject P.M., the hand–field interaction was
SPATIAL COMPATIBILITY
IN ACALLOSALS P. M.: C8110wlagmds
Cmlrob(n=B) 9--
m
I
C
W
4
4 .
+-
.?
!
I
3
2
—
2
FIG. 1. Normal controls. Means and standard errors of RTs for the four combinations of compatible and incompatible reactions performed with uncrossed a;d crossed hands. RTs of compatible responses are shown as means of responses with the right hand to the right flash and with the left hand to the left flash in the uncrossed-hands condition, and with the left hand to the right flash and with the right hand to the left flash in the crossed-hands condition. RTs of incompatible responses are shown as means of responses with the left hand to the right flash and with the right hand to the left flash in the uncrossed-hands condition, and with the right hand to the right flash and with the left hand to the left flash in the crossed-hands condition.
U
c
C
FIG. 4. Subject P.M. with callosal agenesis. Means and standard errors of RTs for the four combinations of compatible and incompatible reactions performed with uncrossed and crossed hands. See ~egendto Fig. 1 f& explanations.
R.B.:Callo8d a@imb 5
E m 4
u.
C0mpbbeal1080t0my
-. I
C .
T
Subiect . R.B. with callosal agenesis. Means and standard errors of RTs for the four combinati&s of compatible and incompatible reactions performed with uncrossed and crossed hands. See legend to Fig. 1 for explanations. F
1 .
U
— l
FIG. 2. Subiect M.E. with complete callosotomy. Means and standard erro~s of RTs for the four-combinations of compatible and incompatible reactions performed with uncrossed and crossed hands. See legend to Fig. 1 for explanations.
1.D. : AnbrlIxc91hN010MY m
c~
I
T
m
~
r
4
4 4 4
—
FIG. 3. Subject I.D. with anterior callosotomy. Means and standard errors of RTs for the four combinations of compatible and incompatible reactions performed with uncrossed and crossed hands. See legend to Fig. 1 for explanations.
not significant,but there was, nevertheless, a tendency for a clear-cut advantage for RT of ipsilateral over contralateral responses. Why this acallosal subject did not show exactly the same general pattern of results as that of M.E. and R.B. remains to be determined. A somewhat different pattern of results has also been reported by Di Stefano et al. (lOA) who however did not distinguish between agenetic subjects and a single partially callosotomized patient. These findingssuggest that rules for coding the stimulus and response sets in the performance of a typical spatial compatibility task are similar in the normal and acallosal brain; yet, unlike normal subjects, in acallosal subjects there is a “neural pathway” effect which appears to be additive with those of spatial stimulus–response compatibility and hand placement. This neural pathway effect is similarto that found in the same subjectsin simple RT tasks, in which the other two effects are absent because performance does not require the coding of the positionsof either stimulusor response (l). As already mentioned, the simple RT tasks which yield differences between ipsilateral and contralateral manual responses to literalized flashes in normal
AGLIOTI, TASSINARI
subjects are likely to depend on relatively straightforward cortico-cortical connections, akin perhaps to the transcortical connections that are thought to account for reflex-like responses of the motor cortex to inputs from muscle spindles (22). In acallosal subjects, the main interhemispheric cortical link required for contralateral responses is missing and is substituted by less efficient extracallosal connections which account for the increase in contralateral over ipsilateral RT. The absence of a neural pathway effect in spatial choice RT tasks in normal subjects has been attributed to its masking by the much greater effects from spatial stimulus–response compatibility and hand placement. However, at least according to the additive factor logic, the effect should be unmasked by appropriate subtraction procedures (7). Although previous limited analyses and metanalyses of results for spatial compatibility tasks hinted at the possible occurrence of RT differences in favor of ipsilateral over contralateral RT, these differences were never found to be significant (2,4,7,8,14).The finding of a small insignificant difference in favor of contralateral RTs in the normal subjects of the present experiment would seem to argue that the previous contrary, equally insignificant findings were indeed chance results, and that the normal RT pattern in spatial compatibility tasks is unaffected by the simple neural pathway effect. The neural substrates involved in choosing the literality of the response as a function of the literality of the stimulus are probably quite different from the straightforward connections assumed to mediate simple visuomotor RTs of each hand to a literalized visual stimulus; hence it makes sense that the simplest and most direct anatomical connections between visual fields, cerebral hemispheres and hand muscles can influence simple but not choice RT. Why, then, can a neural pathway effect become apparent in subjects lacking the corpus callosum? Is this simply due to the mere size of the difference between ipsilateral and contralateral RTs in acallosals, which in simple tasks is so much greater than in normal subjects? This possibility would be in agreement with the hypothesis that a neural pathway effect cannot be seen in spatial compatibility tasks in normal subjects
AND BERLUCCHI
because it is too small compared to the effects arising from the coding of stimuli and response and the matching of codes. Since, as discussed above, we tend to reject this possibility, we propose the following alternative interpretation. We hypothesize that in normal subjects performing spatial compatibility tasks, both hemispheres participate in the decisive processes leading to a response to a literalized visual stimulus due to rapid interhemispheric communication via the corpus callosum. Once a decision is reached, the speed of activation of the appropriate motor mechanisms is independent of the side of entry of the literalized stimulus. By contrast, in acallosal subjects, the decisive processes would be largely limited to the hemisphere initially receiving the literalized stimulus, so that the subsequent activation of the appropriate motor mechanisms would vary in speed depending on whether the executive motor command must be issued by the decisional or the opposite hemisphere. In the latter case, a time-consuming interhemispheric transfer would be needed, accounting for the occurrence of the neural pathway effect and its additivity with the other effects arising from the earlier decisional stages. Indirect support for our hypothesis comes from the well established evidence that (1) digital and manual responses are known to be controlled exclusively by the contralateral hemisphere (1,5,6,13,31); (2) each hemisphere of the acallosal brain was shown to be cognizant of both sides of extrapersonal space (10,12,23); and (3) while each hemisphere is able to code and direct attention to specific locations in both the visual hemifield, the speed of responses to stimuli in such locations was found to be, as in the present study, much greater for the contralateral than for the ipsilateral hand (26). A
The studies described in this paper have been carried out with financial support from the ConsiglioNazionale delle Ricerche, the Ministero dell’Universit2t e della Ricerca Scientific e Tecnologica, the Human Frontier Science Programme Organization and the Human Capital Programme of the European Community.
R 1. Aglioti, S.; Berlucchi, G.; Pallini, R.; Rossi, G. F.; Tassinari, G. Hemispheric control of unilateral and bilateral responses to literalized light stimuli after callosotomy and in callosal agenesis. Exp. Brain Res. 95:151–165;1993. 2. Anzola, G. P.; Bertoloni, G.; Buchtel, H. A.; Rizzolatti, G. Spatial compatibility and anatomical factors in simple and choice reaction time. Neuropsychologia 15:295–305;1977. 3. Bashore, T. R. Vocal and manual reaction time estimates of interhemispheric transmission time. Psychol. Bull. 89:352–368; 1981. 4. Berlucchi, G.; Crea, F.; Di Stefano, M.; Tassinari, G. Influence of spatial stimulus–response compatibility on reaction time of ipsilateral and contralateral hand to literalized light stimuli. J. Exp. Psychol. Hum. Perc. Perf. 3:505-517;1977. 5. Berlucchi, G.; Aglioti, S.; Marzi, G. A.; Tassinari, G. Corpus callosum and simple visuomotor integration. Neuropsychologia 33:923-936;1995. 6. Bogen, J. E. Physiological consequences of complete or partial commissural section. In: Apuzzo, M. L. J., ed. Surgery
of the Third Ventricle. Baltimore: Williams & Wilkins; 1987:175-194. 7. Bradshaw, J. L.; Umiltil, C. A reaction time paradigm can simultaneously index spatial compatibility and neural pathways effects: a reply to Levy. Neuropsychologia22:99–101;1984. 8. Brebner, J.; Shephard, M.; Cairney, P. Spatial relationships and S-R compatibility. Acta Psychological37:93-106;1972. 9. Clarke, J. M.; Zaidel, E. Simple reaction times to Iateralized light flashes: Varieties of interhemispheric communication routes. Brain 112:849+70;1989. 10. Corballis, M. E. Visual integration in the split brain. Neuropsychologia33:937-959;1995. lOA. Di-Stefano, M.; Sauerwein, H. C.; Lassonde, M. Influence of anatomical factors and spatial compatibility on the stimuhrsresponse relationship in the absence of the corpus callosum. Neuropsychologia30:177-185;1992. 11. Dutta, A.; Proctor, R. W. Persistence of stimulus–response compatibility effects with extended practice. J. Exp. Psychol.: Learn. Mere. Cogn. 18:801-809;1992.
SPATIAL COMPATIBILITY
IN ACALLOSALS
12. Gazzaniga, M. S. Perceptual and attentional processes following callosal section in humans. Neuropsychologia 25:119–133;1987. 13. Gazzaniga, M. S. Principles of human brain organization derived from split-brain studies. Neuron 17:217-228;1995. 14. Heister, G.; Schroeder-Heister, P.; Ehrenstein, W. H. Spatial coding and spatio-anatomical mapping: Evidence for a hierarchical model of spatial stimulus–response compatibility. In: Proctor, R. W.; Reeve, T. G., eds. Stimulus–response compatibility. Amsterdam: Elsevier; 1990:117-143. 15. Hoptman, M. J.; Davidson, R. J. How and why do the two cerebral hemispheres interact? Psychol. Bull. 116:195–219;1994. 16. Jeeves, M. A. Psychologicalstudies of three cases of congenital agenesis of the corpus callosum. In: Ellinger, E. G. Functions of the corpus callosum. London: J. & A. Churchill; 1965:73–94. 17. Jeeves, M. A. A comparison of interhemispheric transmission times in acallosals and normals. Psychon. Sci. 16:245-246;1969. 18. Milner, A. D. Visual integration in callosal agenesis. In: Lassonde, M.; Jeeves, M. A., eds. Callosal agenesis—a natural split-brain? New York: Plenum Press; 1994:171–183. 19. Milner, A. D.; Lines, C. R. Interhemispheric pathways in simple reaction time to literalized light flash. Neuropsychologia 20:171-179;1982. 20. Milner, A. D.; Jeeves, M. A.; Silver, P. H.; Lines, C. R.; Wilson, J. G. Reaction times to literalized visual stimuli in callosal agenesis: stimulus and response factors. Neuropsychologia 23:323-331;1985. 21. Nicoletti, R.; Anzola, G. P.; Luppino, G.; Rizzolatti, G.; Umiltil, C. Spatial compatibility effects on the same side of the body midline. J. Exp. Psychol., Hum. Perc. Perf. 8:664-673; 1982. 22. Phillips, C. G. Motor apparatus of the baboon’s hand. Proc. Roy. S 173:141–174;1969. 23. Plourde, G.; Sperry, R. W. Left hemisphere involvement in left spatial neglect from right-sided lesions. A commissurotomy study. Brain 107:95–106;1984.
24. Proctor, R. W.; Reeve, T. G., eds. Stimulus–response compatibility. Amsterdam: Elsevier; 1990:89-116. 25. Proctor, R. W.; Dutta, A. Do the same stimulus–response relations influence choice reactions initially and after practice? J. Exp. Psycho].,Learn. Mere. Cogn. 19:922-930;1993. 26. Reuter-Lorenz, P. A.; Fendrich, R. Orienting attention across the vertical meridian: Evidence from callosotomy patients. J. Cognit. Neurosci. 2:232-238;1990. 27. Riggio, L.; Gawryszewski, L.; Umilth, C. What is crossed in crossed-hand effects? Acta Psychological62:89–100;1986. 28. Ringo, J. L.; Doty, R. W.; Demeter, S.; Simard, P. Y. Time is of the essence: A conjecture that hemispheric specialization arises from interhemispheric conduction delay. Cerebral Cortex 4:331-343;1994. 29. Sergent, J.; Myers, J. J. Manual, blowing, and verbal simple reactions to literalized flashes of light in commissurotomized patients. Perc. Psychophys.37:571-578;1985. 30. Smith, K. U. Bilateral integrative action of the cerebral cortex in man in verbal association and sensori-motor coordination. J. E Psychol. 37:367-376;1947. 31. Sperry, R. W.; Gazzaniga, M. S.; Bogen, J. E. Interhemispheric relationships: the neocortical commissures: syndromes of hemispheric disconnection. In: Vinken, P. J.; Bruyn, G.W., eds. Handbook of clinical neurology, vol. 4. Amsterdam: NorthHolland; 1969:4:273-290. 32. Tassinari, G.; Morelli, M.; Berlucchi, G. Interhemispheric transmission of information in manual and verbal reaction-time tasks. Hum. Neurobiol. 2:7745; 1983. 33. Tassinari, G.; Aglioti, S.; Pallini, R.; Berlucchi, G.; Rossi, G. F. Interhemispheric integration of simple visuomotor responses in patients with partial callosal defects. Behav. Brain Res. 64:141-149;1994. 34. Umilta, C. A.; Nicoletti, R. Spatial stimulus–response compatibility. In: Proctor, R. W.; Reeve, T. F., eds. Stimulus–response compatibility. Amsterdam: Elsevier; 1990:89-116.