Exp Brain Res (1996) 112:496-504 �9 Springer-Verlag 1996

David P. Carey �9 Eric L. Hargreaves Melvyn A. Goodale

Reaching to ipsilateral or contralateral targets: within-hemisphere visuomotor processing cannot explain hemispatial differences in motor control

Received: 22 May 1995 / Accepted: 20 June 1996

Abstract Aiming movements made to visual targets on the same side of the body as the reaching hand typically show advantages as compared to aiming movements made to targets on the opposite side of the body midline in the contralateral visual field. These advantages for ipsi- lateral reaches include shorter reaction time, higher peak velocity, shorter duration and greater endpoint accuracy. It is commonly hypothesized that such advantages are re- lated to the efficiency of intrahemispheric processing, since, for example, a left-sided target would be initially processed in the visual cortex of the right hemisphere and that same hemisphere controls the motor output to the left hand. We tested this hypothesis by examining the kine- matics of aiming movements made by 26 right-handed subjects to visual targets briefly presented in either the left or the right visual field. In one block of trials, the subjects aimed their finger directly towards the target; in the other block, subjects were required to aim their move- ment to the mirror symmetrical position on the opposite side of the fixation light from the target. For the three ki- nematic measures in which hemispatial differences were obtained (peak velocity, duration and percentage of movement time spent in deceleration), the advantages were related to the side to which the motor response was directed and not to the side where the target was present- ed. In addition, these effects tended to be larger in the fight hand than in the left, particularly for the percentage of the movement time spent in deceleration. The results are interpreted in terms of models of biomechanical con- straints on contralateral movements, which are indepen- dent of the hemispace of target presentation.

D.R Carey ( ~ ) 1 . E.L. Hargreaves 2 �9 M.A. Goodale Department of Psychology, The University of Western Ontario, London, Ontario, Canada, N6A 5C2

Present addresses: 1 Department of Psychology, Kings College, University of Aberdeen, Old Aberdeen AB24 2UB, Scotland; e-mail: d.carey @abdn.ac.uk 2 Department of Psychology, McGill University, Stewart Building, 1205 Docteur Penfield Avenue, Montreal, PQ, Canada H3A 1B 1; e-mail: erk@ blaise.psych.mcgill.ca

Key words Motor control �9 Hemispace �9 Kinematics �9 Handedness - Interhemispheric transmission �9 Human

Introduction

Because sensory and motor systems in primates are largely dependent on the contralateral side of the brain, many movements require participation of both hemi- spheres. For example, if a visual stimulus is presented in the field contralateral to the hand used to respond to the stimulus, some sort of interhemispheric communication will be required for a successful response. Transmission of information between the hemispheres has been exten- sively studied in split-brain patients (Sergent and Myers 1985; Tassinari et al. 1994), in patients with congenital malformations of the corpus callosum (Di Stefano et al. 1992; Jakobson et al. 1994; Jeeves and Silver 1988; Las- sonde and Jeeves 1994; Rugg et al. 1985) and in neuro- logically intact subjects (Bryden and Bulman-Fleming 1994; Marzi et al. 1991; Rugg et al. 1984). In neurologi- cally intact subjects, the principal means of investigating inter-hemispheric transmission has been the so-called crossed-uncrossed difference in reaction time. The ratio- nale behind such experiments is this: if a visual stimulus is presented in the visual field on the same side as the hand which responds (the so-called uncrossed condi- tion), information about the target does not have to cross the corpus callosum in order to reach motor cortices con- trolling the responding hand, as it would have to do if the stimulus appeared in the contralateral visual field (the "crossed" condition; Fig. 1). Such studies have shown a small but reliable reaction-time advantage for stimuli presented in the same hemifield as the responding hand of about 2-3 ms (Berlucchi et al. 1971; Di Stefano et al. 1980; see Bashore 1981; Marzi et al. 1991; Hoptman and Davidson 1994 for reviews. Stimulus-response compati- bility effects and related theories will not be discussed in detail, but the interested reader should consult Proctor and Reeve 1990).

Fig. la, b Ipsilateral targets are processed in the same hemisphere that processes the initial visual stimulus, a A right-sided target is initially processed in left visual/visuomotor cortices (A). That in- formation can be sent within-hemisphere to left motor/premotor cortex (B). b A left-sided target is initally processed in right visu- al/visuomotor cortices (C). Information for the left motor/premo- tot cortex may cross the callosum in the spleninm (De Lacoste et al. 1985; Degos et al. 1987) and be transferred to homologous re- gions in the left hemisphere (A). The existence of a pathway from C directly to B is unknown, although Johnson et al. (1993) suggest that the visual guidance of reaching is dependent upon pathways from A to B in the macaque

A similar phenomenon has been reported in experi- ments in which subjects actually make a movement to- wards the visual target. Careful quantification of such movements has uncovered hemispatial advantages for movements made to ipsilateral targets in measures such as movement duration, peak velocity, accuracy and con- sistency of the movement trajectory (Carson et al. 1990, 1992, 1993; Chua et al. 1992; van Der Staak 1975; Elliot et al. 1993; Fisk and Goodale 1985; Ingum and Bjork- lund 1994; Levin 1996; Prablanc et al. 1979). Much as in the uncrossed condition described above, such targets are processed initially in the same hemisphere as the motor, premotor and somatosensory cortices with direct connec- tions to the reaching hand. Unlike the crossed-uncrossed studies, however, the subject is required to reach into the hemifield in which the target has appeared. The shorter durations and greater accuracy observed for reaches

497

made into ipsilateral space could be a consequence of more efficient within-hemisphere visuomotor transmis- sion of target information and/or visual feedback from the reaching limb, or they could be related to motor out- put being better organised for movements that do not cross the body midline. Also unlike the differences found in the typical reaction-time experiment, the size of the ipsilateral advantage in reaching is often very large, i.e. 30-50 ms or more in duration and movement onset times (i.e. van Der Staak 1975; Fisk and Goodale 1985; Jeannerod 1987). The simple type of information transfer in reaction-time studies might be mediated by large, fast- conducting myelinated fibres in the callosum (Hoptman and Davidson 1994), and these fibers are relatively rare (Aboitiz et al. 1992). To initiate a spatially appropriate motor response into the contralateral space might require the participation of many more callosal fibres, including the more slowly conducting axons. Whatever the exact mechanisms involved are, it seems unlikely that reaction- time differences of such magnitude result from a simple model of increased transmission time across the corpus callosum.

In these reaching studies, the side of stimulus presen- tation has never been dissociated from the side of the re- quired motor response (except by Chua et al. 1993, where "mirror-image" pointing was used to increase the spatial complexity of an aiming task). The typical ipsilat- eral movement advantages for movement onset, peak ve- locity, duration and accuracy have been explained in two alternative ways. Fisk and Goodale (1985) offered the suggestion that kinematic advantages stem from the fact that initiation and control of movement is primarily the responsibility of the hemisphere contralateral to the limb being used. Therefore, visual information about a target that is in the contralateral hemifield has to cross the cor- pus callosum to reach the motor programming centres necessary to generate and control the movement (see Fig. lb).

Alternatively, the hemispatial effects might not have to do with processing the visual input within-hemisphere per se, but instead be a by~product of some sort of bio- mechanical constraint on contralaterally directed move- ment. For example, the displacement of the centre of mass of the limb will usually be larger for a contralateral movement and could account for some of the observable differences in kinematics. Another possibility is that more muscle groups have to be recruited in order to make a movement across the body midline (such as the pectorals, deltoid and latissimus dorsi; see Happee and Van der Helm 1995). Unfortunately, the few studies that have examined electromyographic (EMG) activity in several muscle groups simultaneously have restricted arm movements to the sagittal plane (e.g. Happee 1992; Happee and Van der Helm 1995), have grouped the data in such a way that hemispatial differences could not be examined (e.g. Bock 1994; Koshland and Hasan 1994) or have used widely varying starting points and endpoints (Karst and Hasan 1991a,b). To date, no study has investi- gated differential muscle activation to examine potential

498

Fig. 2a, b The antipointing task. In half of the trials (b), subjects were required to reach to the mirror symmetrical posi- tion on the opposite side of the fixation point when a target ap- peared

differences in motor programming requirements in ipsi- lateral compared with contralateral movements, identical in distance and direction from the body midline.

Nevertheless, Fisk and Goodale suggest that a biome- chanical explanation of this sort might be too simple. They found that ipsilateral reaches of different lengths showed peak velocity scaling: longer required move- ments resulted in higher peak velocity (thus durations for these different movements remained very similar). Such effects were not seen in the contralateral reaches; instead movement duration increased for longer reaches. Fisk and Goodale (1985) argue that because peak velocity is achieved well before the hand crosses the body midline during a contralateral movement, the lower peak veloci- ties that are commonly observed are unlikely to be a con- sequence of greater mechanical constraints. Bradshaw and colleagues have also argued that biomechanical con- siderations may not be responsible for differences in movements of this sort (Bradshaw et al. 1990; but see Carson et al. 1990; Chua et al. 1992 for alternate views).

Despite these objections, a biomechanical hypothesis has yet to be examined in an experimental fashion. Re- cording from many of the proximal and distal muscles of the arm during movements might reveal differences in the execution of ipsilateral compared with contralateral reaches. As an alternative to that type of complicated EMG study, examination of movement kinematics might reveal whether or not input or output stages of the visuo- motor transformation are crucial for hemispatial differ- ences in movements to visual targets. The present experi- ment does so by dissociating stimulus position from mo- tor response by requiring subjects to point (using the tip of the index finger) to the mirror-symmetrical position on the opposite side of the fixation point from the pre- sented stimulus position ("antipointing", after Guitton et al.'s 1985 "antisaccades", and Chua et al.'s 1993 "mental rotation" pointing task - see Fig. 2). By comparing these antipointing movements with normal target-directed movements, it was possible to tease apart the relative contributions of inter-hemispheric communication and biomechanical constraints to the kinematics of contralat- eral aiming movements. In addition, some investigators have suggested that the left hemisphere-right hand ad- vantages in motor control stem from superior utilization of feedback (e.g. Lomas and Kimura 1976; Todor and Doane 1978). For this reason both hands were tested in

hand-visible and hand-invisible conditions. I f visual and somatosensory feedback comparison is facilitated or en- hanced in the ipsilateral hemifield, then removing visual feedback about hand position might be expected to atten- uate the hemispatial effects. Finally, the performance of left and right hands was examined in order to examine the possibility of any asymmetries in hemispatial effects on movement kinematics, using a larger sample than in the original Fisk and Goodale (1985) study. Particular emphasis in the current study was placed on reaction time, movement time, peak velocity and movement accu- racy - kinematic markers that have often shown advanta- ges for ipsilateral movements.

Materials and methods

Subjects

Subjects were 26 graduates, senior undergraduates and research assistants from the University of Western Ontario. Subjects were male, right-handed (9/9 items performed with the right hand on a modified version of the Edinburgh Handedness Inventory; Oldfield 1971) and had a mean age of 26.5 years (range 22-32 years) at time of testing. For a different analysis, sighting dominance was assessed and several non-right sighters were recruited for partici- pation in the study. Twelve subjects comprised the left/anomalous sighting group, the remaining 14 were right eye-sighters. It should be noted that, owing to the higher proportion of right-sighting sub- jects in the general population (Porac and Coren 1976), to achieve approximately equal sample sizes this sample includes a much higher proportion of non-right sighters than a random sample of right-handers would. However, sighting dominance did not affect any of the main effects or interactions relevant for the present analysis, so will not be discussed further.

Apparatus

Pointing movements were recorded using a WATSMART opto- electronic recording system (Northern Digital; Waterloo, Ontario, Canada). In our laboratory, ten-trial samples of a stationary, infra- red light-emitting diode (IRED) revealed the largest difference of 0.41, 0.86 and 1.03 mm in the x, y and z dimensions (Jakobson and Goodale 1989). For the present study, calibrations from all ses- sions had a mean root square error of less than 2.00 mm sampled from a minimum of 21 of 24 IREDs from a calibration cube.

The visual stimuli were 0.25 ~ light-emitting diodes (LEDs), embedded in a target wedge covered with black, non-reflective speaker cloth (as used by Jakobson and Goodale 1989; Goodale et al. 1990). The wedge was located 32 cm in front of a starting posi- tion (indicated tactually by a thumbtack), and the targets were lo- cated 2 cm above the table surface. The start position was located

10 cm from the edge of the table closest to the subject, placed at the subjects' midline and on the same axis as the central LED. Four lateralized targets were presented to the left and the right of the central LED (which served as a fixation light at the start of each trial). The innermost targets were 3 cm lateral (5.4 ~ to the central LED; the additional three targets in each hemispace were 6 cm apart (16 ~ 27 ~ and 38 ~ from the fixation light).

Procedure

Subjects' heads were positioned as comfortably as possible in a chin-rest and they wore a black glove with two attached IREDs (at the tip and the base of the index finger). The black glove served to minimize potential visual feedback from the moving hand in hand- invisible conditions. Eight calibration trials were collected in which subjects made pointing movements to continuously illumi- nated LEDs and were permitted to adjust the final position of the index fingertip until they were satisfied that they had perfectly oc- cluded the LED. The experimenter then collected a 100-Hz 1.5-s sample of the finger position before collecting the next calibration trial.

Subjects reached in four 40-trial blocks: once in each combina- tion of hand visible/invisible and pointing/antipointing. For the an- tipointing trials, subjects demonstrated their comprehension of the task by showing what would be correct movements to positions manually indicated by the experimenter. Order of blocks was com- pletely randomized. At the beginning of each block, subjects re- ceived eight practice trials prior to data collection. During each block, subjects were frequently reminded of the appropriate task for that block. Subjects were also instructed to report any instruc- tional set errors (i.e. antipointing during a pointing block or vice versa) for subsequent deletion of those trials. Subjects used one hand in each session, and sessions were separated by at least 2 days. In both blocks, target presentation was restricted to a 400-ms period, in order to minimize potential visual feedback during hand-invisible reaching (i.e. if the moving hand passed over the target or occluded it at movement end).

Results

Data analysis

After data collection, raw WATSMART files were con- verted to a three-dimensional format and filtered with a 7-Hz, second-order low-pass Butterworth filter (as used by Goodale et al. 1990; Jakobson and Goodale 1989). This type of filter removes high-frequency noise from the data in a two-pass procedure, which preserves the po- sition of the signal in the time domain and has a sharp cut-off of frequencies greater than 7 Hz. Filtered files were used to compute peak velocity (centimetres per sec- ond), movement onset time and movement duration (both in milliseconds), and the percentage of the total reach spent decelerating (movement duration minus time to peak velocity/movement duration x 100).

Accuracy of the reaches were estimated with three different measures, all based on comparisons of index finger IRED endpoints during experimental trials with the coordinates of index finger IRED endpoints specified in calibration trials. For antipointing trials, accuracies were calculated relative to the symmetric target position on the opposite side of fixation. For this analysis, means were collapsed across all four targets on the same side of the fixation light. Repeated-measures ANOVAs with Hand, Hand visibility, Task, Sighting dominance (right

499

eye dominance vs anomalous dominance) and Side (left- sided vs right-sided target presentation) were performed.

The usual differences between hand-visible and hand- invisible reaching were obtained. For the purposes of the present report, however, only those analyses relevant to ipsilateral-contralateral differences in movement kine- matics will be considered. I f advantages for ipsilateral targets were dependent to any extent on factors related to the arm movement rather than target position per se, then an interaction between Hand, Task and Side should have occurred, such that for antipointing the usual direction of the difference (e.g. "ipsilateral faster than contralateral") would have been reversed (or attenuated if side of stimu- lus and side of motor response both play a role in pro- ducing the hemispatial differences).

Accuracy

Subjects showed equivalent accuracy in both hemispaces (although they were somewhat less accurate on the anti- pointing task overall). No significant ipsilateral advanta- ges were found in any measure of terminal accuracy in the pointing task and thus could not be attenuated or re- versed in antipointing.

Movement onset time

The HandxTaskxSide interaction was significant (F1,24=5.47, P<0.03), but the data from pointing trials did not provide any evidence for ipsilateral target-move- ment onset advantages, therefore it is impossible in this dataset to see any evidence of attenuation of such advan- tages in the antipointing trials. This result differs from that found by Chua et al. (1992), who found that ipsilat- eral movements had shorter movement onset times than contralateral movements, in pointing movements as well as in "mirror" movements. Nevertheless, movement on- set time seems to be a variable where ipsilateral advanta- ges are not always obtained (cf. Carson et al. 1992; Fisk and Goodale 1985, experiment 2 - except in the condi- tion where central fixation was maintained throughout the movement; M. Harvey, personal communication).

Movement duration

Movement durations provide very strong support for the importance of hemispace of the movement rather than target position being the crucial variable (Fig. 3). The Hand by Task by Side interaction (F1,24=398.05, P<0.0001) and post hoc tests revealed that antipointing reversed the normal movement duration advantages seen in pointing to ipsilateral targets. As described by Chua et al. (1992), for both hands, pointing into the ipsilateral side of space produced significantly shorter movement durations than pointing into contralateral space (P<0.01, for all). In fact, the ordering of the means was exactly as

500

500

450

Mean Duration

400 (msec)

350

300

500

450

Mean Duration 400

(msec)

350

300

Right H a n d

Left Hand

Left Right

Side of Stimulus

= = Pointing

Antipointing

Fig. 3 Mean movement duration as a function of Hand, Task and Side of stimulus. Ipsilateral movements are completed more quickly, regardless of target hemispace. Error bars SEM

200

175

Mean Peak Velocity 150 (cm/sec)

i25

100

200

175

Mean Peak Velocity 150

(cm/sec)

125

100

Right Hand

Lef t H a n d

i

Left Right

Side of Stimulus

.- �9 Pointing

Antipointing

Fig. 4 Mean peak velocity as a function of Hand, Task and Side of stimulus

predicted for ipsilateral side of movement and right-hand advantages. Mean differences in movement duration be- tween ipsilateral and contralateral movements are pro- vided in Table 1. The four-way interaction between Hand, Task side and Hand visibility also reached signifi- cance (F1,24=13.57, P<0.001), but post hoc analyses did not provide an unambiguous interpretation. Nevertheless, inspection of the data in Table 1 makes it clear that there is little evidence for attenuation of hemispatial effects on movement duration in hand-invisible conditions. All of these mean differences between ipsilateral and contralat- eral reaches are significant at the .01 level.

Peak velocity

Movements made towards the contralateral side of the body were different from movements into the ipsilateral hemispace in peak velocity. A significant Hand by Task by Side interaction (F1,24=111.08, P<0.0001; Fig. 4) and associated post hoc tests showed that pointing in ipsilat- eral hemispace produced higher peak velocities, inde- pendent of stimulus hemifield (Q=11.45, P<0.01, for right-handed pointing; Q=10,24, P<0.01 for right-hand- ed antipointing; Q=3.81, P<0.05, for left-handed point- ing; Q=4.30, P<0.05, for left-handed antipointing). The four-way interaction between Hand, Task, Side and Hand visibility did not reach significance (F1,24=0.09, P<0.77), again suggesting that hemispatial differences in move- ments do not depend on the presence of visual feedback of the reaching limb.

Table 1 Mean difference (milliseconds) in movement duration (ipsilateral-contralateral reach) as a function of Hand, Hand visi- bility and Task

Hand

Left Right

Point Antipoint Point Antipoint

Hand Visible -51.6 -45.1 -49.2 -45.8 Hand Invisible -72.6 -56.8 -59.9 -48.7

Percentage decelerating

A significant Task by Side (F1,24=5.29, P<0.03) interac- tion may be explained by a tendency to spend more of the movement decelerating when left-sided movements were made compared with right-sided movements. As in the other measures where hemispatial effects were found in pointing, antipointing reversed the usual pattern of de- celeration duration (Hand by Task by Side; F1,24=34.99, P<0.0001). The ordering of the cell means is perfect for contralateral movements having shorter deceleration phases than ipsilateral movements and the right hand spending less time decelerating than the left hand (Fig. 5). Newman-Keuls results suggest that movements into contralateral hemispace for the right hand had sig- nificantly shorter deceleration periods than all other con- ditions (P<0.05). The Hand by Task by Side by Hand visibility interaction was not significant (F1,24=0.03, P<0.86), again suggesting that these hemispatial effects are independent of the availability of visual feedback of the reaching limb.

Mean Percentage Decelerating

Mean Percentage Decelerating

64

60

56

i

52 ~

64

60

56

i i

52 f

Right Hand

Left Hand

-" -" Pointing

o--~ Antipointing

Left Right

Side of Stimulus

Fig. 5 Mean percentage of the movement spent in deceleration as a function of Hand, Task and Side of st imulus

Overview

In the case of deceleration duration, movement duration and peak velocity, advantages for ipsilateral targets were obtained in the pointing task and for the contralateral tar- gets in the antipointing task. When hemispatial effects were not equivalent in the two hands, they were always more obvious in the right hand (i.e. peak velocity and percentage decelerating). These effects were not depen- dent on visibility of the reaching limb.

Discussion

By comparing antipointing movements with pointing movements, we were able to demonstrate that all of the apparent advantages in ipsilateral space could be ex- plained by the biomechanical characteristics of the movement rather than the location of the target. These results are consistent with the prediction made by Fisk and Goodale (1985) that ipsilateral spatial advantages are related to side of motor response rather than side of stimulus presentation. In a second study, they showed that the differences were hemispatial and not hemireti- nal. The present study goes further by showing that tar- get hemispace is largely irrelevant; in fact we predict that in experiments where subjects reach (without any target) into one hemispace for a block of trials, and then into the other for a second block, the same differences in reach kinematics across the hemispace would be found as seen in this experiment.

These data suggest that models of ipsilateral-contra- lateral movement differences based on within-hemi- sphere visual input to the appropriate motor control cen- tres are not tenable explanations for this phenomenon.

501

An alternative to within compared with between hemi- sphere models has been suggested (Jeannerod 1988; see also van Der Staak 1975). This argument is largely based on anatomical experiments in rhesus macaques, which have provided strong evidence for proximal muscles be- ing controlled by the ipsilateral hemisphere (Brinkman and Kuypers 1972; Haaxma and Kuypers 1975; Law- rence and Kuypers 1968a,b). If this anatomical story is also true in Homo sapiens, then the differences between ipsilateral and contralateral movements might be expli- cable by greater ipsilateral cortical control of contralater- al movements. Of course, this hypothesis requires the further assumption that ipsilateral (or perhaps bilateral) control of a muscle results in less accurate, poorly coor- dinated movement (Jeannerod 1988; also see Jakobson et al. 1994; Muller et al. 1991).

A great deal of recent evidence, however, suggests that the Kuypers story may not be as clear-cut as most text- book accounts would suggest. First, Savaki et al. (1993) have positron emission tomography (PET) data that sug- gests that split-brain monkeys can cross-cue a "blind" hemisphere with gross information about target location, which can be sufficient to guide whole-arm movements but not fine-tune movements of the fingers and wrist. Therefore, the monkeys in Kuypers' original studies might have had spared arm movement for reasons other than ipsilateral control of the proximal musculature. Sec- ond, several recent investigations using PET, magnetic and anodal stimulation and other physiological tech- niques are suggestive of a greater degree of contralateral control for the proximal musculature in our species (Colebatch et al. 1990, 1991; Matelli et al. 1993; Triggs et al. 1994), or at least a more complicated account than an exclusive monosynaptic pathway to only the contralat- eral distal muscles (Basu et al. 1994; Burke et al. 1994; Tanji et al. 1988; Trope et al. 1987; Wasserman et al. 1994). Third, although many would argue that contralat- eral paresis and apraxia respect the proximal-distal dis- tinction, newer evidence suggests otherwise (Colebatch et al. 1986; Lakke et al. 1984; Poeck et al. 1982). Fourth, the disruption of finger but not whole-arm movements by lesions that disconnect parietal cortex from frontal targets (i.e. Haaxma and Kuypers 1975) may not be a conse- quence of differences in the contralateral control of the proximal muscles in motor cortex per se. Glickstein has suggested that the lesions made by Brinkman and Kuy- pers (1972) may have produced the effects by disconnect- ing pontocerebellar circuits from the parietal cortex, rath- er than by removing visual inputs to frontal cortex re- sponsible for finger movements (Glickstein 1980, 1990). Recent neurophysiology suggests a role in hand move- ment control for the anterior region of the intraparietal sulcus (Taira et al. 1990; Gallese et al. 1996), which may have also been damaged by the lesions intended to deaf- ferent frontal cortex. Further investigation of the proxi- mal-distal distinction in Homo sapiens and nonhuman primates is required.

If interhemispheric versus intrahemispheric control models and ipsilateral-proximal control models are not

502

the most likely explanations for kinematic differences in movements which vary across the hemispace, then what sort of model remains? Fisk and Goodale (1985) were sceptical of biomechanical explanations, primarily be- cause their obtained peak velocity differences did not scale in a linear fashion across contralateral space. (Car- son and colleagues disagree: they have argued that dif- ferences across the hemispace in the patterns of move- ment duration for near versus far targets strongly suggest the inclusion of biomechanical factors; Carson et al. 1990; Chua et al. 1992). However, some recent physio- logical/kinesiological analyses do suggest that a less neuropsychological explanation of ipsilateral/contralater- al differences in movement kinematics may be called for.

A strong candidate has been suggested by Gordon et al. (1994), who have recently reported that the variance of inertial forces at the hand, which depends on move- ment direction, can account for many differences in movement kinematics, including movement duration, peak velocity, acceleration characteristics and terminal accuracy. They argue that hand movements in directions perpendicular to the axis of the upper limb have higher inertial loads than movements made in directions parallel to the axis of the upper arm (see also Begbie 1959; Corrigan and Brogden 1949). Of course, in typical aim- ing movement studies, with the hand on a central starting position close to the body, ipsilateral movements of the hand are much more parallel to the upper arm axis than contralateral movements. Unfortunately the angular dif- ferences between the targets in one hemispace were fair- ly small in the current study, so effects related to target eccentricty were not analyzed, as they have been in past studies by Carson and colleagues (Carson et al. 1990; Chua et al. 1992). Nevertheless, our ipsilateral and con- tralateral groups of movements provide a good approxi- mation of "parallel" and "perpendicular" movement cat- egories, respectively.

The simulations of the inertial forces in the study by Gordon et al. (1994) were well matched to actual data obtained from two-dimensional movements in nine hu- man subjects. For example, the higher peak velocities and accelerations found for ipsilateral movements were well predicted by the estimated inertial differences be- tween the two types of movement: Although the move- ments in the study by Gordon et al. (1994) were two-di- mensional, the authors argue persuasively that their re- sults would extend to the three-dimensional case (see also Flanagan et al. 1993; Flash 1987; Ghez et al. 1993; Hogan 1985; Jordon et al. 1994; Turner et al. 1995). In- deed, a study by Koshland and Hasan (1994) suggests that muscle activation patterns for reproducing an end- point in three-dimensional space tend to minimize the differential inertial effects at the wrist that are produced by torques in the proximal parts of the arm. These data suggest that the central nervous system attempts to com- pensate for inertial differences for different arm move- ments, but that the compensation is not complete. If the model by Gordon et al. (1994) can indeed account for the type of hemispatial effects seen in this study, then

varying the degree of "perpendicularity" of the two parts of the arm with different targets should produce predict- able trends in movement kinematics. One of us (D.RC.) is currently investigating this possibility.

Of interest to neuropsycholgists in this regard is that sensorimotor and perhaps even attentional systems seem well acquainted with hemispatial differences in these classes of movement, whatever the specific biomechani- cal (or neurological) origins of those differences might be. For example, Fisk and Goodale (1985) found a "yok- ing" of saccadic latencies to arm movement latencies in a reaching task: an eye movement to a particular target was delayed by as much as 50 ms if that target was about to be acquired with a contralateral limb movement (in spite of the conjugacy of eye movements - eye move- ments cannot be ipsilateral or contralateral the way arm movements can). There are also rather intriguing sugges- tions in the literature suggesting that right-handed sub- jects may attend more frequently to right space and/or their right hand in unimanual or bimanual tasks (Brad- shaw et al. 1989; Fagot et al. 1994; Honda 1984; Peters 1987; Rizzolatti et al. 1995; Tipper, et al. 1992). In non- human primates, there is some evidence to suggest stronger correlations between the preferred movement directions of motor cortex ceils and the actual direction of movement in ipsilateral rather than contralateral hemi- space (Caminiti et al. 1990).

Finally, the hemispatial differences in the percentage of the movement spent in deceleration have been report- ed for the left but not the right hand in right-handed sub- jects (Carson et al. 1993). In the present study, ipsilateral increases in deceleration duration were seen in the right hand only. Carson et al. (1993) varied task instructions and found their effects only in a condition where accura- cy (and not speed) was emphasized. Roy et al. (1994), like us, report longer deceleration durations in the right hand of their right-handed subjects. In two other aiming movement tasks in our laboratory we also have seen in- creased deceleration duration in the right hemispace in right-handed movements. The significance of this effect is as yet unknown and is not present in all right-handed subjects.

In summary, this experiment provides unambiguous evidence of the origins of hemispatial differences in the production of aiming movements. Contrary to previous suggestions, intra- and inter-hemispheric processing de- mands do not explain differences in the kinematics of reaches across the hemispace. Rather, complex biome- chanical demands may account for slower and less accu- rate contralateral reaching. Further study is necessary to determine whether or not inertial forces are the principal cause, or whether models which incorporate differences in the neural control of the proximal and distal muscula- ture are ultimately responsible for these kinematic ef- fects.

Acknowledgements We would like to thank A. David Milner and Philip Surette for computer programming, and Lorna Jakobson, John Orphan, Rick Cornwall, Duncan Roland and Michael Burt for excellent technical assistance. Our thanks also to Richard Car-

503

son and a second, anonymous reviewer for their helpful comments on an earlier draft of this manuscript. This research was supported by grant MA-7269 from the Canadian Medical Research Council to M.A. Goodale. Most of the research was carried out while D.EC. held a studentship from the Canadian Medical Research Council. Much of the data analysis and manuscript preparation was supported by a Teaching/Research Fellowship to D.P.C. from the University of St. Andrews, St. Andrews, Scotland.

References

Aboitiz F, Scheibel AB, Fisher RS, Zaidel E (1992) Individual dif- ferences in brain asymmetries and fiber composition in the hu- man corpus callosum. Brain Res 598:154-161

Bashore TR (1981) Vocal and manual reaction time estimates of inter-hemispheric transmission time. Psychol Bull 89:352-368

Basu AR Turton A, Lemon RN (1994) Activation of ipsilateral up- per limb muscles by transcranial magnetic stimulation in man (abstract). J Physiol (Loud) 479:144

Begbie GH (1959) Accuracy of aiming in linear hand-movements. Q J Exp Psychol 11:65-75

Berlucchi G Crea, F Di Stefano M, Tassinari G (1971) Influence of spatial stimulus-response compatibility on reaction time of ipsilateral and contralateral hand to lateralized light stimuli. J Exp Psychol Hum Percept Perform 3:505-517

Bock O (1994) Scaling of joint torque during planar arm move- ments. Exp Brain Res 101:346-352

Bradshaw JL, Bradshaw JA, Nettleton NC (1989) Direction and location of movement in kinesthetic judgments of extent. Neu- ropsychologia 27:1139-1151

Bradshaw JL, Bradshaw JA, Nettleton NC (1990) Abduction, ad- duction and hand differences in simple serial movements. Neuropsychologia 28:917-931

Brinkman J, Kuypers HGJM (1972) Split-brain monkeys: cerebral control-of ipsilateral and contralateral arm, hand and finger movements. Science 176:536-539

Bryden MR Bulman-Fleming MB (1994) Laterality effects in nor- mal subjects: evidence for interhemispheric interactions. Be- hay Brain Res 64:119-129

Burke D, Gracies JM, Mazevet D, Meunier S, Pierrot-Deseilligny E (1994) Non-monosynpatic transmission of the cortical com- mand for voluntary movement in man. J Physiol (Lond) 480: 191-202

Caminiti R, Johnson PB, Burnod Y, Galli C, Ferraina S (1990) Shift of preferred directions of premotor cortical cells with arm movements performed across the workspace. Exp Brain Res 83:228-232

Carson RG, Chua R, Elliott D, Goodman D (1990) The contribu- tion of vision to asymmetries in manual aiming. Neuropsy- chologia 28:1215-1220

Carson RG, Goodman D, Elliott D (1992) Asymmetries in the dis- crete and pseudocontinuous regulation of visually guided reaching. Brain Cogn 18:169-191

Carson RG, Goodman D, Chua R, Elliott D (1993) Asymmetries in the regulation of visually guided aiming. J Mot Behav 25: 21-32

Chua R, Carson RG, Goodman D, Elliott D (1992) Asymmetries in the spatial localization of transformed targets. Brain Cogn 20:227-235

Colebatch JG, Gandevia SC (1989) The distribution of muscular weakness in upper motor neuron lesions affecting the arm. Brain 112:749-763

Colebatch JG, Rothwell JC, Day BL, Thompson PD, Marsden CD (1990) Cortical outflow to proximal arm muscles in man. Brain 113:1843-1856

Colebatch JG, Deiber M-R Passingham RE, Friston KJ, Frackowiak RSJ (1991) Regional cerebral blood flow during voluntary arm and hand movements in human subjects. J Neurophysiol 65:1392-1401

Corrigan RE, Brogden WJ (1949) The trigonometric relationship of precision and angle of linear pursuit-movements. Am J Psychol 62:90-98

Degos JD, Gray F, Lourarn F, Ansquer JC, Poirer J, Barbizet J (1987) Posterior callosal infarction: Clinicopathological corre- lations. Brain 110:1155-1171

De Lacoste MC, Kirkpatrick JB, Ross ED (1985) Topography of the human corpus callosum. J Neuropathol Exp Neurol 44: 579-591

Der Staak van C (1975) Intra- and interhemispheric visual-motor control of human arm movements. Neuropsycholgia 13:439- 448

Di Stefano M, Morelli M, Marzi CA, Berlucchi G (1980) Hemi- spheric control of unilateral and bilateral movements of proxi- mal and distal parts of the arm as inferred from simple reac- tion time to lateralized light stimuli in man. Exp Brain Res 38:197-204

Di Stefano M, Sanerwein HC, Lassonde M (1992) Influence of an- atomical factors and spatial compatibility on the stimulus re- sponse relationship in the absence of the corpus callosum. Neuropsychologia 30:177-185

Elliot D, Roy EA, Goodman D, Carson RG, Chua R, Maraj BKV (1993) Asymmetries in the preparation and control of aiming movements. Can J Exp Psychol 47:570-589

Fagot J, Lacreuse A, Vauclair J (1994) Hand-movement profiles in a tactual-tactual matching task: effects of spatial factors and laterality. Percept Psychophys 45:347-355

Fisk JD, Goodale MA (1985) The organization of eye and limb movements during unrestricted reaching to targets in ipsilater- al and contralateral space. Exp Brain Res 60:159-178

Flanagau JR, Ostry DJ, Feldman AG (1993) Control of trajectory modifications in target-directed reaching. J Mot Behav 25: 140-152

Flash (1987) The control of hand equilibrium trajectories in multi- joint arm movements. Biol Cybern 57:257-274

Gallese V, Fadiga L, Fogassi L, Luppino G, Murata A (1996) A parietal-frontal circuit for hand grasping movements in the monkey: evidence from reversible inactivation experiments. In Thief P, Karnath H-O (eds) Parietal lobe contributions to ori- entation in 3-D space. Springer, Berlin Heidelberg New York

Ghez C, Gordon J, Ghilardi MF (1993) Programming of extent and direction in human reaching movements. Biomed Res [Suppl 1] 14:1-5

Glickstein M (1980) Posterior parietal cortex and visual control of the hand. Behav Brain Sci 3:503

Glickstein ME (1990) Brain pathways in the visual guidance of movement and the behavioral functions of the cerebellum. In: Trevarthen E (ed) Brain circuits and the functions of the mind: essays in honor of Roger W. Sperry. Cambridge University Press, New York, pp 157-167

Goodale MA, Milner AD, Jakobson LS, Carey DP (1990) Kine- matic analysis of limb movements in neuropsychological re- search: subtle deficits and recovery of function. Can J Psychol 44:180-195

Gordon J, Gilhardi MF, Cooper SE, Ghez C (1994) Accuracy of planar reaching movements. II. Systematic errors resulting from inertial anisotropy. Exp Brain Res 99:112-130

Graziano SA, Yap GS, Gross CG (1994) Coding of visual space by premotor neurons. Science 266:1054-1057

Guitton D, Buchtel HA, Douglas RM (1985) Frontal lobe lesions in man cause difficulties in supressing reflexive glances and in generating goal-directed saccades. Exp Brain Res 58:455-472

Haaxma H, Kuypers HGJM (1975) Intrahemispheric cortical con- nections and visual guidance of hand and finger movements in the rhesus monkey. Brain 98:239-260

Happee R (1992) Goal directed arm movements. I. Analysis of EMG records in shoulder and elbow muscles. J Electromyogr Kinesiol 2:165-168

Happee R, Van tier Helm FCT (1995) The control of shoulder muscles during goal directed movements, an inverse dynamic analysis. J Biomech 28:1179-1191

Hogan (1985) The mechanics of mulit-joint posture and move- ment control. Biol Cybern 52:315-331

Honda H (1984) Functional between-hand differences and outflow eye position information. Quart J Exp Psychol 36:75-88

Hoptman MJ, Davidson RJ (1994) How and why do the two cere- bral hemispheres interact? Psychol Bull 116:195-219

504

Ingum J, Bjorklund R (1994) Effects of flunitrazepam on respons- es to lateralized visual stimuli: evidence for cerebral asymme- try of execution of manual movements to targets in contralat- eral and ipislateral visual space. Psychopharmacology 114: 551-558

Jakobson LJ, Goodale MA (1989) Trajectories of reaches to pris- matically-displaced targets: evidence for "automatic" visuo- motor recalibration. Exp Brain Res 78:575-587

Jakobson LJ, Servos P, Goodale MA, Lassonde M (1994) Control of proximal and distal components of prehension in callosal agenesis. Brain 117:1107-1113

Jeannerod M (1988) The neural and behavioural organization of goal-directed movements. Clarendon Press, Oxford

Jeeves MA, Silver PH (1988) The formation of finger grip during prehension in an acallosal patient. Neuropsychologia 26:153- 159

Johnson PB, Ferraina S Caminiti R (1993) Cortical networks for visual reaching. Exp Brain Res 97:361-365

Jordon MI, Flash T, Arnon Y (1994) A model of the learning of arm trajectories from spatial deviations. J Cogn Neurosci 6:359-376

Karst GM, Hasan Z (1991a) Initiation rules for planar, two-joint arm movements. J Neurophysiol 66:1579-1593

Karst GM, Hasan Z (1991b) Timing and magnitude of electro- myographic activity for two-joint arm movements in different directions. J Neurophysiol 66:1594-1604

Koshland GE Hasan Z (1994) Selection of muscles for initiation of planar, three-joint arm movements with different final ori- entations of the hand. Exp Brain Res 98:157-162

Lakke JPWF, Weerden TW van, Staal-Schreinemachers A (1984) Axial apraxia, a distinct phenomenon. Clin Neurol Neurosurg 86:291-294

Lassonde M, Jeeves MA (1994) Callosal agenesis: a natural split- brain? (Adv Behav Biol 42) Plenum, London

Lawrence DG, Kuypers HGJM (1968a) The functional organiza- tion of the motor system in the monkey. I. The effects of bilat- eral pyramidal lesions. Brain 9h1-14

Lawrence DG, Kuypers HGJM (1968b) The functional organiza- tion of the motor system in the monkey. II. The effects of le- sions of the descending brain-stem pathway. Brain 91:15-36

Levin MF (1996) Interjoint coordination during pointing move- ments is disrupted in spastic hemiparesis. Brain 119:281-293

Lomas J, Kimura D (1976) Intrahemispheric interaction between speaking and sequential manual activity. Neuropsychologia 14:23-34

Marzi CA, Bisiacchi R Nicoletti R (1991) Is interhemispheric transfer of visuomotor information asymmetric? Evidence from a meta-analysis. Neuropsychologia 29:1163-1177

Matelli M, Rizzolatti Bettinardi V, Gilardi MC, Perani D, Rizzo G, Fazio F (1993) Activation of precentral and mesial motor areas during the execution of elementary proximal and distal arm movements: a PET study. Neuroreport 4:1295-1298

Muller F, Kunesch E, Binkofski F, Freund H-J (1991) Residual sensorimotor functions in a patient after right-sided hemisphe- rectomy. Neuropsychologia 29:125-145

Oldfield RC (1971) The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9:97-114

Peters M (1987) A nontrivial motor performance difference be- tween right handers and left handers: attention as an interven- ing variable in the expression of handedness. Can J Psychol 41:91-99

Poeck K, Lehmkuhl G, Willmes K (1982) Axial movements in ide- omotor apraxia. J Nenrol Neurosurg Psychiatry 45:1125-1129

Porac C, Coren S (1976) The dominant eye. Psychol Bull 83: 880-897

Prablanc C, Echallier JF, Komilis E, Jeannerod M (1979) Optimal response of the eye and hand motor systems in pointing at a visual target. I. Spatio-temporal characteristics of eye and hand movements and their relationships when varying the amount of visual information. B iol C ybern 35:113-124

Proctor RW, Reeve TG (1990) Stimulus-response compatibility: an integrated perspective. North-Holland, Amsterdam

Rizzolatti G, Riggio L, Sheliga BM (1995) Space and selective at- tention. In: Umilita C, Moscovitch M (eds) Conscious and un- conscious processing and cognitive function. (Attention and performance XV) MIT Press, Cambridge MA

Roy EA, Kalbfleisch L, Elliot D (1994) Kinematic analyses of manual asymmetries in visual aiming movements. Brain Cogn 24:289-295

Rugg MD, Lines CR, Milner AD (1984) Visual evoked potentials to lateralized visual stimuli and the measurement of interhemi- spheric transmission time. Neuropsychologia 22:215-225

Rugg MD, Milner AD, Lines CR (1985) Visual evoked potentials to lateralised stimuli in two cases of callosal agenesis. J Neu- rol Neurosurg Psychiatry 48:367-373

Savaki HE, Kennedy C, Solokoff L, Mishkin M (1994) Visually- guided reaching with the forelimb contralateral to a blind hemisphere-a metabolic mapping study in monkeys. J Neuro- sci 13:2772-2789

Sergent J, Myers JJ (1985) Manual, blowing, and verbal simple re- actions to lateralized flashes of light in commissurotomized patients. Percept Psychophys 37:571-578

Taira M, Mine S, Georgopolous AR Murata A, Sakata H (1990) Parietal cortex neurons of the monkey related to the visual guidance of hand movement. Exp Brain Res 83:29-36

Tanji J, Okano K, Sato KC (1988) Neuronal activity in cortical motor areas related to ipsilateral, contralateral, and bilateral digit movements of the monkey. J Neurophysiol 60:325-343

Tassinari G, Aglioti S, Pallini R, Berlucchi G, Rossi GF (1994) In- terhemispheric integration of simple visuomotor responses in patients with partial callosal defects. Behav Brain Res 64: 141-149

Tipper SR Lortie C, Baylis GC (1992) Selective reaching: Evi- dence for action-centred attention. J Exp Psychol 18:891-905

Todor JI, Doane T (1978) Handedness and hemispheric asymme- try in the control of movements. J Mot Behav 10:295-300

Triggs WJ, Calvanio R, MacDonell RAL, Cross D, Chiappa KH (1994) Physiological motor asymmetry in human handedness: evidence from transcranial magnetic stimulation. Brain Res 636:270-276

Trope I, Fishman B, Gur RC, Sussman NM, Gur RE (1987) Con- tralateral and ipislateral control of fingers following callosoto- my. Neuropsychologia 25:287-291

Turner RS, Owens JWM, Andersin ME (1995) Directional varia- tion of spatial and temporal characteristics of limb movements made by monkeys in a two-dimensional work space. J Neuro- physiol 74:684-697

Wasserman EM, Pascual-Leone A, Hallett M (1994) Cortical mo- tor representation of the ipsilateral hand and arm. Exp Brain Res 100:121-132