Spatial Vision, Vol. 16, No. 3-4, pp. 347– 364 (2003) Ó VSP 2003. Also available online - www.vsppub.com Pointing errors in immediate and delayed conditions in unilateral optic ataxia P. REVOL 2;4 , Y. ROSSETTI 1;2;4 , A. VIGHETTO 2;3;4 , G. RODE 1;2;4 , D. BOISSON 1;2;3;4 and L. PISELLA 2;4;¤ 1 Service de rééducation neurologique, Hôpital H Gabrielle, Route de Vourles, BP 57, F-69565 Saint Genis-Laval cedex, France 2 Espace et Action, Inserm U534, 16 avenue du doyen Lépine, CP 13, F-69676 Lyon, France 3 Service Neurologique, Hôpital Neuro-cardiologique,59 Bd Pinel, F-69003 Lyon, France 4 Institut Fédératif des Neurosciences, Bâtiment B13, Hôpital Neuro-cardiologique,59, boulevard Pinel, F-69394 Lyon cedex 03, France Received 7 June 2002; revised 9 October 2002; accepted 17 October 2002 Abstract—The present paper provides an analysis of the pointing errors of a patient with unilateral optic ataxia (O.K.) following right hemispheric damage, revealing the type of errors related to the use of the contralesional hand and/ or to the reaching of targets located in the contralesional visual Ž eld. In addition, comparison between immediate and delayed pantomime pointing allow testing of whether pointing deŽ cits of this patient are speciŽ c to real-time visuo-motor control and, subsequently, whether delay could improve his pointing performance. The results show different patterns in the four hand– Ž eld combinations. The following conclusion can be drawn from the results of the delayed condition. In the case of patient O. K., the delay reduced the pointing variability for both hands in the left visual Ž eld but not in the right visual Ž eld. However, the pointing biases did not improve accordingly. As in healthy subjects, target locations tended to be coded in memory with a bias directed toward the Ž xation point. These results are discussed and contrasted with respect to those previously obtained in the literature in patients with bilateral optic ataxia. Keywords: Delay; pointing movement; optic ataxia; human; parietal cortex; visual guidance. 1. INTRODUCTION Visual localization of the goal is the Ž rst stage in the process of performing accurate goal directed movements. Numerous studies in animal and man have shown that visually guided movements imply sensorimotor transformations which convert the retinal visual information in egocentric motor coordinates useful for the organization of movement in space (centred on trunk, shoulder, head or arm: ¤ To whom correspondence should be addressed. E-mail: pisella@lyon.inserm.fr 348 P. Revol et al. e.g. Flanders et al., 1992; Colby, 1998; Desmurget et al., 1998). This visuo-motor processing is made within a neural network including the posterior parietal cortex (PPC) and the premotor cortex, among other brain areas (Jeannerod et al., 1995; Burnod et al., 1999; Battaglia-Mayer and Caminiti, 2002). All of these brain regions are part of the ‘dorsal’ stream of the visual cortical system (see Goodale and Milner, 1992; Milner and Goodale, 1995; Rossetti et al., 2000). When a lesion occurs in the PPC, patients show a characteristic inaccuracy when they reach or grasp visual objects (Balint, 1909; Holmes, 1918). Balint (1909) suggested that these deŽ cits were due to a visuo-motor disconnection that he called ‘optic ataxia’, whereas Holmes (1918) proposed that these same deŽ cits rather result from an impairment in spatial perception that he called ‘visual disorientation’. Following studies (Garcin et al., 1967; Ratcliff and Davies-Jones, 1972; Vighetto, 1980; Vighetto and Perenin, 1981; Perenin and Vighetto, 1988; Jeannerod et al., 1994) tended to conŽ rm that these visuo-motor deŽ cits can occur without perceptual disorder, rather supporting the deŽ nition of Balint (1909). Despite demonstrating an exaggerated grip aperture compared to healthy subjects, these patients are not impaired in object recognition (Perenin and Vighetto, 1988; Goodale et al., 1991; Jeannerod et al., 1994), and can even correctly match the object’s dimension with their Ž nger grip when they do not perform the reaching movement (Goodale and Milner, 1992; Milner and Goodale, 1995). Interpretation of optic ataxia as a speciŽ c visuo-motor disorder was also reinforced by the careful study of reaching behaviour by Perenin and Vighetto (1988). First, the testing of auditory-guided movements showed that optic ataxia appears to be a modality-speciŽ c reaching impairment. Second, although verbal discrimination of dot position was impaired in some patients, a direct causal link between these subtle deŽ cits in visual space perception and the gross misreaching errors was excluded by the authors. Finally, patients showed reaching errors related to both a visual Ž eld effect and to a hand effect. This combination of sensory and motor in uences, as well as the localization of the underlying lesion in-between the visual and motor areas, supported the notion that the deŽ cit lies at the visuo-motor interface rather than at the sensory or motor level. Superimposition of the lesions of 6 left brain-damaged and 2 right brain-damaged patients with optic ataxia using CT scans (Perenin and Vighetto, 1988) revealed a symmetrical converging region including the superior parietal lobule, in and above the intraparietal sulcus, and sparing the human inferior parietal lobule. Recent functional imaging studies conŽ rmed the involvement of this region in distal and proximal visuo-motor guidance during grasping (Grafton et al., 1996; Faillenot et al., 1997; Binkofski et al., 1998; Culham and Kanwisher, 2001) and pointing (Grafton et al., 1992; Kertzman et al., 1997; Desmurget et al., 2001) movements, respectively. Pointing errors associated with optic ataxia have recently been analyzed in two patients with fairly symmetrical bilateral parietal lesions (Milner and Dijkerman, 2001; Milner et al., 2003). Analyzing pointing performances in patient A. T., Milner et al. (1999) noted that the absolute errors increased with tar- Effect of delay on unilateral optic ataxia 349 get eccentricity for both visual hemiŽ elds. In addition, for both patients A. T. and I. G., conŽ dence ellipses of the scatter of Ž nger end positions were abnormally large in both visual hemiŽ elds, and as target eccentricity increased, so did angular errors in direction of gaze (Milner et al., 2003). The questions of whether or not the pointing errors in unilateral optic ataxia are systematic and are directed toward the Ž xation point or the side of the lesion are still debated (Ota et al., 1999). These questions are addressed in the present paper. Until now, pointing deŽ cits associated with unilateral optic ataxia have not been precisely characterized by a detailed error analysis. Thus, the Ž rst aim of the present study is to Ž ll in this gap, analyzing in detail different pointing error parameters (constant and variable errors) and endpoints conŽ dence ellipses, as already done for healthy subjects by Rossetti and Régnier (1995; Rossetti et al., 1996; see Fig. 2). The reaching and grasping errors associated with optic ataxia support the conclusion that the superior parietal lobule lesion disrupts visuo-motor transformations in general. Nevertheless, recent papers have highlighted conditions in which optic ataxic patients exhibit pointing responses similar to those of healthy subjects, suggesting that not all visuo-motor transformations involve this dorsal stream. First, pointing in central vision is known to be essentially unimpaired in optic ataxia (e.g. Perenin and Vighetto, 1983). In addition, Milner et al. (1999) showed that delayed pointing is also performed with good precision. In the same vein, Pisella et al. (2000) showed that the fast visuo-motor guidance subserving the ‘hand automatic pilot’ is speciŽ cally disrupted in bilateral optic ataxia (see also Gréa et al., 2002), whereas slow intentional visuo-motor guidance is preserved. Pisella et al. (2000) concluded from these results that the superior parietal lobule processes mainly fast automatic visuo-motor transformations. Therefore, the existence of an additional system of visuo-motor transformations that does not involve the dorsal stream needs to be advocated. Goal-directed actions do not always imply the dorsal automatic system operating in real-time and using egocentric spatial location. Psychophysical studies have shown that target location could also be computed with respect to other visual stimuli, i.e. using an allocentric frame of reference (Bridgeman et al., 1997, 2000; Rossetti, 1998, 2000; Coello and Magne, 2000; Coello et al., 2000, 2002). This coordinate computation has been proposed to take place within the ‘ventral’ stream of visual processing, including the infero-temporal cortex (Goodale et al., 1994; Bridgeman et al., 1997; Rossetti, 1998, 2000; Milner et al., 1999, 2001; Pisella et al., 2000). This indirect pathway for action (Milner and Dijkerman, 2001) would necessarily be used in bilateral optic ataxia, and may also be used in healthy subjects in conditions where the target is no longer present and a delay is interposed between the stimulus presentation and the movement response (Rossetti and Pisella, 2002). During this delay period, it has been postulated that the intrinsic object properties and the target location could be stored in working memory, and subsequently used for action (Rossetti, 1998, 2000; Milner and Dijkerman, 2001; Milner et al., 1999, 2001, 2003). When goal-directed movements are performed 350 P. Revol et al. in a poor visual environment (such as in experimental conditions), the allocentric coding of the target location necessary for delayed visuo-motor transformations may use the location of the Ž xation point as a spatial landmark. Experimental evidence for this hypothesis has been provided by the study of Sheth and Shimojo (2001). This study showed that in a ‘delayed’ pointing task, the coding of target locations in healthy subjects is spatially compressed toward gaze direction. This bias in target coding was not observed for ‘immediate’ pointing, because healthy subjects probably spontaneously use their ‘dorsal’ stream in this condition. Supposing that allocentric coding is slowly processed, it would explain that angular errors towards the Ž xation point in bilateral optic ataxia are larger in ‘immediate’ pointing. In healthy subjects, a delay modiŽ es the kinematic parameters of movements (Goodale et al., 1994) and worsens pointing accuracy (Milner et al., 1999); this is to be expected if we postulate the use of the dorsal stream for immediate movements, and the use of the indirect visual route for goal-directed action relying on memory processes for delayed pointing. By contrast, both grasping and pointing performances were dramatically improved with the delay in bilateral optic ataxic patients, probably because the use of the intact ‘ventral’ stream takes longer to complete the processing of visual information (Rossetti, 1998; Milner et al., 1999, 2001, 2003; Rossetti et al., 2000). These results corroborate previous Ž ndings obtained in the visual agnosic patient D. F., who exhibited spatially oriented behaviour opposite to that of manually optic ataxic patients. This patient, with a bilateral lesion of the infero-temporal cortex (‘ventral’ stream), can accurately grasp objects that she cannot describe (Goodale et al., 1991). Her intact visuo-motor performance has been attributed to a spared ‘dorsal’ stream processing (Goodale and Milner, 1992; Milner and Goodale, 1995). By contrast, when a delay was interposed between the stimulus presentation and the movement response, D. F. showed a grip size which did not correlate any more the object width, probably due to an impairment of her ‘ventral’ stream (Goodale et al., 1994). Taken as a whole, the spatial behaviours of these brain damaged patients and healthy subjects converge to the idea that there are mainly two systems for spatial representation in the brain specialized for two different functions (see reviews Bridgman et al., 1997, 2000; Rossetti, 1998; Pisella and Rossetti, 2000; Milner and Dijkerman, 2001). The ‘dorsal’ pathway elaborates ‘short-lived’ (Rossetti, 2000) on-line sensorimotor transformations used for goal directed movements, whereas the ‘ventral’ pathway is dedicated to cognitive processing of visual information for action, using long-lasting representation. Neuroanatomical data support this conclusion. Indeed, the dissociation between a ventral and a dorsal stream of visual processing has been initially based on the connectivity of visual areas and the functional properties of the cells participating in each of these streams. Single-unit studies in the monkey also provide evidence for a temporal dissociation between the two visual pathways. Comparison of visual response latencies at various locations in the monkey visual cortex have led Nowak and Bullier (1997) to distinguish what they called a ‘fast brain’ and a ‘slow brain’ system. Parietal areas from the ‘dorsal’ Effect of delay on unilateral optic ataxia 351 stream exhibit particularly short latencies related to their magnocellular neurons (about 40– 80 ms). In addition, numerous short-cuts are found within the dorsal stream and parietal areas directly project onto the premotor cortex (see reviews in Pisella and Rossetti, 2000; Rossetti et al., 2000; Rossetti and Pisella, 2002). By contrast, other extra-striate areas in pertaining to the ventral stream exhibit slow latencies (about 100– 150 ms) and more indirect projections to the primary motor area. Thus, among the various elements of dorsal-ventral dissociation in the motor context, the time constraints play a crucial role (see reviews in Pisella and Rossetti, 2000; Rossetti et al., 2000; Rossetti and Pisella, 2002). Therefore, the second aim of this study was to test whether visually-guided movement deŽ cits are speciŽ c to real-time visuo-motor control in a case of unilateral optic ataxia. This result was previously found in bilateral brain-damaged patients, by means of a comparison of performances between immediate and delayed pointing. In addition, the present study will investigate whether unilateral optic ataxia patients use the allocentric coding of space for ‘immediate’ and/ or ‘delayed’ pointing, knowing that the function of one PPC is spared in unilateral optic ataxia. 2. MATERIAL AND METHODS 2.1. Case description Patient O. K. was a 33-year-old man at the times of testing, 5 months after an ischemic stroke involving the posterior branch of the right sylvian artery. Magnetic resonance imaging revealed a hyperintense signal on T2 sequences located in the right posterior parietal lobe along with a slight damage to the right posterior part of the corpus callosum (see Fig. 1). At clinical assessment, O. K. demonstrated a typical optic ataxia for his left hand in both peripheral visual Ž elds without any pure visual Ž eld or motor deŽ cits, whereas he did not show any sign of unilateral neglect, a Ž nding which is in agreement with the previous conclusion of a double dissociation between these two symptoms (Perenin and Vighetto, 1988; Revol, 2000). In the present paper, pointing performances were quantiŽ ed with the left and the right hand, in both visual Ž elds. 2.2. Procedure Patient O. K. had his head Ž xed by means of a chin rest in front of the experimental set-up. He was asked to point towards 8 targets, displayed in a fronto-parallel plane, while he was looking at a central Ž xation point. Targets were located at 60 mm, 120 mm, 180 mm or 240 mm from the central Ž xation point, on the right visual Ž eld (signed positive) or the left visual Ž eld (signed negative). Pointing movements were performed as accurately and as quickly as possible, either with the right or the left hand. For each movement the starting point was located close to the body, aligned with O. K.’s midsagittal plane. Since the patient was asked to Ž xate the 352 P. Revol et al. Figure 1. The lesion of case O. K. T2 weighted horizontal magnetic resonance imaging section demonstrates an ischemic lesion of the right posterior parietal region. LH: left hemisphere; RH: right hemisphere. central point, pointing towards an eccentric target was performed in peripheral vision. Two conditions were tested. In the ‘immediate’ pointing condition, the patient had to point towards the target as soon as it was presented. The pointing movement began and the target remained visible until the end of the trial. In the ‘delayed’ pantomime pointing condition (Milner et al., 1999), the target was presented for 2 s only, after which it was hidden for 5 s and the patient was asked to point to the target location as soon as it reappeared. As before, O. K. was asked to hold his gaze towards the Ž xation point. Both conditions were intermingled in a ABBA design and the targets were presented randomly, so that a total of 10 pointing movements were performed for each target and each condition. A Ž rst session was performed with the left hand, then a second session was performed separately with the right hand. Pointing movements were recorded with an optoelectronic motion analyzer (Vicon® , Oxford Metrics). The position of a small sphere covered attached with re ecting tape to the tip of the index Ž nger was recorded at a sampling rate of 200 Hz. The end-point recording of each individual trial was used to calculate different accuracy parameters (see Fig. 2): the constant errors measured in amplitude (½) and direction (®) relative to the ideal reach that would hit the target. Amplitude errors (½) were negative when movement distance was shorter than the ideal reach, and angular errors (®) were signed negative when the reach was closer to Effect of delay on unilateral optic ataxia 353 Figure 2. Schematic representation of the constant and variable pointing errors measured for each target in each pointing condition (see text for details). the body than the aimed target. Variable errors were assessed by conŽ dence ellipses (95%) of the scatter of Ž nger end positions (Rossetti and Régnier, 1995). The ellipse surface provided an estimate of the global pointing variability. Although absolute error (d), the unsigned mean distance between the Ž nal pointing positions and the target, is a compound parameter that does not precisely describe the data (Schutz and Roy, 1973), it was also used in the present study in order to allow comparison with previous reports. All variables were submitted to a three-way (Hand £ Delay £ Visual Ž eld) analysis of variance. Moreover, contrasts analysis were performed in order to quantify the effect of one factor among the two others, e.g. effect of the delay for pointing responses with the left hand in the left visual Ž eld. 3. RESULTS 3.1. Qualitative description Movement endpoints and conŽ dence ellipses obtained for the 8 targets in the immediate and delayed conditions were plotted in Figs 3A and 3B for the patient’s left hand and right hand respectively. First, these Ž gures allow a qualitative description of the patient’s pointing errors depending on the visual Ž eld and the hand. The pattern of immediate pointing illustrated in these Ž gures revealed that ataxic errors in this patient result from combined impairments in the contralesional visual Ž eld and in the use of the contralesional hand. Immediate pointing toward the Figure 3. Movement endpoints and conŽ dence ellipses obtained with the left hand (A) or with the right hand (B) in the ‘immediate’ and the ‘delayed’ pointing conditions. The horizontal and vertical axes correspond to the x- and y-coordinates (measured in mm) of the movement end-points with respect to the starting position (i.e. close to the body and aligned with the patient’s midsagittal plane). A conŽ dence ellipse (95%) is drawn for the Ž nal pointing scatter corresponding to each target. Black crosses match the eight target locations and the grey cross indicates the location of the Ž xation point. In the left visual Ž eld, irrespective of the hand used, pointing scatters in the ‘immediate’ condition were the largest. The delay drastically reduced this variability. By contrast, pointing scatters in the right visual Ž eld were smaller and those with the right hand provide a control for unimpaired performances. (A) 354 P. Revol et al. Effect of delay on unilateral optic ataxia Figure 3. (Continued). (B) 355 356 P. Revol et al. left visual Ž eld was the most variable, irrespective of the hand used, and the most inaccurate condition (large ellipses uncentred on the target). In addition, immediate pointing to the right visual Ž eld with the left hand also showed a pathological variability, illustrated by ellipses with relatively large surfaces (but centred on the target). Immediate pointing to the right visual Ž eld with the right hand provided the control performance of the patient. Second, Figs 3A and 3B allow qualitative comparison between immediate and delayed pointing performance for each of the four hand-Ž eld combinations. For the left visual Ž eld, whichever hand was used, the pointing scatters (quantiŽ ed by the ellipse’s surface) appeared clearly smaller in the ‘delayed’ condition. In addition, it appeared that the delayed pointing responses overall tended to be located closer to the gaze direction, i.e. towards the Ž xation point. 3.2. Quantitative analysis 3.2.1. Precision: variable error. Analysis of variance showed a signiŽ cant main effect of Visual Ž eld and Delay on the pointing variability. The surface of the conŽ dence ellipses was signiŽ cantly larger for the left visual Ž eld (F .1; 3/ D 16:32, p < 0:05), and overall the pointing scatters dramatically decreased with the delay (F .1; 3/ D 11:06, p < 0:05). All second order interactions were also signiŽ cant. Therefore, results will be described below for each hand-Ž eld combination. 3.2.2. Accuracy: constant and absolute errors. The amplitude errors (½) showed a signiŽ cant main effect of Hand (F .1; 3/ D 29:92, p < 0:05), probably due to the differential performance in the left visual Ž eld: movements were largely hypometric when the patient pointed with the right hand but largely hypermetric when pointing with the left hand. For the absolute error (d), analysis of variance revealed no signiŽ cant main effect but a signiŽ cant interaction was found between the Delay and the Visual Ž eld (F .1; 3/ D 10:99, p < 0:05). In the left visual Ž eld absolute errors decreased with the delay (23.59 mm to 20.15 mm, p D 0:93), whereas they increased in the right visual Ž eld (7.24 mm to 28.59 mm, p D 0:10; see Fig. 4). In terms of constant errors, a signiŽ cant main effect of Delay was observed for patient O. K.’s amplitude errors (½) and angular errors (®) (F .1; 3/ D 15:97, p < 0:05 and F .1; 3/ D 15:83, p < 0:05, respectively). In fact, the movement amplitude was smaller and the angular error was greater with the delay (see Fig. 3). But a signiŽ cant Hand £ Delay interaction (F .1; 3/ D 17:94, p < 0:05) on the angular errors (®), and a signiŽ cant Hand £ Delay £ Visual Ž eld interaction (F .1; 3/ D 37:35, p < 0:05) on the amplitude errors (½) suggested that this pattern of evolution with the delay needs to be examined separately for each visual Ž eld and hand condition. For the right hand in the right visual Ž eld. This combination served as a ‘control performance’ for the three other combinations. Indeed, patient O. K.’s pointing Figure 4. Histogram representing the absolute errors (mm) obtained with the left or the right hand in the ‘immediate’ (black bars) or ‘delayed’ (hatched bars) pointing conditions. Absolute errors were sizeable in the left visual Ž eld with both hands and were reduced by the delay. By contrast, in the right visual Ž eld absolute errors increased with the delay. Effect of delay on unilateral optic ataxia 357 358 P. Revol et al. movements in this condition were fairly accurate and reproductible (see Fig. 3B). In addition, patient O. K.’s error parameters ½; ® and d increased with the delay (all F s.1; 3/ > 5:96, p < 0:09), which is usually the case for healthy subjects. However no signiŽ cant change was noticed for the ellipse surface. For the left hand in the right visual Ž eld. The ellipse surface was larger than in the right hand and right Ž eld control performance (Fig. 3A), showing that the contralesional arm was also ataxic in the ipsilesional Ž eld, as a result of a hand effect. In terms of accuracy, however, contrary to pointing directed toward the left visual Ž eld, constant errors were comparable to the ‘control performance’ (i.e. close to zero). The ellipse surface was not affected by the delay. As in the control performance (right hand in the right visual Ž eld), all the error parameters, except the ellipse surfaces, increased with the delay (all F s.1; 3/ < 10:41, p > 0:05, see Figs 4 and 5). For the left hand in the left visual Ž eld. In the ‘immediate’ condition, the patient demonstrated large absolute errors (Fig. 4) resulting from sizeable amplitude errors (Fig. 5), moderate angular errors (Fig. 5) and large pointing scatters (Fig. 3A). It was the only condition where the patient demonstrated large amplitude errors (½) in the direction of hypermetria. It was the only condition where both amplitude (F .1; 3/ D 16:42, p < 0:05) and absolute pointing errors (24.35 mm to 15.65 mm in means, p D 0:20, see Fig. 5) were reduced with the delay. The delay also reduced the ellipse surfaces (F .1; 3/ D 16:78, p < 0:05). However, it is worth noting that angular errors became more negative with the delay as a result of a pointing convergence in the direction of the Ž xation point (Fig. 3A). For the right hand in the left visual Ž eld. In this condition as well, the immediate pointing was marked by large absolute errors (Fig. 4) resulting from hypometric errors for the two most eccentric targets on the left (see amplitude errors on Fig. 5), positive angular errors (Fig. 5), and large pointing scatters (Fig. 3B). It was the only condition where the patient demonstrated large positive angular errors (®). The delay signiŽ cantly reduced the ellipse surface (F .1; 3/ D 12:68, p < 0:05) and aggravated the ® error (F .1; 3/ D 14:26, p < 0:05, see Fig. 5). The amplitude errors (½) were statistically weakened with the delay (F .1; 3/ D 28:65, p < 0:05). The mean absolute error (d) remained unchanged (22.83 mm to 24.65 mm, p D 0:81). 4. DISCUSSION The present paper Ž rst aimed to quantify and characterise the ataxic pointing errors of the patient O. K. in each visual Ž eld and for each hand, using several error parameters allowing discrimination between precision and accuracy. To our knowledge it is the Ž rst time that such a detailed description of the pointing errors Figure 5. Histograms representing the amplitude errors (mm) and angular errors (± ) obtained with the left or the right hand, in the ‘immediate’ (black bars) or ‘delayed’ (hatched bars) pointing conditions. Amplitude errors decreased with the delay for points in the left visual Ž eld with the left hand, whereas in all other conditions they slightly increased. For the left hand, alpha errors converged in the direction of the Ž xation point when the delay was introduced. Effect of delay on unilateral optic ataxia 359 360 P. Revol et al. to peripheral targets in unilateral optic ataxia has been provided. The present results conŽ rmed the observations of Perenin and Vighetto (1988) that a unilateral lesion to the right superior parietal lobule leads to a combination of errors related to the visual Ž eld effect (errors in the left visual Ž eld, independent of the hand used), and to the hand effect (errors with the left hand, independent of the visual Ž eld where the target is presented). Figure 3 clearly showed that the border between the two hemiŽ elds corresponded to a drastic change in the pointing performance, and that the performance was consistent for all target positions within each hemiŽ eld. Performance did not change from left to right space in a gradual way. The measure of different error parameters for each hand– Ž eld combination allowed us to distinguish the errors made in the contralesional visual Ž eld (irrespective of the hand used), which were observed on all parameters, and the errors made in the ipsilesional visual Ž eld (with the contralesional hand), which consisted of only abnormally large pointing scatters. If this pattern is conŽ rmed with other unilateral patients, it would suggest that ataxic errors related to a visual Ž eld effect are expressed in terms of accuracy as well as variability (or precision), whereas ataxic errors related to the contralesional hand are expressed only by variable errors. The initial errors during pointing to peripheral targets in the two patients with bilateral optic ataxia published in the literature (A. T. and I. G.) are also expressed in terms of accuracy and variability. Only the right hand was tested. Pointing scatters were abnormally large in the whole visual space for both A. T. and I. G. (Milner et al., 2003). Absolute errors (d) increased with eccentricity in the left and the right hemiŽ elds (reported by Milner et al., 1999 for A. T. and Milner and Dijkermann, 2001 for I. G.). This pattern was observed in the present case only in the left visual Ž eld with the left hand, i.e. when hand and Ž eld effects are added as in bilateral optic ataxia. We also measured amplitude errors (½) and angular errors (®) (see Milner et al., 2003): A. T. consistently exhibited movements with short amplitude, whereas I. G. tended to overshoot the peripheral targets. No consistent pattern emerged from amplitude errors in O. K. Concerning angular errors (®), both A. T. and I. G. consistently exhibited errors in the direction of the center of gaze, increasing with target eccentricity (Milner et al., 2003). This pattern was also observed when the targets were presented within a 2D matrix to a patient with unilateral optic ataxia: the error vectors pointed towards the Ž xation point (Ota et al., 1999, 2002). Note that when Vighetto (1980; Perenin and Vighetto, 1983) measured accuracy for horizontal pointing (amplitude and directional errors were not distinguishable) in three patients with unilateral optic ataxia, all the patients showed a tendency towards increasing hypometria with eccentricity in the contralesional Ž eld with the contralesional hand. This pattern was also consistent with the results obtained in bilateral optic ataxia and argues in favour of the idea proposed by Ratcliff and Davies-Jones (1972) that pointing errors are directed toward the Ž xation point. This idea has been recently renewed by the striking single case description of ‘magnetic misreaching’ by Carey et al. (1997). Jeannerod (1986) interpreted these results as a systematic pointing error toward the side of the lesion. Effect of delay on unilateral optic ataxia 361 However, none of these patterns was observed in the patient O. K., who rather tended to produce leftward errors in the left visual Ž eld (in the direction opposite to both gaze direction and lesion side), especially with the right hand. Further studies using the same paradigm should enable us to determine whether the present result is attributable to between-patient variability factors or whether O. K. himself exhibit an unusual pattern. The second aim of this paper was to test, for each hand– Ž eld combination, the effect of delaying the response on the pointing performance of O. K. The comparison between ‘immediate’ and ‘delayed’ pointing responses allows evaluation of real-time visuo-motor control of optic ataxia, which may rely on the PPC, versus off-line visuo-motor control (see Rossetti and Pisella, 2002). In healthy subjects, variable error tends to increase with delay, as a result of the introduction of a memorisation process between stimulus presentation and pointing (Rossetti and Régnier, 1995; Rossetti et al., 1996; Rossetti, 1998, 2000; Rossetti and Pisella, 2002). Moreover, visual memory has been associated with space compression in a pointing task performed by healthy subjects (Sheth and Shimojo, 2001), i.e. associated with systematic mislocalisation of the targets toward the gaze direction. In the two patients with bilateral optic ataxia (A. T. and I. G.), previous experiments have shown a paradoxical improvement of both the pointing variability and accuracy across the entire visual Ž eld (except in central vision) in conditions of delayed real or delayed pantomime pointing (i.e. with the target present or absent respectively at the time of the go-signal). Improvement of the pointing accuracy with delay was Ž rst observed on absolute errors (d; Milner et al., 1999), and then on angular errors (®I Milner and Dijkermann, 2001; Milner et al., 2003). In these two patients, it was the pattern of immediate pointing to peripheral targets that was compatible with a space compression phenomenon. Delaying the action reduced the initial angular errors (®) oriented toward the gaze direction whereas amplitude errors (½) were not affected by delay (Milner and Dijkermann, 2001; Milner et al., 2003). In the case of patient O. K., as in normal subjects, space compression appeared with delay in all hand-Ž eld combinations (Fig. 3). Nonetheless, as in bilateral optic ataxia, a paradoxical effect of delay was observed on the pointing scatters in patient O. K.: the pointing variability dropped in the left visual Ž eld for both hands. Delay therefore seems to improve the variability of the pointing errors due to the visual Ž eld effect. These results raise the question of a possible dissociation between the errors resulting from Ž eld or hand effect. However, it tends to conŽ rm that optic ataxia is not a general deŽ cit of action or visuo-motor transformation but rather a speciŽ c deŽ cit of real-time action (Milner et al., 1999, 2001; Pisella et al., 2000; Rossetti and Pisella, 2002). This is consistent with the fact that optic ataxia is mostly observable in peripheral vision, i.e. when real-time processing plays a crucial role for action guidance. To be provocative, it could therefore be proposed that optic ataxia might be considered as a temporal, rather than spatial disorder. 362 P. Revol et al. Acknowledgements The authors want to thank Brigitte Delouis and Carine Michel for their technical help. This study was funded by grants from Mc Donnell Pew foundation and PHRC (Hospices Civils de Lyon). REFERENCES Balint, R. (1909). Seelenhamung des ‘Schauens’, optische Ataxie, raümlische Störung des Aufmeirsamkeit, Monatschrift für Psychiatrie und Neurologie 25, 51– 81. Battaglia-Mayer, A. and Caminiti, R. (2002). Optic ataxia as a result of the breakdown of the global tuning Ž elds of parietal neurones, Brain 125 (Pt. 2), 225– 237. Binkofski, F., Dohle, C., Posse, S., Stephan, K. M., Hefter, H., Seitz, R. J. and Freund, H. J. (1998). Human anterior intraparietal area subserves prehension: A combined lesion and functional MRI activation study, Neurology 50 (5), 1253– 1259. Bridgeman, B. (2000). Interactions between vision for perception and vision for behavior, in: Beyond Dissociation: Interaction Between Dissociated Implicit and Explicit Processing, Rossetti, Y. and Revonsuo, A. (Eds), pp. 19– 37. Benjamins, Amsterdam, The Netherlands. Bridgeman, B., Peery, S. and Arnand, S. (1997). Interaction of cognitive and sensorimotor maps of visual space, Perception and Psychophysics 59 (3), 456– 469. Burnod, Y., Baraduc, P., Battaglia-Mayer, A., Guigon, E., Koechlin, E., Ferraina, S., Lacquaniti, F. and Caminiti, R. (1999). Parieto-frontalcoding of reaching: an integrated framework, Exper. Brain Res. 129, 325– 346. Carey, D. P., Coleman, R. J. and Della Sala, S. (1997). Magnetic misreaching, Cortex 33 (4), 639– 652. Coello, Y. and Magne, P. (2000). Determination of target position in a structured environment: selection of an information for action, Eur. J. Cognitive Psychol. 12, 489– 519. Coello, Y., Magne, P. and Plenacoste, P. (2000). The contribution of retinal signal to the speciŽ cation of target distance in a visuo-manual task, Current Psychology Letters 3, 75– 89. Coello, Y., Richaud, S., Magne, P. and Rossetti, Y. (2002). Vision for spatial perception and vision for action: a dissociation between the left-right and near-far dimension, Neuropsychologia 41 (5), 622– 633. Colby, C. L. (1998). Action-oriented spatial reference frames in cortex, Neuron 20 (1), 15– 24. Culham, J. C. and Kanwisher, N. G. (2001). Neuroimaging of cognitive functions in human parietal cortex, Current Opinion in Neurobiology 11, 157– 163. Desmurget, M., Pelisson, D., Rossetti, Y. and Prablanc, C. (1998). From eye to hand: planning goaldirected movements, Neuroscience Behavioral Review 22 (6), 761– 788. Desmurget, M., Gréa, H., Grethe, J. S., Prablanc, C., Alexander, G. E. and Grafton, S. T. (2001). Functional anatomy of nonvisual feedback loops during reaching: a positron emission tomography study, J. Neurosci. 21 (8), 2919– 2928. Faillenot, I., Toni, I., Decety, J., Grégoire, M.-C. and Jeannerod, M. (1997). Visual pathways for object-oriented action and object recognition: functional anatomy with PET, Cerebral Cortex 7, 77– 85. Flanders, M., Tillery, S. I. H. and Soechting, J. F. (1992). Early stages in a sensorimotor transformation, Behavioral and Brain Science 15, 309– 362. Garcin, R., Rondot, P. and de Recondo, J. (1967). Optic ataxia localized in 2 left homonymous visual hemiŽ elds (clinical study with Ž lm presentation), Revue Neurologique (Paris) 116 (6), 707– 714. Goodale, M. A. and Milner, A. D. (1992). Separate visual pathways for perception and action, Trends in Neuroscience 15 (1), 20– 25. Goodale, M. A., Milner, A. D., Jakobson, L. S. and Carey, D. P. (1991). Perceiving the world and grasping it. A neurological dissociation, Nature 349, 154– 156. Effect of delay on unilateral optic ataxia 363 Goodale, M. A., Jakobson, L. S. and Keillor, J. M. (1994). Differences in the visual control of pantomimed and natural grasping movements, Neuropsychologia 32, 1159– 1178. Gréa, H., Pisella, L., Rossetti, Y., Desmurget, M., Tilikete, C., Grafton, S. T., Prablanc, C. and Vighetto, A. (2002). A lesion of the posterior parietal cortex disrupts on-line adjustments during aiming movements, Neuropsychologia 40 (13), 2471– 2480. Grafton, S. T., Mazziotta, J. C., Woods, R. P. and Phelps, M. E. (1992). Human functional anatomy of visually guided Ž nger movements, Brain 115 (Pt. 2), 565– 587. Grafton, S. T., Fagg, A. H., Woods, R. P. and Arbib, M. A. (1996). Functional anatomy of pointing and grasping in humans, Cerebral Cortex 6 (2), 226– 237. Holmes, G. (1918). Disturbances of visual orientation, Brit. J. Ophthalmol. 2, 449– 506. Jeannerod, M. (1986). Mechanisms of visuo-motor coordination: a study of normal and braindamaged subjects, Neuropsychologia 24 (1), 41– 78. Jeannerod, M., Arbib, M. A., Rizzolatti, G. and Sakata, H. (1995). Grasping objects: the cortical mechanisms of visuomotor transformation, Trends in Neuroscence 18, 314– 320. Kertzman, C., Schwarz, U., ZefŽ ro, T. A. and Hallett, M. (1997). The role of posterior parietal cortex in visually guided reaching movements in humans, Exper. Brain Res. 114 (1), 170– 183. Milner, A. D. and Goodale, M. A. (1995). The Visual Brain in Action. Oxford University Press, Oxford, UK. Milner, A. D. and Dijkerman, H. C. (2001). Direct and indirect visual routes to action, in: Out of Mind: Varieties of Unconscious Processes, De Gelder, B., De Haan, E. and Heywood, C. A. (Eds), pp. 241– 264. Oxford University Press, Oxford, UK. Milner, A. D., Paulignan, Y., Dijkerman, H. C., Michel, F. and Jeannerod, M. (1999). A paradoxical improvement of optic ataxia with delay: new evidence for two separate neural systems for visual localization, Proc. Roy. Soc. London (B) 266, 2225– 2230. Milner, A. D., Dijkerman, H. C., Pisella, L., McIntosh, R. D., Tilikete, C., Vighetto, A. and Rossetti, Y. (2001). Grasping the past: delay can improve visuomotor performance, Current Biology 11, 1896– 1901. Milner, A. D., Dijkerman, H. C., McIntosh, R. D., Rossetti, Y. and Pisella, L. (2003). Delayed reaching and grasping in patients with optic ataxia, in: Progress in Brain Research Series: Neural Control of Space Coding and Action Production, Pelisson, D., Prablanc, C. and Rossetti, Y. (Eds). Elsevier, Amsterdam (in press). Nowak, L. G. and Bullier, J. (1997). The timing of information transfer in the visual system, in: Extrastriate Cortex in Primates, Kaas, J., Rochland, K. and Peters, A. (Eds), Cerebral Cortex, Vol. 12, pp. 205– 241. Plenum Press, New York. Ota, H., Fujii, T., Ohtake, H., Suzuki, K. and Yamadori, A. (1999). Deviated reaching toward Ž xation point — a case report of ataxie optique, Japan. J. Neuropsychol. 15 (4), 43 (in Japanese). Ota, H., Fujii, T., Ohtake, H., Tabuchi, M., Meguro, Y., Yamadori, A. and Rossetti, Y. (2002). Deviation to a Ž xation: performance from patients with optic ataxia, in: Euroconference and EBBS Workshop on Cognitive and Neural Mechanisms of Visuomotor Control, La Londe, France. Perenin, M.-T. and Vighetto, A. (1983). Optic ataxia: a speciŽ c disorder in visuomotor coordination, in: Spatially Oriented Behavior, Hein, A. and Jeannerod, M. (Eds), pp. 305– 326. Springer-Verlag, New York. Perenin, M.-T. and Vighetto, A. (1988). Optic ataxia: a speciŽ c disruption in visuomotor mechanisms. I. Different aspects of the deŽ cit in reaching for objects, Brain 111, 643– 674. Pisella, L. and Rossetti, Y. (2000). Interaction between conscious identiŽ cation and non-conscious sensori-motor processing: temporal constraints, in: Beyond Dissociation: Interaction between Dissociated Implicit and Explicit Processing, Rossetti, Y. and Revonsuo, A. (Eds), pp. 129– 152. Benjamins, Amsterdam, Pisella, L., Gréa, H., Tilikete, C., Vighetto, A., Desmurget, M., Rode, G., Boisson, D. and Rossetti, Y. (2000). An ‘automatic pilot’ for the hand in human posterior parietal cortex: toward reinterpreting optic ataxia, Nature Neuroscience 3, 729– 736. 364 P. Revol et al. Ratcliff, G. and Davies-Jones, G. A. (1972). Defective visual localization in focal brain wounds, Brain 95 (1), 49– 60. Revol, R. (2000). Cadres de référence et mouvements d’orientation spatiale. Approche psychophysique chez des sujets sains et des patients avec fésions pariétales. (Frame of reference and spatially oriented behaviors. Psychophysical studies in healthy subjects and with parietal lobe damage patients.) Unpublished PhD thesis dissertation, Lyon. Rossetti, Y. (1998). Implicit short-lived motor representations of space in brain damaged and healthy subjects, Conscious Cognition 7, 520– 558. Rossetti, Y. (2000). Implicit perception in action: short-lived motor representations of space, in: Finding Consciousness in the Brain, Grossenbacher, P. G. (Ed.), pp. 131– 179. Benjamins, Amsterdam, The Netherlands. Rossetti, Y. and Pisella, L. (2002). Several ‘vision for action’ systems: a guide to dissociating and integratingdorsal and ventral functions, in: Attention and Performance XIX: Common Mechanisms in Perception and Action, Prinz, W. and Hommel, B. (Eds), pp. 62– 119. Oxford University Press, Oxford, UK. Rossetti, Y. and Régnier, C. (1995). Representations in action: pointing to a target with various representations, in: Studies in Perception and Action III, Bardy, B. G., Bootsma, R. J. and Guiard, Y. (Eds), pp. 233– 236. Lawrence Erlbaum Ass., Mahwah, N.J., USA. Rossetti, Y., Gaunet, F. and Thinus-Blanc, C. (1996). Early visual experience affects memorization and spatial representation of proprioceptive targets, Neuroreport 7 (6), 1219– 1223. Rossetti, Y., Pisella, L. and Pélisson, D. (2000). Eye blindness and hand sight: temporal constraints and visuo-motor network, Visual Cognition 7 (6), 785– 809. Schutz, R. W. and Roy, E. A. (1973). Absolute error: the devil in disguise, J. Motor Behavior 3, 141– 153. Sheth, B. R. and Shimojo, S. (2001). Compression of space in visual memory, Vision Research 41, 329– 341. Tanné, J., Boussaoud, D., Boyer-Zeller, N. and Rouiller, E. M. (1995). Direct visual pathways for reaching movements in the macaque monkey, Neuroreport 7, 267– 272. Vighetto, A. (1980). Etude neuropsychologique et psychophysique de l’ataxie optique. (Neuropsychological and psychophysical studie of optic ataxia.) Unpublished MD thesis dissertation, Lyon. Vighetto, A. and Perenin, M. T. (1981). Optic ataxia: analysis of eye and hand responses in pointing at visual targets, Revue Neurologique (Paris) 137 (5), 357– 372.