No dissociation between perception and action in patient DF when haptic feedback is withdrawn

Goodale et al. (1991) reported a striking dissociation between vision for perception and action. They examined DF, a human patient who had damage to her ventral visual stream and suffered from visual form agnosia. She was unable to perceive an object's size but could match the opening of her hand to the object's size during grasping. It was concluded that grasping relied on a separate representation of visual size in the dorsal stream and required no visual input from the ventral stream. This observation inspired the influential perception-action model, which claimed separate visual streams for perception and action. However, in grasping (but not in corresponding perceptual tasks), participants receive haptic feedback after each trial. Using this feedback, DF might compensate for her impaired size-vision. I reexamined DF's grasping behavior using a mirror apparatus to dissociate the image of an object from its physical presence. DF's grasping was only normal when she received haptic feedback. Thus, in grasping, DF can rely on haptic feedback to compensate for her deficit in size-perception. This can explain why her grasping is significantly better than her perceptual performance. The findings emphasize the extent of early interstream interactions and highlight the multimodal nature of sensory processing in the dorsal stream.     

Optic ataxia and the function of the dorsal stream: Contributions to perception and action

Optic ataxia (OA) is one of the symptoms pertaining to Bálint's Syndrome. It has been described clinically for nearly 80 years before it became a cornerstone of the most popular dual stream theory of the visual brain. Over the last 10 years a regain of interest for this neurological condition lead to a number of precise analyses of the deficits found in optic ataxia, giving rise to a renewed outline of its very definition and hence of the function(s) of the occipito-parietal (dorsal) stream.In absence of concomitant clinical symptoms, we review evidence that misreaching errors in central vision result from the “hand effect”: an erroneous dynamic spatial processing of proprioceptive information from the hand. When visual feedback of the hand is provided (closed-loop condition), pure optic ataxia is restricted to peripheral vision. This central versus peripheral vision distinction is repeatedly used to argue that action and perception are not unique and dissociated systems. New assessments of optic ataxia patients are provided, confirming on one hand that their visuomotor deficit is specific to peripheral vision (i.e. when the gaze and the hand goals are dissociated), on the other hand that they disclose perceptual deficits in peripheral vision. These results are coherent with the recent demonstration that optic ataxia patients exhibit a general contralesional deficit for dynamic visuo-spatial processing, affecting both hand and eye movements [Gaveau, V., Pélisson, D., Blangero, A., Urquizar, C., Prablanc, C.,Vighetto, A., et al. (2008). A common parietal module for saccade and reach: Eye-hand coordination and saccadic control in optic ataxia. Neuropsychologia, 46, 475–486]. Such module(s) within the dorsal stream could be used for both action and perception in the periphery.It is concluded that optic ataxia cannot be considered as a unitary and specific visuo-manual deficit, and that the modular organisation of the dorsal stream allows for numerous dorsal–ventral interactions for perception and action.     

Concurrent visuomotor behaviour improves form discrimination in a patient with visual form agnosia

It is now well established that the visual brain is divided into two visual streams, the ventral and the dorsal stream. Milner and Goodale have suggested that the ventral stream is dedicated for processing vision for perception and the dorsal stream vision for action [A.D. Milner & M.A. Goodale (1995) The Visual Brain in Action, Oxford University Press, Oxford]. However, it is possible that ongoing processes in the visuomotor stream will nevertheless have an effect on perceptual processes. This possibility was examined in the present study. We have examined the visual form-discrimination performance of the form-agnosic patient D.F. with and without a concurrent visuomotor task, and found that her performance was significantly improved in the former condition. This suggests that the visuomotor behaviour provides cues that enhance her ability to recognize the form of the target object. In control experiments we have ruled out proprioceptive and efferent cues, and therefore propose that D.F. can, to a significant degree, access the object's visuomotor representation in the dorsal stream. Moreover, we show that the grasping-induced perceptual improvement disappears if the target objects only differ with respect to their shape but not their width. This suggests that shape information per se is not used for this grasping task.     

Brain activation during immediate and delayed reaching in optic ataxia

Patients with optic ataxia after lesions of the occipito-parietal cortex demonstrate gross deviations of movements to visual targets in their peripheral visual field. When the same patients point to remembered target locations their accuracy improves considerably. Taking into account opposite findings in a single patient suffering from visual form agnosia due to bilateral occipito-temporal lesions (D.F.), this paradoxical improvement was attributed to brain structures outside the dorsal stream, and supposed to be specifically associated with delayed movement execution. This conclusion was based on the still unverified assumption that the dorsal system is almost completely lacking any localization function in patients with optic ataxia who demonstrate the paradoxical delay effect. We thus investigated brain activity associated with immediately executed and delayed movements in a patient with optic ataxia due to extensive bilateral lesions (I.G.) and in 16 healthy subjects using functional magnetic resonance imaging. Our analysis revealed robust and indistinguishable activation of intact dorsal occipital and parietal areas adjacent to the patient's lesions for both types of movements. In healthy subjects, we found the same visuomotor network activated during immediate and delayed movements as well as additionally higher signal increases for movements to visible targets than for delayed movements in bilateral occipito-parietal and occipito-temporal areas. Our results suggest that in healthy subjects as well as in the optic ataxia patient I.G. dorsal areas are not only involved in immediate but also in delayed reaching. This observation questions the hypothesis that residual visuospatial abilities in patients with optic ataxia could only be mediated by a system outside of the dorsal stream.     

Visuospatial neglect in action

It is well established that patients with hemispatial neglect present with severe visuospatial impairments, but studies that have investigated visuomotor control directly have revealed diverging results, with some investigations finding impairments mirroring the perceptual difficulties of these patients, while others have shown that such neglect patients perform relatively better in action tasks. In this review we attempt to reconcile these diverging findings, addressing differences in the type of visuomotor tasks studied but also highlighting the diverging neuroanatomy that seems to be driving the differences in performance. We argue that there are different types of actions and that these in turn depend on different cortical networks (Goodale, Westwood, & Milner, 2004; Milner & Goodale, 2006). Patients with visuospatial neglect, in contrast to patients with optic ataxia, are relatively unimpaired at performing target-directed tasks even towards stimuli located in their 'neglected' field. We relate these findings to the view that for the on-line guidance of action, spatial information is coded in egocentric coordinates and depends on the visuomotor networks of the visual dorsal stream. Furthermore, based on recent lesion-symptom mapping studies, we postulate that deficits in on-line actions that are observed after right-brain damage are associated with damage to the visuomotor control network, in particular with damage to the basal ganglia, frontal and parieto-occipital regions. On the other hand, clear neglect-specific deficits emerge when the action is off-line and not directly target-driven, thus requiring relational metrics or scene-based coordinates (as is the case for example in delayed and mirrored (anti-pointing) reaches). We review recent studies that support our argument that such deficits in off-line actions are associated with damage to occipito-temporal and parahippocampal cortex, perhaps as part of the ventral visual stream or areas where information from the two visual streams is combined.     

Dorsal and ventral stream interaction: contributions from optic ataxia

In monkeys and humans, two functionally specialized cortical streams of visual processing emanating from V1 have been proposed: a dorsal, action-related system and a ventral, perception-related pathway. Traditionally, a separate organization of the two streams is assumed; the extent of functional interaction is unknown. After lesions of the dorsal stream in patients with optic ataxia, it has recently been shown that the ventral perception-related system might contribute to visuomotor processing if movements rely on remembered target positions. The ventral pathway thus seemed to participate in goal-directed movements, a function that previously has been assigned exclusively to the dorsal stream. We wondered whether different types of pointing movements are controlled by switching between two separated cortical pathways or whether a variable interaction of interconnected systems should be assumed. Our study investigated two acute stroke patients with optic ataxia following lesions of the dorsal stream in a delayed pointing task. The delays ranged from 0 to 10 sec. The patients' pointing error decreased in a linear manner with the length of time. The finding suggests a gradual change between dorsal and ventral control of reaching behavior, rather than a sudden switch between two separated cortical processing streams. Although our observations with two patients require further validation, the results suggest that the ventral and dorsal systems interact closely in the sensorimotor control of reaching behavior.     

Dissociating 'what' and 'how' in visual form agnosia: a computational investigation.

Patients with visual form agnosia exhibit a profound impairment in shape perception (what an object is) coupled with intact visuomotor functions (how to act on an object), demonstrating a dissociation between visual perception and action. How can these patients act on objects that they cannot perceive? Although two explanations of this 'what-how' dissociation have been offered, each explanation has shortcomings. A 'pathway information' account of the 'what-how' dissociation is presented in this paper. This account hypothesizes that 'where' and 'how' tasks require less information than 'what' tasks, thereby allowing 'where/how' to remain relatively spared in the face of neurological damage. Simulations with a neural network model test the predictions of the pathway information account. Following damage to an input layer common to the 'what' and 'where/how' pathways, the model performs object identification more poorly than spatial localization. Thus, the model offers a parsimonious explanation of differential 'what-how' performance in visual form agnosia. The simulation results are discussed in terms of their implications for visual form agnosia and other neuropsychological syndromes.     

The role of V5/MT+ in the control of catching movements: an rTMS study

Milner and Goodale described a model which distinguishes between two visual streams in the brain. It is claimed that the ventral stream serves object recognition (i.e. vision for perception), and the dorsal streams provides visual information for the guidance of action (i.e. vision for action). This model is supported by evidence from the domain of spatial vision, but it remains unclear how motion vision fits into that model. More specifically, it is unclear how the motion complex V5/MT contributes to vision for perception and vision for action. We addressed this question in an earlier study with the V5-lesioned patient LM. Can a motion-blind patient reach for moving objects? We found that she is not only impaired in perceptual tasks but also in catching, suggesting a role for V5/MT+ in vision for both perception and action. However, LM's lesion goes beyond V5/MT+ into more dorsal regions. It is thus possible, that the catching deficit was not produced by damage to V5/MT+ itself. In this case, one would expect that selective interference with V5/MT+ would have no effect on catching. In the present study we tested this prediction by applying rTMS over V5/MT+ of the left hemisphere while healthy subjects were either performing a catching or a reaching task. We found that V5-TMS reduced the speed of the catching but not the reaching response. These results confirm that V5/MT+ is not only involved in perceptual but also in visuomotor tasks.     

Symptoms and lesion localization in visual agnosia]

There are two cortical visual processing streams, the ventral and dorsal stream. The ventral visual stream plays the major role in constructing our perceptual representation of the visual world and the objects within it. Disturbance of visual processing at any stage of the ventral stream could result in impairment of visual recognition. Thus we need systematic investigations to diagnose visual agnosia and its type. Two types of category-selective visual agnosia, prosopagnosia and landmark agnosia, are different from others in that patients could recognize a face as a face and buildings as buildings, but could not identify an individual person or building. Neuronal bases of prosopagnosia and landmark agnosia are distinct. Importance of the right fusiform gyrus for face recognition was confirmed by both clinical and neuroimaging studies. Landmark agnosia is related to lesions in the right parahippocampal gyrus. Enlarged lesions including both the right fusiform and parahippocampal gyri can result in prosopagnosia and landmark agnosia at the same time. Category non-selective visual agnosia is related to bilateral occipito-temporal lesions, which is in agreement with the results of neuroimaging studies that revealed activation of the bilateral occipito-temporal during object recognition tasks.     

Immediate and delayed actions share a common visuomotor transformation mechanism: A prism adaptation study

It is a key assumption of the perception/action model that the dorsal stream relies on current visual input and does not store visual information over an extended period of time. Consequently, it is expected that action which is guided by memorized visual information, so-called delayed action, cannot use information from the dorsal stream but must rely instead on the ventral stream input. However, it is currently unclear how the information from the ventral stream can be used to guide an action. This issue is particularly challenging given that the perception/action model also assumes that ventral stream input is not particularly useful for guiding actions since the information it provides is coded relative to the visual scene and not relative to the observer. We describe two possible solutions to this problem and suggest that they can be tested using the prism adaptation paradigm. Subjects in our study were adapted to optical prisms using either an immediate or a delayed pointing task. In both cases, subjects showed the typical post-exposure negative aftereffect. Moreover, there was almost complete transfer of the aftereffect between immediate and delayed pointing. This is particularly surprising given the long history of findings showing little transfer between motor tasks for which separate neural representations are assumed. In this context our findings suggest a substantial overlap in the visuomotor transformation processes used for immediate and delayed pointing.     

Immediate and delayed reaching in hemispatial neglect

Milner and Goodale (The visual brain in action, Oxford: Oxford University Press, 1995; The visual brain in action, 2nd ed., Oxford: Oxford University Press, 2006) propose a model of vision that makes a distinction between vision for perception and vision for action. One strong claim of the model is that the dorsal stream's control of action is designed for dealing with target stimuli in the ‘here and now’, yet when time is allowed to pass and a reaction has to be made on the basis of a visual memory, the ventral stream is required for successful performance.Regarding the syndrome of hemispatial neglect, Milner and Goodale further claim that the visual dorsal stream is relatively spared in these patients. In the current study we tested whether neglect patients would indeed be unimpaired in immediate pointing, yet show inaccurate pointing in a condition where a delay is interposed between the presentation of the stimulus and the response signal (in particular in left space). We tested the ability of nine neglect patients (and healthy and right hemisphere no neglect control groups) to perform reaches towards immediate and delayed targets, placed in left, central and right locations. Neglect patients showed no accuracy impairments when asked to perform an immediate action. Conversely, when pointing towards remembered leftward locations, they markedly overshoot the target or failed to initiate a reach altogether. These results confirm that patients with neglect are not specifically impaired when performing ‘here and now’ actions, but rather present deficits when the visuomotor task taps into more perceptual ‘off-line’ representations thought to depend on ventral visual stream activation.     

Attention for action?: Examining the link between attention and visuomotor control deficits in a patient with optic ataxia

The classic definition of ‘pure’ optic ataxia suggests that these patients’ visuomotor impairments are independent of perceptual or attentional deficits. More recent work suggests that some patients with optic ataxia also have difficulty attending to targets in their ataxic field. Thus, an important question is whether these attentional deficits might be related to the well-known problems in visuomotor control evident in these patients. To investigate this question we had controls (N = 5) and CF, a patient with optic ataxia in his left visual field, perform tasks that required them to detect or reach towards targets presented in either central vision, or at different target eccentricities in the periphery. As expected, CF was less accurate than controls when reaching to targets in his ataxic (left) visual field, and was much slower than controls to detect the presence of targets in his ataxic field. The reaction times to lift the hand in the pointing and the detecting conditions were correlated in the ataxic field of patient CF, suggesting a common attentional deficit in both tasks. Importantly, although CF was slower to detect targets in the ataxic field, and less accurate to reach towards those same targets, the two deficits did not follow the same pattern. Specifically, only reaching errors in the ataxic field were strongly modulated by target eccentricity. These results suggest that dorsal posterior parietal lesions result in attention and visuomotor control problems in optic ataxia that arise from damage to independent mechanisms.     

Visuomotor performance in a patient with visual agnosia due to an early lesion

We tested a patient with visual agnosia who had suffered severe bilateral brain damage early in life, on a series of visuomotor tasks. The broad pattern of results confirms that S.B., like the extensively tested patient D.F., shows an impressive array of preserved skills, despite his severe perceptual problems. Also like D.F., S.B. shows certain subtle visuomotor difficulties that can be related to the idea that his partially intact occipito-parietal areas are unable to benefit from interactions with the apparently severely damaged occipito-temporal regions. Unlike D.F., however, he is able to make accurate discriminations of simple visual features, such as object width and orientation, albeit with very slow response times. We hypothesize that several factors such as the early onset of S.B.'s lesion and the long period since his brain lesion have allowed his brain to compensate to a degree what has been impossible in D.F., whose brain damage occurred in adulthood. This may include an element of 'rewiring' and self-monitoring of visuomotor processes that allow S.B. to achieve perceptual access to visual information processed in the dorsal stream: information that is normally only available for on-line visuomotor control.     

Deficits in peripheral visual attention in patients with optic ataxia

Earlier research has suggested that optic ataxia, a deficit in reaching in peripheral vision, can be isolated from Balint's syndrome as it is primarily a visuomotor disorder, independent of perceptual or attentional deficits. Yet almost no research has examined the attentional abilities of these patients. We examined peripheral visual attention in two patients with unilateral optic ataxia. Results indicated that both patients were slower to respond to targets in their ataxic visual field, irrespective of cuing condition (i.e. validly, invalidly, and no cue conditions), consistent with an overall decrease in the salience of stimuli in the ataxic field. Attentional deficits in peripheral vision are therefore an important factor to consider when examining visuomotor control deficits in optic ataxia.     

Correlated deficits of perception and action in optic ataxia

Optic ataxia, following dorsal stream lesions, is characterised by impaired visuomotor guidance. Recent studies have found concurrent perceptual deficits, but it is unclear whether these are functionally related to the visuomotor symptoms. We studied the ability of a well-documented patient (IG) with bilateral optic ataxia to react to sudden target jumps by correcting ongoing reaches or by explicitly reporting the jump direction. IG showed deficient reach corrections, especially for target jumps to the visual periphery, and was similarly slow to discriminate the same jumps perceptually. Across six test conditions, in which the retinal locations of target jumps were varied, her perceptual slowing mirrored her reaching deficit precisely. These findings confirm perceptual impairments after dorsal stream lesions, and imply a shared functional basis with the classical visuomotor symptoms of optic ataxia. Additionally, we show that the online correction deficit is determined dually by the retinal location to which the reach must be diverted, and the location to which it is initially directed. We suggest that this deficit, and its perceptual counterpart, can be traced to a slowed contralesional orienting of attention in optic ataxia.     

Electrophysiological evidence for selective impairment of optic flow perception in autism spectrum disorder

People with autism spectrum disorder (ASD) often show inferior global motion performance with superior performance in detail form perception, suggesting dysfunction of the dorsal visual stream. To elucidate the neural basis of impaired global motion perception in ASD, we measured psychophysical threshold and visual event-related potentials (ERPs) with a 128-channel system in 12 ASD and 12 healthy control adults. Radial optic flow (OF) and horizontal motion (HO) were used as the visual stimuli. The former was related to the ventro-dorsal stream formed by the inferior parietal lobule, while the latter was conveyed from the dorso-dorsal stream formed by the superior parietal lobule. No significant group differences were observed in the motion thresholds for both OF and HO. N170 and P200 were elicited as major components of ERPs in both groups. However, the latencies of both components for OF but not HO were significantly prolonged in ASD compared with the control group. Our ERP results suggest that ASD has a selective impairment for OF processing even though the psychophysical thresholds are preserved. Therefore, we provide the first electrophysiological evidence for altered function of the higher-level dorsal visual stream in ASD, specifically the ventro-dorsal stream closely related to OF perception.     

Impaired delayed but preserved immediate grasping in a neglect patient with parieto-occipital lesions

Patients with optic ataxia, a deficit in visually guided action, paradoxically improve when pantomiming an action towards memorized stimuli. Visual form agnosic patient D.F. shows the exact opposite pattern of results: although being able to grasp objects in real-time she loses grip scaling when grasping an object from memory. Here we explored the dissociation between immediate and delayed grasping in a patient (F.S.) who after a parietal-occipital stroke presented with severe left visual neglect, a loss of awareness of the contralesional side of space. Although F.S. had preserved grip scaling even in his neglected field, he was markedly impaired when asked to pretend to grasp a leftward object from memory. Critically, his deficit cannot be simply explained by the absence of continuous on-line visual feedback, as F.S. was also able to grasp leftward objects in real-time when vision was removed. We suggest that regions surrounding the parietal-occipital sulcus, typically damaged in patients with optic ataxia but spared in F.S., seem to be essential for real-time actions. On the other hand, our data indicates that regions in the ventral visual stream, damaged in D.F but intact in F.S., would appear to be necessary but not sufficient for memory-guided action.     

Do visual illusions probe the visual brain? Illusions in action without a dorsal visual stream

Visual illusions have been shown to affect perceptual judgements more so than motor behaviour, which was interpreted as evidence for a functional division of labour within the visual system. The dominant perception-action theory argues that perception involves a holistic processing of visual objects or scenes, performed within the ventral, inferior temporal cortex. Conversely, visuomotor action involves the processing of the 3D relationship between the goal of the action and the body, performed predominantly within the dorsal, posterior parietal cortex. We explored the effect of well-known visual illusions (a size-contrast illusion and the induced Roelofs effect) in a patient (IG) suffering bilateral lesions of the dorsal visual stream. According to the perception-action theory, IG's perceptual judgements and control of actions should rely on the intact ventral stream and hence should both be sensitive to visual illusions. The finding that IG performed similarly to controls in three different illusory contexts argues against such expectations and shows, furthermore, that the dorsal stream does not control all aspects of visuomotor behaviour. Assuming that the patient's dorsal stream visuomotor system is fully lesioned, these results suggest that her visually guided action can be planned and executed independently of the dorsal pathways, possibly through the inferior parietal lobule.     

The end point of the ventral visual stream: face and non-face perceptual deficits following unilateral anterior temporal lobe damage

While it has been claimed that the ventral visual stream ends in the inferior aspects of the anterior temporal lobe (ATL), little is known about whether this region is important for visual perception. Here the performance of two patients with unilateral ATL damage was assessed across four visual perception tasks that parametrically varied stimulus similarity. Patients performed normally on difficult judgments of circle size or face age but were impaired on face identity and dot pattern matching tasks. Portions of the ATL, most likely the ventral surface, may have a functional role in visual perception tasks requiring detailed configural processing, most commonly used to discern facial identity.