Are optic ataxia and visual form agnosia complementary parts of a double dissociation, or just two different neuropsychological disorders?
Dissociation, hence abandonment of the monolithic view of visual processing, was primarily proposed by Schneider in 1969. Subsequently, Ungerleider & Mishkin (1982) distinguished between the differing processing strategies of instead, two proposed streams, as being associated with appreciation of object qualities, and location. Although influential, this model was superseded by Milner & Goodale (1992, 1995), who postulate stream segregation in a comparable, but pivotally diverse manner. Their modified model corresponds to disparity between visuomotor control and perceptual representation. The most compelling evidence to support this dissociation comes from neuropsychology. After studying patients with lesions in either temporal or parietal regions (namely the ventral and dorsal streams), Milner & Goodale noted opposing behavioural consequences. In light of this, they concluded the ventral stream (from early visual areas to the temporal lobe) is devoted to vision for perception, damage customarily resulting in optic ataxia (OA). Dorsal stream processing (early visual areas to the posterior parietal cortex (PPC)) is thought to facilitate vision for action, visual form agnosia (VFA) being the result of damage to this stream. Because the characteristic deficits demonstrated by patients with these conditions contrast so well, they provide crucial evidence for double dissociation, the very scaffolding for the dichotomic view of the visual system.
For the purpose of this essay, I will investigate supporting, and challenging evidence concerning the proposed double dissociation between OA and VFA, thought to form the majority of evidence in favour of Milner & Goodale’s dual system hypothesis. I will offer conclusions with reference to pertinent, empirical studies that scrutinise evidence involving differences in: the behaviour, spatial processing, and temporal abilities of the two disorders.
Before continuing, it is crucial to investigate what is meant by the term ‘double dissociation’. Dissociation can be described as evidence that separate systems mediate two different variables (Rossetti & Revnsuo, 2000). In terms of dissociation in the human visual system, much evidence comes from single-case lesion studies. A major difficulty for relying on evidence of this calibre is that behavioural consequences are a potential by-product of reorganisation of the brain tissue surrounding the lesion, rendering the functional specificity of lesion location questionable. More appropriate evidence for functional autonomy of two related processes therefore comes from double dissociations; “one case of A and not B, and another case of B and not A” (Bridgeman, 2000, p.20). Coined by Tueber (1955), double dissociations are believed to help to elaborate on neurophysiological understanding of the brain, although criticisms of this claim are evident (Passingham, Stephan & Kotter, 2002). Confirmation of the proposed double dissociation between VFA and OA necessitates the satisfaction of various criteria. Firstly, one must empirically diagnose each condition. Secondly, these diagnosed patients must be tested in identical conditions. Lastly, exactly opposing behavioural consequences must be demonstrated by the patients with each disorder. If these specifications are met, it is reasonable to accept that OA and VFA form complimentary parts of a double dissociation. If not, discernment of double dissociation between the disorders cannot be justified.
Initial contrasting differences between OA and VFA evidence themselves in behaviour. Patient D.F. has a bilateral lesion of the occipito-temporal cortex, resulting in severe VFA. Her symptoms include the inability to discriminate line orientation and simple geometric shapes. She can however, draw from memory, name colours and distinguish between surface materials and shades, which can sometimes facilitate object recognition. Interestingly, D.F. can use visual cues she is consciously unaware of to achieve veridical reaching and grasping of objects.
Patients with OA typically have damage to the parietal-occipital junction (POJ) in the proposed dorsal stream (Karnath & Perenin, 2005) and demonstrate the opposite deficits. They can easily describe orientations and identify objects but are unable to utilise this information to enable proficient interaction with them.
These behavioural consequences are compliant with Milner & Goodale’s model of vision for action and vision for perception. Studies using fMRI accolade this particular theory of double dissociation, revealing that D.F. displays different regions of activation than controls when viewing pictures, but similar when grasping objects (James et al, 2003). Even areas in healthy controls were differentially activated depending on whether the task was perceptual or action based (Cavina-Pratesi, Goodale, & Culham, 2007).
In light of this evidence, it seems acceptable to assume a double dissociation between these two disorders. However, many researchers refute this conclusion and offer reason to repudiate the idea that OA and VFA form complimentary parts of a double dissociation.
Firstly, it has been suggested that the symptoms exhibited by patients are not as divergent as first projected. D.F’s action abilities are limited. She can execute accurate actions consistently when asked to post a simple plaque, but makes errors with T-shaped objects (Goodale et al, 1994). Critics have argued that if OA and VFA are complimentary parts of a double dissociation corresponding to perception and action, one would expect D.F. to perform adept actions regardless of object complexity, just as OA patients identify objects irrespective of intricacy. In the same vein, it was noted that D.F. also fails to plan actions based on certain visual cues and tends to grasp in unconventional, often uncomfortable ways (Dijkerman et al, 2009). Researchers have since however, suggested that the dorsal stream is capable of processing simple, not complex objects (McIntosh, Dijkerman, Mon-Williams, & Milner, 2004) and that distinction between the two conditions may correspond to action planning (ventral) and action programming (dorsal) (Milner & Goodale, 2008), thus promoting the theory of double dissociation. However, complications with this hypothesis have also been offered (see Kitadono & Humphreys, 2007), suggesting interactions between these functions.
In relation to this point, many would argue that due to ubiquitous interaction between the two streams, and hence between behavioural consequences of damage to them, one cannot assume double dissociation. ‘Ventral’ perception clearly triggers and inhibits action, as we constantly decide upon actions based on what we perceive. Additionally, recognition of objects also affects ‘dorsally’ controlled actions towards them. When considering the disorders in question, it has been deduced that OA patients are successful in executing actions towards familiar objects (Jeannerod et al, 1994). If double dissociation between the two disorders relies on the fact that one condition produces preserved perceptual abilities with deficits in visuomotor abilities, and the other eliciting the converse, this observation is a blatant exception. However, it can be argued that practised movements such as those executed towards familiar objects become learned and automatic and no longer require dorsal processing. Modification of the specific behavioural double dissociation present may be required.
It appears then that interactions exist. However, Milner & Goodale’s (2008) response emphasises their agreement with evident interaction between streams, but they maintain fundamentally segregation. Distinction is therefore likely to correspond to something other than perception and action. However, they redefine their interpretation of the words ‘perception’ and ‘action’ as referring to “‘unconscious’ or ‘preconscious’ perception of objects and events…that could potentially reach conscious awareness” (Milner & Goodale, 2008, p775), and, the ‘implementation of action’ respectively. Continuous modifications assist in re-evaluating the potential double dissociation between VFA and OA. However, by relaxing definition and independence of the proposed systems, it becomes harder to ever fully reject or accept theory, and may perhaps diminish the importance of new related findings, making them easily manipulated to coincide.
Another issue raised involving discrepancies over the current double dissociation concerns the fact that is it also possible to dissociate neglect, environmental-dependency syndrome, and blindsight from these disorders (Rossetti & Pisella, 2002). If confirmed, these dissociations remove any credibility of the VFA/OA double dissociation. Confirmation would also suggest that the visual system may be more complicated than proposed by Milner & Goodale. A further dissociation between reaching and grasping (e.g. Jackson et al, 2009) has been observed within the visuomotor stream. This has been supported by neuropsychological evidence revealing that some OA patients show impairments for grasping but not reaching (Binkofski et al, 1998). This suggests that the OA VFA double dissociation may not be as definitive as initially postulated, and suggests that there is at least one additional, corresponding sub-stream within the dorsal pathway. In fact, many would argue that the visual system is a “far more complex organisation with multiple parallel visual-to-motor connections” (Pisella et al, 2006, p.2734), with neuroimaging data being consistent with this claim (e.g. Rushworth et al, 2006). In light of these discoveries, it may be more accurate to view OA and VFA as two different neuropsychological disorders that result from damage to specific regions within a vast, complex network of systems.
Lastly, although many monkey studies also support the idea of double dissociation, it should be noted that damage in humans is never exclusive to specific areas. Overlap will undoubtedly occur given the very nature of lesions. Unfortunately, most of the VFA patients studied have damage caused by carbon monoxide (CO), which affects grey and white matter bilaterally. Conclusions built upon these patients may be an artefact of the specific tissues affected; therefore double dissociations observed may also be a proxy of systematic damage. Although recent research involving VFA caused by stroke (Karnath et al, 2009) by-in-large supports Milner & Goodale’s model, minor differences emphasise the difficulties concerning inferring fact from CO patients such as D.F.
The dual system theory also assumes double dissociation between the spatial coding of the ventral and dorsal streams, evidenced again, by preserved and abolished abilities demonstrated by patients with OA and VFA (Schenk, 2006). Because D.F. has preserved abilities in executing accurate actions towards objects, she must have the ability to code visual information egocentrically, using absolute metrics. In contrast, OA patients must code visual information allocentrically, using relative metrics to allow a rich visual representation to be formed in context. This theory is also supported by empirical brain imaging data from neurologically intact participants (see Valyear et al, 2006). Although this spatial dissociation appears complimentary to the widely accepted behavioural dissociation, some believe that it supersedes the former.
Schenk (2006) showed that D.F.’s perceptual performance could be improved to the standard of controls in an egocentric task, whereas her visuomotor abilities were eradicated in allocentric task conditions. Again it seems likely that the dissociation in VFA is not between perception and action, but is perhaps spatial in nature. Incongruities over the mode of visuospatial processing actually employed during particular tasks however, result in equivocal conclusions.
Additionally, to completely support a double dissociation of spatial coding strategies, each should be existent in one stream exclusively. However, evidence is suggestive of the contrary, indicating the existence of allocentric coding within dorsal stream regions (Pizzamiglio, Guariglia, & Cosentino, 1998).
Further support for this assumption comes from utilisation of illusory effects. Because OA patients are thought to code spatial information allocentrically, their actions are subsequently privy to the effects of illusion. Visual illusions have rarely been utilised in testing patients with OA or VFA but they do exist. Coello and colleagues (2008) tested patient I.G. with OA in a size contrast illusion. It was predicted that due to damage of the dorsal stream, I.G.’s performance would be sensitive to illusion for both perceptual and action based tasks. However, they noted analogous results from I.G. and the controls on three different tests.
Central versus Peripheral
Further motive to doubt the double dissociation between OA and VFA arises from the fact that these patients are often tested under different conditions (one of the criteria for confirming double dissociation). It has been reported that most reaching deficits demonstrated by OA patients, even in bilateral patients, are observed in peripheral vision (e.g. Pisella et al, 2006) and that OA patients can execute accurate reaching a grasping in central vision under basic conditions (e.g. Grea et al, 2002). However, this ability is not consistent (Goodale & Wolf, 2009). Authors in favour of this critique suggest that misreaching in central vision is a result of the ‘hand effect’; misrepresentation of proprioceptive input received from the hand (Pisella et al, 2009). If this action deficit is restricted to peripheral vision alone, it may be proposed that OA forms a dissociation between central and peripheral vision (Rossetti et al, 2003), perhaps further indicating the existence of sub-streams, while removing emphasis from behavioural double dissociations observed between OA and VFA. Event-related fMRI studies of healthy subjects support the central/peripheral dissociation theory, revealing varying activation when reaching either in central or peripheral vision (Prado et al, 2005). It may therefore be justifiable to maintain that dorsal stream involvement only occurs when movements are guided in the periphery. However, the only time we tend to interact with familiar objects is in this area, a much easier task for OA patients. This may help to explain why patients with OA are much less debilitated than VFA patients in everyday life. If OA patients need to interact with unfamiliar objects, they can compensate by slowing their movements and turning their head to ensure the object is in foveal space.
Additionally, it has been established that OA patients may exhibit perceptual difficulties in peripheral vision (Rossetti et al, 2005). If this is the case, then the fundamental dissociation proposed between OA and VFA (i.e. perception versus action) is fallible and it is more likely that these are simply two different neuropsychological disorders. If OA and VFA both exhibit perceptual and action related deficits in certain circumstances we can completely reject the double dissociation hypothesis, at least in terms of these specific functions.
It is important to consider, however, whether the central/peripheral dissociation actually opposes the double dissociation between OA and VFA. Although it suggests interaction and potential sub-streams, some evidence for double dissociation between the discussed disorders seems to remain; in general, OA patients exhibit abilities in object recognition in central vision while VFA patients do not. In fact, the central versus peripheral argument may well compliment the current dissociation, given the different neuronal properties proposed to lie in each stream. The ventral stream processes high –resolution information to accomplish accurate perception. It is therefore reasonable that its processing would not extend to low-resolution information in the periphery. It may be that further modifications must be implemented to determine the full extent of the proposed double dissociation.
Under ‘free vision’ conditions, whereby patients’ view is not restricted to central or peripheral vision, OA patients often demonstrate accurate actions towards objects. However, deficits appear when reaching for objects experimentally displaced at movement onset (Rossetti, Pisella, & Vighetto, 2003), inferring visuomotor system involvement in online correction of action and highlighting the temporal limitations of the perceptual stream.
Investigation into the temporal differences of perceptual and visuomotor processing has provided the most compelling evidence in support of the double dissociation between OA and VFA. A critical discovery made by Goodale and colleagues (1994) was that D.F.’s accurate grip scaling was removed when a delay was introduced. Later testing revealed that this was also true for reaching (Milner et al, 1999). Although this can be explained simply through decay of visual representation, OA patients exhibit a paradoxical improvement with delay (e.g. Milner et al, 1999), allowing for the slower processing constraints of ventral areas to activate.
Subsequent research suggests that automatic, on-line actions are preserved in D.F. These actions are under ‘automatic pilot’, operate on a different time-scale than those serving vision for perception and do not reach awareness (Pisella et al, 2000). Conversely, ventral processing maintained in OA patients does reach awareness.
These opposing consequences clearly support the double dissociation theory – VFA patients’ performance deteriorates with delay, while OA improves. However, despite improvement, OA performance remains inferior to that of controls, which suggests that for optimal performance in delayed actions, an interaction of ventral and dorsal processing is required. Furthermore, the issue of testing conditions again becomes relevant. The effects of delay are reported in central vision for VFA patients and peripheral for OA.
To conclude, the opposing consequences of these disorders offer persuasive evidence to support double dissociation within the human visual system. However, if an exact double dissociation was present between these disorders in terms of temporal differences, immediate action performance in VFA should be comparable to delayed actions in OA, and, this would be true when tested under identical conditions.
Confirming the double dissociation between OA and VFA is critical in providing evidence to support Milner & Goodale’s dual system hypothesis. If this double dissociation is refuted, it is possible that their entire theory should be rejected, as a large proportion of favourable evidence comes from patients with these disorders.
However, there are a number of difficulties in relying on double dissociations of this ilk when inferring neuronal, anatomical and functional separation. Firstly, behavioural dissociations do not necessarily imply different underlying neural networks. Secondly, evidence from lesion studies is always weakened by the fact that brain reorganisation, or spared regions, may be responsible for behavioural outcomes. Additionally, diagnoses of patients are not always rigorous. Most evidence against double dissociation comes from experimental work with neurologically intact participants.
If we accept that the visual system is more complicated than initially postulated, the double dissociation paradigm becomes limited in value (Jeannerod & Jacobson, 2005). Perhaps then, the dual visual system theory is more applicable to primates. As we evolve, cognitive demands increase and actions become more complex. It seems reasonable that more networks would develop to cope with this. Pisella and colleagues (2006) infer that there are in fact three parallel pathways in the human visual system; the dorso-dorsal pathway, the ventral stream pre-frontal pathway, and the ventro-dorsal pathway. The dorso-dorsal pathway is responsible for immediate visuo-motor control (disturbances resulting in OA). The ventral stream pre-frontal pathway involves “mediate” control for intention (VFA being the result of damage to this pathway). Lastly, the ventro-dorsal pathway mediates complex planning and programming. In this assumption, double dissociation can be refuted are there are more than two parallel streams in action.
Finally, conclusions will always be problematic given that the proposed streams are theoretical in nature. If you are in favour of a dual system, problems can be explained by communication between streams and the location of spared and damaged regions. If you are against it, problems become evidence for interaction. Indeed, the very search for double dissociation assumes a dualist attitude, and, given the copious number offered in research, may be too successful to posit accurate conclusions.
In summary, assuming double dissociation between OA and VFA may be detrimental in advancing understanding of the visual system. In light of the evidence above, it seems there is insufficient evidence to support the most demonstrative arguments in favour of the double dissociation, with failure to meet the aforementioned criteria. Therefore unless a unifying theory is suggested, it would appear that some basic separation is apparent, but that OA and VFA are not complimentary parts of a double dissociation, but two separate neuropsychological disorders.
Aglioti, D., DeSouza, J.F.X. & Goodale, M.A. (1995) Size-contrast illusions deceive the eye but not the hand. Current Biology, 5, 679-685.
Bridgeman, B., (2000). Interactions between Vision of Perception and Vision for Behaviour. In Y. Rossetti & A. Revonsuo (Eds), Beyond dissociation: interaction between dissociated implicit and explicit processing (pp 17-38). Amsterdam: John Benjamins Publishing Co.
Bruno, N. (2001), When does action resist visual illusions? Trends in Cognitive Neuroscience, 9, 379-382.
Carey, D.P. (2001). Do action systems resist visual illusions? Trends in Cognitive Sciences, 5, 109-113.
Cavina-Pratesi, C., Goodale, M.A. & Culham, J.C. (2007). fMRI reveals a dissociation between grasping and perceiving the size f real 3D objects. PLoS ONE 2(5): e424. doi:10.1371/journal.pone.0000424.
Coello, Y., Danckert, J., Blangero, A. & Rossetti, Y. (2008). Do visual illusions probe the visual brain?: Illusions in action without a dorsal visual stream. Neuropsychologia, 65, 390-391.
Culham, J.C., Danckert, S.L., DeSouza, J.F.X., Gati, J.S., Menon, R.S. & Goodale, M.A. (2003) Visually guided grasping produces fMRI activation in dorsal but not ventral stream brain areas. Experimental Brain Research, 153, 180-189.
Danckert, J.A., Sharif, N., Haffenden, A.M., Schiff, K.C. & Goodale, M.A. (2002). A temporal analysis of grasping in the Ebbinghaus illusion: planning versus online control. Experimental Brain Research, 144, 275-280.
Dassonville, P., Bridgeman, B., Bala, J.K., Thiem, P. & Sampanes, A. (2004). The induced Roelofs effect: two visual systems of the shift of a single reference frame? Vision Research, 44, 603-611.
Dijkerman, H.C., McIntosh, R.D., Schindler, I., Nijboer, T.C.W. & Milner, A.D. (2009). Choosing between alternative wrist postures: Action planning needs
perception. Neuropsychologia, 47, 1476-1482.
Dijkerman, H.C., Le, S., Demonet, J.F., & Milner, A.D. (2004). Visuo-motor
performance in a patient with visual agnosia due to an early lesion. Brain
Research. Cognitive Brain Research, 20, 12–25.
Franz, V.H., Gegenfurtner, K.R., Bulthoff, H.H. & Fahle, M. (2000) Grasping visual illusions: No evidence for a dissociation between perception and action. Psychological Science, 11, 20-25.
Franz, V. H., Hesse, C. & Kollath, S. (2009). Visual illusions, delayed grasping and memory: No shift from dorsal to ventral control. Neuropsychologia, 47, 1518-1531.
Goodale, M. A., Jakobson, L. S.,&Keillor, J.M. (1994). Differences in the visual
control of pantomimed and natural grasping movements. Neuropsychologia,
Goodale, M.A., Meenan, J.P., Bulthoff, H.H., Nicolle, D.A., Murphy, K.J. & Racicot, C. I. (1994). Separate neural pathways for the visual analysis of object shape in perception and prehension. Current Biology, 4, 604-610.
Goodale, M. A.,& Milner,A.D. (1992). Separate visual pathways for perception
and action. Trends in Neuroscience, 15, 20–25.
Goodale, M.A. & Westwood, D.A. (2004). An evolving view of duplex vision: separate but interacting cortical pathways for perception and action. Cognitive Opinion in Nerobiology, 14, 203-211.
Himmelbach, M., & Karnath, H.O. (2005). Dorsal and ventral stream interaction:
Contributions from optic ataxia. Journal of Cognitive Neuroscience,
Jackson, S.R., Newport, R., Husain, M., Fowlie, J.E., O’Donoghue, M. & Bajaj, N. (2009) There may be more reaching than meets the eye: re-thinking optic ataxia. Neuropsychologia, 47, 1397-1408.
Jeannerod, M., Jacob, P. (2005) Visual cognition: a new look at the two-visual systems model. Neuropychologia, 2, 301-312.
Karnath, H.O., & Perenin, M.T (2005). Cortical control of visually guided
reaching: Evidence from patients with optic ataxia. Cerebal Cortex, 15(10),
Karnath, H.O., Ruter, J., Andre, M. & Himmelbach, M. (2009). The anatomy of object recognition-visual form agnosia caused by medial occipitotemporal stroke. Journal of Neuroscience, 29, 5854-5862.
Khan, A. Z., Pisella, L., Vighetto, A., Cotton, F., Luaute, J., Boisson, D., Salemme, R., Crawford,J.D. & Rossetti,Y. (2005) Optic ataxia errors depend on remapped, not viewed, target location. Nature Neuroscience, 8, 418-420.
Milner, A.D. & Goodale, M.A. (2008) Two visual systems re-viewed. Neuropsychologia, 46, 774-785.
Milner, A.D., Paulignan, Y., Dijkerman, H.C., Michel, F., & Jeannerod, M.
(1999).A paradoxical improvement of optic ataxia with delay: New evidence
for two separate neural systems for visual localisation. Proceedings of the
Royal Society of London B, 266, 2225–2230.
Milner, A D., Perrett, D.I., Johnston, R.S., Benson, P.J., Jordan, T.R., Heeley,
D.W., et al. (1991). Perception and action in ‘visual form agnosia’. Brain,
Pisella, L., Binkofski, F., Lasek, K., Toni, I.& Rossetti, Y. (2006). No double-dissociation between optic ataxia and visual agnosia: Multiple sub-streams for multiple visuo-manual integrations. Neuropsychologia, 44, 2734-2748.
Pisella, H., Grea, H., Tilikete, C., Vighetto, A., Desmurget, M., Rode, G., Boisson, D. & Rossetti, Y. (2000) An ‘automatic pilot’ for the hand in human posterior parietal cortex: toward reinterpreting optic ataxia. Nature Neuroscience, 3, 729-736.
Pisella, L., Sergio, L., Blangero, A., Torchin, H., Vighetto, A. & Rossetti, Y. (2009) Optic ataxia and the function of the dorsal stream: contributions to perception and action. Neuropsychologia, 47, 3033-3044.
Prado, J., Clavagnier, S., Otzenberger, H., Scheiber, C., Kennedy, H. & Perenin,
M. T. (2005). Two cortical systems for reaching in central and peripheral
vision. Neuron, 48(5), 849–858.
Rizzolatti, G., Matelli, M. (2003). Two different streams form the dorsal visual system: anatomy and functions. Experimental Brain Research, 153, 146-157.
Rossetti, Y., & Pisella, L. (2002). Several ‘vision for action’ systems: A guide
to dissociating and integrating dorsal and ventral functions. In W. Prinz &
B. Hommel (Eds.), Attention and performance. XIX. Common mechanisms
in perception and action (pp. 62–119). Oxford University Press
Rossetti, Y., Pisella, L., Vighetto, A. (2003). Optic ataxia revisited: Visually guided action versus immediate visuomotor control. Experimental Brain Research, 153, 171-179.
Rossetti, Y., & Revonsuo, A. (2000). Beyond dissociations: Recomposing the
mind-brain after all? In Y. Rossetti & A. Revonsuo (Eds.), Beyond dissociation:
Interaction between dissociated implicit and explicit processing (pp.
1–16). Amsterdam: Benjamins.
Schenk, T. (2006). An allocentric rather than perceptual deficit in patient D.F. Nature neuroscience, 9, 1369-1370.
Schenk, T. & McIntosh, R.D. (2010). Discussion paper: Do we have independent visual streams for perception and action? Cognitive Neuroscience, 1, 52-78.
Van der Kamp, J., Rivas, F., van Doorn, H. & Savelsbergh, G. (2008). Ventral and dorsal system contributions to visual anticipation in fast ball sports. International Jounral of