What are the effects of lesioning the ventral/dorsal stream?

Question 1

Effects of dorsal stream lesion?

Answer 1

= optic ataxia

Ratcliff & Davies-Jones (1972) measured localisation accuracy in a group of men with brain injury
- Found localisation deficits sig. correlated with high posterior parietal (dorsal) lesions

Perenin & Vighetto (1988) studied 10 optic ataxia patients
When reaching for two objects simultaneously in each hemifield, largest errors were from hands in contralesional VF
- Therefore, ataxia deficits related to hemisphere of damage.

When orienting hand through slit, right-damaged PS made errors only in contralesionion VF regardless of hand used. However, left-damaged PS displayed both VF and hand effect
- Differential hand-effect suggests there is hemispheric assymmetry in the functional organisation of motor-guided mechanisms.

Inferior parietal sulcus only region damaged in all 10 cases
- IPS critical for optic ataxia

Question 2

Effects of ventral stream lesion?

Answer 2

Goodale et al (1991) studied patient DF who had damage to B. areas 18 and 19 (ventral) and thus displayed visual form agnosia

• DF could only match card to slot once it became practical (could post a card through but could not indicate most suitable card to fit through)
• DF could not identify shapes or their width but could pick them up with correct grip
• Therefore, DF’s lesions caused deficits in conscious visual perception and recognition but not automatic visuomotor movement

Question 3

Optic ataxia and visual form agnosia compared in the same study?

Answer 3

Goodale et al (1994) tested DF (ventrolateral occipital lesion) and RV (occipitopariteal lesion) in discrimination and grasping tasks

• Discrimination task: DF could not determine if the two shapes were the same or different (RV could)
• Grasping task: RV could not pick up objects (DF could)
• Therefore, ventral and dorsal stream lesions result in dissociable deficits

Question 4

visual form agnosia + fMRI

Answer 4

James et al (2003) scanned DF while conducting object identification task and reaching & grasping task

Intact line drawings vs. scrambled line drawings…
- Unlike controls, no difference in activation

Intact colour and grayscale images vs scrambled images

• Both controls and DF showed differential activation, however, spatial pattern differed…
• DF had no LOC activations (due to lesion) and showed greater activation in striate and pre-striate cortex to scrambled images (opposite of controls)

When analysing the time-course of activation…

• Stronger activation to named objects
• This correlation between identification success and activation in these regions is suggestive of a recognition mechanism in DF in the absence of LOC

Explanation————————

Blob channel (inc. V1, V2, V4, V8) is relatively intact in DF

• Therefore, colour and texture information may be aiding object recognition of images but not line drawings
• This suggests there is separate channels for processing form information (LOC) and surface information such as colour and texture
• This latter information is perhaps what is used by DF

THEREFORE lesions to the ventral stream results in a reliance on the use of surface information in object recognition (as supported by intact pre/striate cortices) rather than the usual form information (as supported by the damaged LOC)

Question 5

Premise of the argument against the use of mentioned studies in supporting the two-streams hypothesis

Answer 5

(Pisella et al, 2006) Behavioural evidence for double-dissociation is incomplete as those with OA and VFA have not been tested under appropriately matched conditions

Question 6

Evidence that DF shows visuomotor deficits

Answer 6

Dijkerman et al (2009) found DF does not select grip posture to minimize awkwardness during grasping, as do normal PS.
- May reflect impaired action planning (high level selection of the action) but intact action programming (low-level parameterization)

Even stronger evidence: Hesse et al (2012) found DF’s visuomotor performance is only preserved when the target is presented centrally - reaching and grasping in peripheral vision were abnormal.

• These findings replicated by Rossit et al (2018) - showed ‘remarkable innacuracy’ when reaching to peripheral targets.
• Most optic ataxic PS show most severe deficits in periphery
• Therefore, DF’s visuomotor abilities are similar to those with OA which undermines suggestion that the disorders form a double dissociation.

Question 7

Two reasons for DF showing symptoms of optic ataxia.

Answer 7

1. DF actually has OA as well as VFA. Why?…
   - DF’s superior parietal occipital cortex (SPOC) is not anatomically nor functionally intact in either hemisphere (Bridge et al, 2013)
   - OA patients often have damage in SPOC + healthy PS who receive TMS to SPOC show similar peripheral vision-motor deficits.
   - If DF does have OA, then she can no longer be considered as an appropriate single-case model for testing dissociation of the two conditions.
2. There is less dissociation between VFA and OA in general, suggesting a less dichotomy between processing action and perception.

Question 8

Evidence that optic ataxia PS show symptoms of VFA

Answer 8

(Pisella et al, 2009)
OA PS have in tact object perception in central vision. However, when OA PS were assessed for object perception in peripheral vision, there was significant deficiencies similar to those with VFA .

Question 9

What do the contradictions of the double dissociation tell us?

Answer 9

Together, this evidence suggests that each patient population is displaying deficits associated with the supposedly opposite condition

This suggests that the assumption that there is separate ‘what’ and a ‘where’ pathways is an oversimplification.

If not two, separate, hierarchically organised streams, then what??

• deHaan and Cowey (2011) propose a ‘patchwork’ model in which there are no functional highways and no presupposed hierarchy between the different visual abilities.
• Instead, many visual brain areas may arrange themselves from task to task into novel functional networks (McIntosh & Schenk, 2009)

Question 10

Conclusion

Answer 10

Initially, there seemed clear evidence that PS with damage to ventral areas showed deficits in visual perception which dissociated from the deficits in visuomotor action resulting from damage to dorsal areas of the visual system.

However, recent work has discovered that this dissociation is less clear when the two patient populations are tested under matched conditions
- E.g DF shows dificts akin to those in optic ataxia and those with object ataxia show deficits aking to those with visual form agnosia.

Although the two-stream model has served as a useful heuristic by which to study different aspects of visual processing, the breakdown in the double-dissocation suggests other models of visual processing should be considered (e.g. patchwork model)

In summary, although lesions to the ventral and dorsal stream appear to results different deficits, each of these populations show deficits associated with the other population when tested with more stringent methodology
- Therefore, for a clear double dissociation to be made, pure cases of OA and VFA need to be found that show no deficits associated with the other condition.