Two visual systems model

consciousnessSpatial coordinates and phenomenology in the two visual systems model

Pierre Jacob & Frederique de Vignemont, Institut Jean Nicod & New York University

In:- Perception, Action & Consciousness (2010) – Eds. – Gangopadhyay, N., Madary, M. & Spicer, F.

Goodale and Milner introduced the concept of the ‘two visual systems’ model in 1992. This involves a functional distinction between vision used for perception, involving the separation of an object from its background, and vision used for action. In this proposal, the ventral stream is related to perception and is conscious, while the dorsal stream is related to action, and is unconscious.

Goodale and Milner’s research relates particularly to a patient known as D.F. who is impaired in the identification of the shape and size of objects, but who can reach and grasp objects accurately. The tests to establish this were manual rather than verbal, therefore the results cannot be accounted for by problems with language processing. This phenomenon is regarded by Goodale and Milner as being similar to blindsight, which is also usually related to the distinction between the ventral and the dorsal stream.

Other patients, who have damage to the dorsal stream, but not the ventral, can recognise objects but have difficulties in reaching to grasp them. This is argued to point to two independent methods for processing the same initial visual stimulus instantiated in the ventral and dorsal streams. The ventral stream projects from the primary visual cortex to the inferior temporal cortex, while the dorsal also projects from the primary visual area but to the superior parietal area. The ventral is concerned with perception, while the dorsal is concerned with spatial action. Further to this the ventral appears to be related to conscious processing and the dorsal to unconscious processing.

In reaching and grasping, as performed by the dorsal stream, the subject uses egocentric coordinates based on their own body for an action performed by the body. The ability of the patient D.F. to grasp objects is assumed to use egocentric coordinates based on her hands. Perceptual judgements are flexible in being able to use either egocentric coordinates or allocentric coordinates based on an object external from the body. Allocentric perception is about comparisons between external objects, and as such it is prone to visual illusions. New visual perceptions may also be linked to older perceptions stored in the memory. Perception handled by the ventral stream enlarges the subject’s knowledge of their surroundings.

A counter argument to the two visual systems model has been that the distinction is not between perception and action but between allocentric and egocentric approaches. However, the authors argue that where the patient D.F. failed in certain tasks, it was because she was using the manual-dorsal information to attempt to perform a perceptual task. D.F. is viewed as being able to detect a target on the basis of egocentric coordinates based on her fingers, but her dorsal stream is not thought to be capable of providing spatial information that required an allocentric frame of reference.

One view of perception is that it allows a target to be discriminated from its background and from other potential targets. After this has been achieved vision-for-action can take over the control of actual movements. There is some argument as to whether patients with ventral stream damage are actually aware of the shape and size of objects, but are simply unable to report them. However, the authors argue that in cases where perceptual illusions might arise, there is no cognitive dissonance apparent between the ability to grasp an objection and some perception that might be at odds with what is grasped, meaning that there is in fact no conscious visual perception. However, the task of grasping by itself allowed the patient to gain an above chance level of locating the position of the object. Moreover a further experiment showed that the patient D.F. could actually only discriminate the width of the objection in question, but could not describe the overall shape. In conclusion, the authors consider it unlikely that the patient D.F. is visually aware of the shapes and features of the objects that she can grasp.

Tags: , Posted by

Leave a Reply