Talk:What and where pathways

Reviewer A

Altogether this is a nice little review of the subject. Here are some weak points in need of revision.

1. “What type of stimulus representation is impaired after IT lesions? A great deal of research has been devoted to this question. Generally, such experiments measure whether monkeys can discriminate different stimuli, before and after experimental lesions in IT cortex – and from any such differences, object-processing mechanisms are inferred. A current summary might be that the IT cortex represents aspects of the intrinsic shape of the stimulus that are useful for recognition (e.g., the salient features in a face and their relative positions), generalizing across idiosyncratic aspects of stimulus appearance that depend on specific viewing conditions.” This is misleading, indeed wrong, since it seems to apply to both humans and monkeys Many types of stimulus representations have been impaired after IT lesions in monkeys including discrimination of objects, patterns, colors and size. The only criterion is that the discrimination task is relatively difficult as measured by the performance of normals matched in experience. There is no evidence that “intrinsic “ shape is crucial. Indeed the ‘face’ example is particularly misleading as there is no evidence at all for a specific impairment in face discrimination after IT lesions in monkeys.

2. The OA, OB, OC terminology introduced by von Economo was useful in its day. However the modern delineation of these areas into V1, V2, V3 V4, and MT etc makes these earlier designations of interest only to the historian and very confusing for the reader of a general brief review such as this one. Fig 2 a should be dropped

3. VP is now well established as the ventral part of V3 and should be called that in the figure.

4. There should be some indication that the “two stream” idea is an over simplification and that the “dorsal” stream is divided into a dorsal one (many parietal areas) involved in space and one running down STS involved in movement/. Indeed early parts of the ventral stream have pattern and color components

5. Fig. 3 is not schematic: it is totally imaginary and speculative. There is good evidence for clustering of properties in IT and maybe some for vertical organization but there is no evidence for the columns shown in the figure, except maybe for the face ones.

6. An ordinary reader will be confused about the relation of LOC in humans and V4 and IT in monkeys. This should e clarified even if only to say explicitly that it is unclear.

7. There is no good evidence for IT cells sensitive to any objects except for faces and face components and hands (or body parts) and not for “other biological stimuli.”

8. ..‘evoked potential studies in humans with implanted electrodes have also shed light on the mechanisms by which the primate brain represents objects.” This needs expansion or at least a reference.

Charlie Gross

Revisions for Reviewer A

1. The following text was deleted:

"What type of stimulus representation is impaired after IT lesions? A great deal of research has been devoted to this question. Generally, such experiments measure whether monkeys can discriminate different stimuli, before and after experimental lesions in IT cortex – and from any such differences, object-processing mechanisms are inferred. A current summary might be that the IT cortex represents aspects of the intrinsic shape of the stimulus that are useful for recognition (e.g., the salient features in a face and their relative positions), generalizing across idiosyncratic aspects of stimulus appearance that depend on specific viewing conditions (see below)."

2. Figure 2A was deleted.

3. Will list VP as follows: “ventral part of V3”

4. Added the following text:

"Although the two-stream idea has served as a useful scheme with which to understand the organization of the visual system, it is an oversimplification in several ways. For instance, some areas in the dorsal stream exhibit selectivity for simple, two-dimensional geometric shapes that are comparable to those observed in the ventral pathway (Sereno and Maunsell, 1998)."

5. Deleted some text and figure, so text reads now:

"This latter idea is supported by evidence suggesting that IT cells are anatomically grouped into functionally similar patches, columns, or areas (see below). Based on findings from single-unit and optical recording, Tanaka (1996) has suggested that IT cells that respond to common visual features are grouped together into cortical columns that run perpendicular to the cortical surface. Newer evidence from functional magnetic resonance imaging in awake monkeys is consistent with the notion that cells with common functional properties are grouped together in IT cortex (see below) – although details of this organization remain incompletely understood."

6. Added the following text (in bold): Later neuroimaging studies revealed that the cortical regions activated by such stimuli includes classical IT cortex in macaques (Tsao et al., 2003; Denys et al., 2004); see Figure 6 – note, however, that the relationship between LOC in humans and areas V4 and IT in monkeys is unclear and is the object of current studies.

7. Deleted “other biological stimuli”.

8. Added reference in bold:

More recently, neuroimaging and evoked potential studies in humans with implanted electrodes (Allison et al., 1999) have also shed light on the mechanisms by which the primate brain represents objects.

Revisions for Reviewer B

1. Reviewer's comment: does the monkey lesion have to be bilateral? what does this mean given that a single hemispheric lesion in humans will suffice for agnosia?

The lesion in monkeys needs to be bilateral to generate the impairment. The lesion in humans producing agnosia is also typically bilateral, although a few cases with unilateral lesions (restricted to right hemishpere) have been documented. This is likely because visual processing/recognition in humans has become right lateralized. Some conjecture (e.g., Behrmann) that this is due to the development of language cortex (including basal temporal regions) in the left hemisphere, pushing visual processing into the right. However, we believe that these issues would unduly complicate the text and have opted to leave the text in the original form.

2. Reviewer’s comment: one of the critical issues that perhaps needs to be addressed is that the binary distinction originally thought to differentiate the ventral and dorsal streams is no longer so clear: for example, many studies reveal evidence of 'where' processing in ventral stream both for parts of an object and how they relate to each other but also for frank locations. By the same token, recent fMRI adaptation studies show adaptation for objects, both 2d and 3d in parietal cortex, suggesting object representation in the so-called 'where' cortex.

We have added the following text:

"Although the two-stream idea has served as a useful scheme with which to understand the organization of the visual system, it is an oversimplification in several ways. For instance, some areas in the dorsal stream exhibit selectivity for simple, two-dimensional geometric shapes that are comparable to those observed in the ventral pathway (Sereno and Maunsell, 1998)."

3. Reviewer's comment:there is growing recognition that there might be substantial computation in V1 too though such as segmentation and context effects rather than simply sensitvity to oriented line edges etc

We have added the following text:

"(note, however, that V1 is implicated in complex processes of segmemtation and context-dependent effects [see Gilbert and Sigman, 2007])"

4. Reviewer's comment: this is a bit confusing as some of this material has already been covered above and is now reintroduced so this is repetitive.

We have deleted this part of the caption text:

"According to this model, there are two major processing pathways, or “streams,” in the visual cortex, both originating in the primary visual cortex: a ventral stream, directed into the temporal lobe and crucial for the identification of objects, and a dorsal stream, directed into the parietal lobe and crucial for spatial perception and visuomotor performance."

5. Reviewer's comment: did I miss it or is there a discussion of receptive field size needed here somewhere?

We have added the following text:

"At the same time, receptive field sizes increase as one advances through the occipitotemporal pathway."

6. Reviewer's comment:relative to the detail assigned to the discussion of ventral cortex, the discussion of dorsal cortex is sparse and lacks detail. This makes the review somewhat unbalanced.

While this is true, we have chosen to focus on the dorsal stream for brevity. This is now explicitly stated in the revised overall summary:

"What and where pathways refer a proposed organization of the visual system based on neuroanatomical, electrophysiological, and lesion studies. It describes two information processing streams originating in the occipital cortex, dorsal (which goes to parietal cortex) and ventral (which goes to temporal cortex), which exhibit relative specialization in object recognition (what) and spatial vision (where). This entry will present the general ideas of the what-and-where proposal, with an emphasis on the organization of the what processing stream."