Analysis of figure-ground organization is a crucial stage of visual processing that almost certainly depends on reciprocal interaction among neurons distributed widely within and across the visual areas of the cerebral hemisphere. If the neural processes underlying figure-ground organization indeed are emergent (Center Aim 1) and distributed (Center Aim 2), in the sense that they depend on interaction between multiple areas, then single-neuron studies of figure-ground organization should be valuable as a means for constraining and testing PDP models (Center Aim 3). Accordingly, the general aim of the proposed project is to characterize patterns of distributed neuronal activity in the visual cortex of the monkey during the perceptual segregation of the visual displays into figure and ground.
The project will focus on two visual cortical areas (V2 and TE). Area V2 was selected because it contains neurons sensitive to border-ownership. Area TE was selected because it contains neurons sensitive to object identity.
Multielectrode single-neuron recording will be carried out while monkeys view ambiguous figure-ground displays in which either of the two regions can be seen as figure. The results will be analyzed to determine whether, as one region or the other is perceived as figure, neuronal activity co-varies within and across areas. Experiment 1 will assess whether the activity of multiple shape-sensitive neurons in area TE co-varies with the monkey.s report of which region is seen as figure. Experiment 2 will assess whether the activity of multiple border-ownership-sensitive neurons in area V2 co-varies with the monkeys. behavioral report. Experiment 3 will address the question whether prior training with unambiguous displays can bias behavioral report and neuronal responses elicited by the ambiguous displays. Experiment 4, involving simultaneous recording in areas V2 and TE, will address the issue of timing, asking whether there are circumstances under which neuronal activity reflecting figure-ground organization emerges first in area V2 or TE.
The results of these experiments and of modeling studies influenced by them will help to improve our current very limited understanding of neural mechanisms that underlie visual perceptual organization in healthy individuals and in patients whose vision has been compromised through brain injury.
Key Personnel:
| Name | Organization | Role on Project |
| Tai Sing Lee | Carnegie Mellon/CNBC | Principal Investigator |
| Carl R. Olson | Carnegie Mellon/CNBC | Co-PI |
