Computations by networks of neurons: fMRI adaptation studies in monkeys

A great deal is understood about the properties of single neurons processing visual information. In contrast, less is known about the collective characteristics of networks of cells that may underlie sensory capacities of animals. By measuring the blood oxygenation level-dependent (BOLD) signal in the macaque cortex we studied the emergent properties of populations of neurons processing motion across different brain areas. We used a visual adaptation paradigm to localize a distributed network of visual areas which process information about direction of motion, and also studied the dynamics of adaptation of the bold signal elicited by moving stimuli. We found that the BOLD signal in areas MT and V2/V3 adapted faster than in V1 reflecting the difference in motion processing between these areas. Moreover, the strength of the directionally selective bold signal in V1 was much greater than the one estimated on the basis of established facts from single cell electrophysiology. We propose an hypothesis that may account for this difference based on the postulate that neuronal selectivity is a function of the state of adaptation and therefore neurons classically thought to lack information about certain attributes of the visual scene, may nevertheless receive and process this information. The implementation of this hypothesis can arise as a result of intra- and inter-area cellular connections, such as feedback from higher areas. This network property may be a universal principle whose computational goal is to enhance the ability of neurons in earlier visual areas to adapt to statistical regularities of the input and therefore increase their sensitivity to detect changes along these stimulus dimensions