The inferotemporal cortex of the macaque monkey mediates the recognition of objects in the visual world. The purpose of the research presented in this dissertation was to investigate the neural mechanisms underlying two poorly understood aspects of object recognition. The first experiment addressed the question of how visual features are integrated in IT. In this study, we sought to determine whether feature selectivity for shape and color is integrated by IT neurons via a conjunction-coding mechanism, or via linear summation. We demonstrate that visual responses of most IT neurons encode shape and color information in a linear manner. Our results shed light on the computational strategy that the brain employs to construct a versatile representation of the visual world.The purpose of the second experiment was to investigate the neural mechanisms underlying repetition priming. Repetition priming is a form of rapid visual learning, whereby previous experience with an object allows for faster, more efficient perceptual processing of that object upon subsequent encounters. This behavioral process is believed to be dependent on activity reductions in single IT neurons, but this hypothesis has never been tested. Indeed, repetition priming has never been demonstrated before in monkeys. To address this issue, we adapted the experimental paradigm of repetition priming for use in primate physiology. We demonstrate that repetition priming at the level of behavior is accompanied by repetition suppression at the level of single neurons in IT. We further demonstrate that repetition suppression in IT results in a proportional scaling reduction of visual responses, and not in a sharpening of the stimulus selectivity. These findings constrain the possible mechanisms whereby visual response plasticity in IT could contribute to behavioral priming.
展开▼
