Project Summary The project goal is to continue exploring the relationship between experience, expertise, and modular organization in the temporal lobe. There are distinct regions of the brain, reproducible from one person to the next, specialized for processing the most universal forms of human expertise; such as face recognition, speech processing, and reading. Face-selective regions in the temporal lobe have been identified using functional magnetic resonance imaging (fMRI) in both humans and macaques, and in humans visual recognition of letters and words is also localized, to the same part of the temporal lobe, but more lateral and posterior and in the opposite hemisphere. Because of the importance of social interactions in primates, one could imagine a face-specific region being generated by natural selection, but it is unlikely that a cortical region specific for written words could have evolved, given that humans have been using written language for only a few thousand years, and literacy has been widespread for at most a few hundred years. The fact that most people have intensive early experience with both faces and symbols prompted the question of whether intensive early experience could cause monkeys to develop anatomical specializations for stimuli they never naturally use. The project revealed that intensive symbol training can induce the formation of artificial domains in macaque inferotemporal cortex (IT), but only when they were trained as juveniles, not as adults. Furthermore these artificial domains are localized within IT according to a pre-existing proto-architecture. The proposed project is to probe and explore this proto architecture to characterize the constraints on the modifiability of IT. Previous results indicated that this proto-architecture could be a retinotopic-map based shape organization. Therefore the one goal is to find out whether the proto-architecture is more constrained by shape or by retinotopy, by pitting the two features against each other during the training regimen, and using fMRI after training to find out whether novel domains localize according to viewing eccentricity or shape. The major goal is to use chronic multi-electrode arrays to explore the constraints on the modifiability of response properties of individual neurons in IT, and to look for differences in the modifiability of neuronal response properties as a function of the age of the animal and as a function of anteroposterior location along IT. Chronic arrays will be implanted in different regions along the anterior/posterior extent of IT in animals of different ages. Response tuning across the array will be assessed using a large image set. Then the animals will be intensively trained on a subset of those shapes, and, as they are trained, response tuning across the arrays will be monitored. Chronic array electrophysiology and fMRI will be used in parallel and in conjunction to explore the constraints on the formation of training-induced dom...