While much is understood about neural mechanisms that analyze the physical attributes of the visual environment, little is known about the culmination of these analyses: the assignment of categories that give stimuli meaning. Our group has developed a novel paradigm for studying visual categories. A morphing system is used to blend prototypes of """"""""cats"""""""" and """"""""dogs"""""""" into single images, with each image having a certain proportion of cat vs. dog. This allows us to continuously vary shape and precisely define a category boundary. Monkeys are trained to judge whether two successively presented images are from the same category. We recently found a neural correlate of category information in the prefrontal cortex (PFC): single neurons that show a sharp change in neural activity at the boundary between categories but relatively little differences in activity within a category. We now plan to use this paradigm to address fundamental questions about category representations. Using multiple electrode techniques, we will simultaneously record neural activity from the PFC and the inferior temporal cortex (TIC), a visual cortical area thought to be important for visual categories. This will allow us to directly compare and contrast their neural properties and relative timing and thus afford a precise assessment of the respective contributions to category-based behaviors. We will also test for category-coding in the hippocampus, a region that shares a close anatomical and functional relationship with the TIC and PFC. To explore whether categorization and identification have common neural substrates, we will record PFC and ITC activity while monkeys switch back and forth between categorizing stimuli and identifying individual category members. To determine if category-coding neurons are highly specialized (like """"""""face cells""""""""), we will test them with a wide range of real world stimuli. To determine whether multiple category memberships are represented by separate neural ensembles or instead multiplexed onto single neurons, we will record while monkeys switch between categorizing a set of stimuli under two different category schemes. Because categorization is central to cognition, data from this project has the potential to impact on a wide range of behaviors and human disorders. The ability to quickly glean concepts and meaning from experience is disrupted in a variety of neuro psychiatric disorders such as autism and schizophrenia. By identifying brain structures important for these abilities, discerning their relative roles, and uncovering their neural mechanisms, we can open a path to drug and behavioral therapies designed to alleviate their dysfunction.
Showing the most recent 10 out of 17 publications
