The behavioral functions of upper and lower cortical layers

PROJECT SUMMARY/ABSTRACT The cerebral cortex mediates all of human and animal cognition, encompassing a diverse set of abilities including sensation, perception, decision making, and motor planning. Dysfunctions of the cerebral cortex are thought to underlie numerous neurological and psychiatric disorders. A major obstacle both to understanding normal behaving and to treating pathology is the high degree of complexity of cortical circuitry, which has remained largely enigmatic. The conventional view of neocortex has been that sensory processing begins in layer 4 (L4), which was identified a century ago as the principal target of thalamic axons carrying information from our sensory organs. Sensory transforms are widely believed to occur as excitation spreads serially along the densest axonal pathways (thalamus?L4?L2/3?L5/6). Recently we discovered that the cerebral cortex, rather than being a monolithic structure, may contain two entirely separate processing systems, activated by the same signals arising from the thalamus. L4 is thus not an obligatory distribution hub for cortical activity, and thalamus activates two distinct ?strata? of cortex in parallel. This proposal's goal is to identify the behavioral and computational roles of the upper (L2-4) and lower strata (L5/6) as well as the interactions between them. We will investigate the behavioral roles of these layers in the mouse whisker system. Specific layers will be optogenetically disrupted in a series of tactile behavioral tasks, in which task complexity is progressively increased. Interlaminar interactions will also be studied by recording electrophysiologically from specific layers during behavior and using novel machine learning techniques designed to identify the type of computation performed in different levels of ?deep networks?. The dimensionality of the representation in a layer will be estimated under normal behaviors and when specific layers are inactivated. Identifying fundamental functions of upper versus cortical layers will likely pave the way for future studies in other neocortical systems and in higher-order species. Moreover, as the different layers contain molecularly and biophysically distinct cell types and project to distinct downstream targets, specific neurological disorders may involve dysfunction of specific pathways, cell types, and layers. Establishing the behavioral and computational roles of these elements may contribute to development of targeted therapies.