The neuronal origins of human language remains an important unsolved problem in cognitive neuroscience. In particular, neuroscientists lack a detailed physiological model of how widely-separated cortical areas coordinate their activity over multiple time-scales in order to enable language use. Evidence suggests, however, that neuronal oscillations play a key role in the large-scale integration and coordination of distinct brain regions. Oscillations of different frequencies are shaped by different populations of GABAergic interneurons that regulate the precise spike timing of pyramidal cells. Furthermore, neuronal oscillations in all mammals are associated with distinct behavioral states, sensory and motor processing, and cognitive operations such as the deployment of attention, working memory, navigation, and (in humans) language use. However, the exact mechanisms by which oscillatory activity relates to cognitive function remains unknown. Our experiments involve recording the electrical activity from the surface of the brain in patients undergoing neurosurgical treatment for epilepsy, and focus on 1) clarifying the role of the recently-described high gamma (HG, 80-160 Hz) band in cortical areas associated with language processing, and 2) describing the relation of HG to other brain rhythms, particularly in coordinating distinct brain regions. HG appears to be the most robust marker of fast local cortical activation in early sensory and motor areas yet found. The response of HG to task demands and stimulus events in higher-order sensorimotor or association areas needs to be clarified further.
The specific aims of our proposal address this important knowledge gap. Our language task uses real verbs and acoustically-matched nonwords to isolate cortical areas associated with semantic processing, as opposed to purely auditory processing. This experiment will address competing hypotheses in cognitive neuroscience as well as provide data critical for improving the risky, decades-old process of functional electrical stimulation mapping of cortex prior to neurosurgery. More broadly, determining the spatio-temporal dynamics of HG activity in relation to experimental task demands, and how HG interacts with better understood functional frequency bands such as the beta (12-30 Hz) and theta rhythms (4-8 Hz), may lead to many public health advances such as faster, safer brain mapping prior to neurosurgery, identification of cortical areas rendered inoperative due to epileptic seizure foci or tumor growth, and recovery of function via brain-machine interface prosthetics.
