PROJECT SUMMARY Social interaction and communication begin in early infancy, and, although these are fundamental human functions, little is known about the underlying neural mechanisms that regulate them particularly in Autism Spectrum Disorder (ASD). ASD is a neurodevelopmental disorder characterized by significant disabilities in language and social skills, and the specific neural mechanisms that lead to these disabilities remain active topics for investigation. Emerging theoretical directions converge on problems with eye-contact as a salient component of these communication and social disabilities. Technical limitations, however, associated with imaging of two or more individuals during natural communication and mutual eye contact have been a primary obstacle to these investigations. To overcome this technical impasse, we employ a rapidly developing brain imaging technology, functional near-infrared spectroscopy (fNIRS) allowing simultaneous neural imaging of two individuals during valid interactions to observe the neural effects of eye-to-eye contact and actual dialogue. Functional NIRS detects active neural tissue based on the blood-oxygen-level-dependent (BOLD) signal by measuring variations in the absorption spectra associated with oxyhemoglobin and deoxyhemoglobin. Because detectors and emitters are surface mounted on the head, they are relatively insensitive to head movement, and, as such, fNIRS is well suited for investigations of neural events engaged during active interpersonal interactions between two participants. The neural mechanisms that underlie atypical interpersonal interactions and eye contact in adult ASD are the focus of this proposal. Pilot studies confirm the feasibility of all aspects of this research project. Dyads consisting of a confederate and a participants with typical development (TYP) or ASD will be compared during neuroimaging while engaged in natural interaction and communication. We introduce a computational approach based on wavelet analysis to quantify regional cross-brain coherence between the two participants and hypothesize that cross-brain coherence associated with speech and eye contact will be reduced in ASD relative to the TYP cohort. Cross-brain computations also form the basis for a model of dynamic neural processes based on neural “send and receive” functions during communication. We hypothesize that these dynamic “cross-brain communication” systems unify and coordinate the roles of language production and reception (Broca's and Wernicke's Areas), respectively, with visual reception involving face specializations (fusiform gyrus). Computational comparison of cross-brain connectivity effects as well as conventional functional connectivity and segregation/contrast effects during live communication both with and without direct eye contact provides a transformational technical, empirical, computational, and theoretical advance toward understanding the dynamic neural mechanisms associated with social a...