ABSTRACT Interpreting human facial expressions is a core human social capacity negatively impacted in diverse neuropsychiatric conditions. Failures of human social cognition contribute to disease burden in mood disorders, as well as autism spectrum and anxiety disorders, reinforcing and exacerbating the social isolation and hopelessness of patients across diagnostic categories. The proposed study examines the RDoC subconstruct of “Facial Communication” and the domain of “Negative Valence Systems” using behavioral cognitive tasks that place heavy demands on the capacity to interpret emotional facial expressions. Prior efforts to develop neuromodulation paradigms for treatment-resistant cases of neuropsychiatric diseases characterized by socio-affective dysfunction have fallen short, in part due to an absence of defined neural dysfunctional-state signals. The overall goal of this R01 proposal is to map the spatiotemporal dynamics of social affective processing and to examine selective modulation of these dynamics in humans undergoing invasive intracranial monitoring for treatment-resistant epilepsy and depression. Pursuing this signal from a novel platform with invasive intracranial recording electrodes provides much-needed spatial and temporal resolution to characterize the neural dynamics of socio-affective processing. We will leverage first-in-human intracranial neural recording opportunities created by a novel therapeutic platform termed “stereotactic electroencephalography-informed deep brain stimulation” (stereo-EEG-informed DBS; Parent Study UH3-NS103549), as well as the powerful platform of intracranial stereotactic recording and stimulation in patients undergoing epilepsy surgical evaluation at Baylor College of Medicine. The sEEG-informed DBS trial provides unique opportunities for intracranial recording of affect-relevant network regions in patients with treatment-resistant depression (TRD). Recordings in identical regions in epilepsy patients who themselves often demonstrate mild-moderate depressive symptoms will provide a wide dynamic range across the symptom spectrum. To provide critical data on the spatiotemporal dynamics of socio-affective processing we will leverage these two human intracranial recording and stimulation cohorts to study the precise structural (Aim 1), functional (Aim 2), and causal (Aim 3) properties of the affective salience network. Greater understanding of the social processing circuitry mediated by the affective salience network may be used to drive therapeutic innovation, pioneering a new paradigm that improves socio-emotional function across a wide variety of neuropsychiatric conditions. The results from this proposal have the potential to improve the lives of patients with dysfunction in social affective processing, with implications for a wide range of neuropsychiatric diseases.