Research on the perpetration of intimate partner violence (IPV) reveals a wide array of important influences, ranging from neurophysiology to culture. Yet prominent theories are narrow in focus and existing interventions are only modestly effective. This commentary highlights the need for a biopsychosocial model of IPV that is coherent, integrative, nonreductionistic, and facilitates effective practice, and illustrates how the social information processing (SIP) approach can provide an organizing framework for this effort. SIP involves recursive steps leading from the decoding of social cues to the generation, selection, enactment, and evaluation of responses in light of interpersonal goals and with constant reference to a central “database” of social knowledge. Factors such as posttraumatic stress, alcohol intoxication, and neurocognitive deficits increase risk for IPV through specific alterations in SIP processes such as the attribution of negative partner intent and perceived acceptability of violence. From this perspective, violence is an actively selected response option and therefore individuals are personally responsible for choosing to act in an abusive manner. However, a wide array of biological, psychological, and social factors influence SIP processes. Further understanding of SIP in IPV will clarify change targets for intervention, facilitate the development of new intervention strategies, and provide efficient ways to evaluate intervention effects.