This paper addresses the question of how incidental findings (IFs) in clinical research should be managed by researchers, focusing in detail on IFs discovered in neuroimaging research. It begins by engaging the larger research ethics issue of whether researchers have any obligations of clinical care to participants, and assesses the content and merits of one particular framework for answering this question, Richardson and Belsky's ancillary care model. From here the paper develops an organizational structure for integrating the ancillary care model with existing research ethics standards, with the aim of better understanding their respective domains. It makes a distinction between incidental findings that are anticipated by informed consent documents, and those that are unanticipated, arguing that this distinction is critical for evaluating researcher obligations. Finally, it takes on the issue of incidental findings in neuroimaging research, translating the standards discussed into recommendations for both unanticipated and anticipated findings.