Neurologic and neuropsychologic impairment are important sequelae of cardiac surgery in general and of coronary artery bypass graft surgery in particular. Although estimates of incidence vary, the numbers affected are considerable. Despite the ubiquity of such effects and the general consensus that impairments originate from ischemic injury secondary to microemboli produced during surgery, the nature of the underlying brain injuries remains poorly understood. Precise, and preferably quantitative, definition of the localization and nature of the underlying injuries is a precondition for the rigorous evaluation of the efficacy of prophylactic measures. The ability of magnetic resonance imaging (MRI) to detect surgically related lesions and the course of brain swelling is described, as are potential improvements in imaging sensitivity. Results of an experimental program studying chemical sequelae of surgery in a pig model are presented. MR spectroscopy can provide noninvasive information on the biochemical changes in brain and brain metabolism that permit empirical evaluation of various neuroprotective interventions. Functional MRI provides a means of studying the neuropsychologic mechanisms most often affected by cardiac surgery. Experimental data are presented that demonstrate that two such mechanisms, selective attention and working memory, can be imaged successfully. Perfusion mapping, combined with functional imaging, allows for the quantitative study of flow and functional activation. Applied to structures such as the cingulate, these techniques permit comparison of surgical sequelae with processes such as normal aging. MRI technology offers the possibility of improved anatomic, chemical, and functional definition of the effects of cardiac surgery on the brain.