Two scientific developments that occurred in the last third of the twentieth century contributed to long-simmering uncertainty about the nature of typical concussive brain injury (CBI). On the one hand, neuropsychologists labored to define a profile of results, from desk-top paper-and-pencil testing, that was reliably associated with brain damage. Multiple reports observed that psychological tests usually normalized by three months post-injury. On the other hand, by 1967, neuroscience had discovered evidence that CBI seems to cause lasting damage to both neurons and axons, including cell death. More recently, abundant neuroimaging evidence has also demonstrated long-term post-concussive brain change. These are not incompatible facts. It has simply required another half-century for scholars to explain: neuropsychological testing is sensitive to some early effects but apparently insensitive to most lasting effects of CBI. It has become clear that meaningful progress in this field requires authentic biomarkers with demonstrable predictive validity rather than inferences from behavior. Functional neuroimaging holds promise.