The limits of the current diagnostic systems have led to a variety of efforts to identify alternate expressions of mental disorders that are broader than the DSM or ICD diagnostic criteria needed to diagnose them (e.g., Research Diagnostic Criteria, or RDoC). These newer approaches reflect a growing consensus that characterizing multiple, dimensional phenotypes can advance the search for identifying etiological or modulatory factors of psychiatric disorders. Moreover, the failure of Mendelian genetics to make progress in psychiatric disorders has led some to propose an alternative genetic focus on the traits associated with illnesses rather than on the diagnosis per se. These alternate phenotypes or endophenotypes (e.g., neuroanatomical, psychophysiological or neuropsychological abnormalities) of disorders may be more specific and amenable to objective measurement than clinical symptoms, and presumably reflect variation among smaller numbers of genes than more complex, clinical symptoms that are hypothesized to be more distal from their genetic origins. This has led to a dramatic growth of using endophenotypes in genetic studies, and consideration of their potential usefulness in the development of targets for early interventions. In this chapter, we review neuropsychological and structural neuroimaging endophenotypes for schizophrenia and for schizophrenia spectrum disorders, with an emphasis on key conceptual criteria for assessing endophenotypes, including their relationships to schizophrenia, to non‐psychotic relatives, and to heritability. Future directions for establishing the validity of endophenotype research are also discussed.