The powerful genome-wide association studies (GWAS) revealed common mutations that increase susceptibility for schizophrenia (SZ) and bipolar disorder (BD), but the vast majority were not known to be functional or associated with these illnesses. To help fill this gap, their impact on human brain structure and function has been examined. We systematically discuss this output to facilitate its timely integration in the psychosis research field; and encourage reflection for future research. Irrespective of imaging modality, studies addressing the effect of SZ/BD GWAS risk genes (ANK3, CACNA1C, MHC, TCF4, NRGN, DGKH, PBRM1, NCANandZNF804A) were included. Most GWAS risk variations were reported to affect neuroimaging phenotypes implicated in SZ/BD: white-matter integrity (ANK3andZNF804A), volume (CACNA1CandZNF804A) and density (ZNF804A); grey-matter (CACNA1C, NRGN, TCF4andZNF804A) and ventricular (TCF4) volume; cortical folding (NCAN) and thickness (ZNF804A); regional activation during executive tasks (ANK3, CACNA1C, DGKH, NRGNandZNF804A) and functional connectivity during executive tasks (CACNA1CandZNF804A), facial affect recognition (CACNA1CandZNF804A) and theory-of-mind (ZNF804A); but inconsistencies and non-replications also exist. Further efforts such as standardizing reporting and exploring complementary designs, are warranted to test the reproducibility of these early findings.