Psychiatric disorders are thought to arise as a result of interactions between genetic vulnerability and environmental risk factors. However, research methods to actually investigate the pattern of hypothesized interactions have only recently been developed. In this article, we review the evidence that genes increase the risk for psychosis by making individuals more sensitive to environmental risk factors (genotype-environment interaction), or by making individuals more likely to select high-risk environments (genotype-environment correlation). It is likely that at least some of the impact of genes on the occurrence of psychosis is mediated through (sensitivity for) environmental risk factors such as a dysfunctional early family rearing environment, paternal absence, use of cannabis, complications of birth and pregnancy, stressful life events and unknown environmental risk factors associated with urban life and membership of certain ethnic groups. With the advent of molecular genetics, further knowledge about possible genotype-environment interactions is urgently required in order to develop and improve strategies for the prevention and early treatment of psychosis.