Genes involved in psychiatric disorders are difficult to identify, and those that have been proposed so far remain ambiguous. As it is unrealistic to expect the development of, say, a ‘schizophrenic' or ‘autistic' mouse, mice are unlikely to have the same role in gene identification in psychiatry as circling mice did in the discovery of human deafness genes. However, many psychiatric disorders are associated with intermediate phenotypes that can be modeled and studied in mice, including physiological or anatomical brain changes and behavioral traits. Mouse models help to evaluate the effect of a human candidate gene mutation on an intermediate trait, and to identify new candidate genes. Once a gene or pathway has been identified, mice are also used to study the interplay of different genes in that system.