If human behaviour is sometimes difficult to understand, it is even harder to study. But there is a growing need to do so if we are to discover the biological basis of behavioural disorders and psychiatric illnesses. In response to this need, and thanks to an ever-expanding array of technologies, the field of behavioural genetics is growing and is generating increasing numbers of behavioural mutants in various experimental organisms. Nevertheless, and as discussed by Marla Sokolowski in our November 2001 issue, in a review of fly behavioural genetics research, this field faces some fundamental questions. Not least among these is what to classify as a behaviour (are sleep patterns a behaviour, for example?) and whether genes exist that function specifically to determine behaviour.

That article is now complemented by a review in this issue by Maja Bućan and Ted Abel on the use of the mouse as a model organism for behavioural genetics research. It looks at this research field from a geneticist's perspective and assesses the ways in which mouse genetics can be used to address some of the factors that prevent investigators from identifying causal relationships between specific behaviours and the genes and cellular mechanisms that underlie them. These factors include pleiotropy, the difficulties of phenotyping behaviour, and the effects of genetic background and the environment on behavioural phenotypes. Some of the recommendations made by the authors — such as crossing mutations onto different genetic backgrounds and using only wild-type littermates, not heterozygotes, as controls — might sound to many geneticists like basic, good lab practice, but in a field of mixed disciplines, such as behavioural research, these are valuable points to raise.