As the use of transgenic and null mutation techniques in the development of animal models of disorders increases, the importance of selecting the appropriate genetic background also increases. The genetic background of the mouse strains used as models for various disorders is critical because of the potential for epistatic effects on the expression of transgenes and null mutations. Twelve strains of inbred mice and seven F1 hybrids were tested in multiple behavioural tasks including open-field locomotor activity, Y-maze activity, auditory and tactile startle and prepulse inhibition of startle response. Differences across genotypes were found for all variables measured. The range of variability among genotypes was dependent on the specific measure so careful consideration must be made in selecting a strain for testing a particular behaviour. Because of the polygenic nature of each of the behavioural phenotypes, the impact of a single gene manipulation may vary depending on the genetic background on which it is expressed. Moreover, quantitative trait loci methods could be applied to these behaviours.