The phenotypic analysis of transgenic or knockout mice may be complicated by the effects of unpredictable variables such as the test environment or the experience of the experiment (Crabbe, et al. Science, 284:1670-2, 1999); the background strain of the animal (Crawley et al, Psychopharmacology, 132:107-124, 1997); the ubiquity of the targeted gene and the diversity of its role (Lijam et al, Cell, 90:895-905); and species-specific behaviours exhibited under stress (Gerlai and Clayton, TiNS, 22:47-51, 1999). The aim of this section of the course will be to review methods and tests for assessing the behaviours of transgenic or knockout mice which may confound the interpretation of data obtained from tests of learning and memory. It may also serve to illustrate the range of tests that are available for assessing mice that have been generated to better understand the basis of affective disorders such as anxiety and depression. Amongst other topics, the relative value and reliability of conditioned (e.g. fear potentiated startle) and unconditioned (e.g. elevated plus-maze) tests for determining levels of anxiety will be reviewed (Dawson and Tricklebank, TiPS, 16:33-36, 1995). The latter test will be used to demonstrate the role automated experimental equipment can play in tests which usually require a human observer.