Rigorous experimental design can minimize the high risk of false positives and false negatives in the behavioral phenotyping of a new transgenic or knockout mouse. Use of well established, quantitative, reproducible behavioral tasks, appropriate Ns, correct statistical methods, consideration of background genes contributed by the parental strains, and attention to litter and gender issues, will maximize meaningful comparisons of -/-, +/-, and +/+ genotypes. Strategies developed and used by our laboratory are described in this review. Preliminary observations evaluate general health and neurological reflexes. Sensory abilities and motor functions are extensively quantitated. Specific tests include observations of home cage behaviors, body weight, body temperature, appearance of the fur and whiskers, righting reflex, acoustic startle, eye blink, pupil constriction, vibrissae reflex, pinna reflex, Digiscan open field locomotion, rotarod motor coordination, hanging wire, footprint pathway, visual cliff, auditory threshold, pain threshold, and olfactory acuity. Hypothesis testing then focuses on at least three well-validated tasks within each relevant behavioral domain. Specific tests for mice are described herein for the domains of learning and memory, feeding, nociception, and behaviors relevant to discrete symptoms of human anxiety, depression, schizophrenia, and drug addiction. An example of our approach is illustrated in the behavioral phenotyping of C/EBPdelta knockout mice, which appear to be normal on general health, neurological reflexes, sensory and motor tasks, and the Morris water task, but show remarkably enhanced performance on contextual fear conditioning.
Copyright 1999 Elsevier Science B.V.