normalcy, therefore, may or may not serve as clinical indicators of a disease state. Various transgenic and knockout mice that exhibit severe behavioral and physiological phenotypes appear abnormal relative to their control littermates, but are normal for their genotype. For example, it is appropriate to evaluate Huntington’s disease transgenic mice for signs of stress and distress only relative to their own “normal” behavior, taking into account their particular genetic makeup, their abnormal motor patterns, and reduced weight gain (Mangiarini et al. 1996).

Many parameters have an effect on species-specific normal behavior and should be taken into consideration when behavioral characteristics are used to determine normalcy or the presence of stress and distress. Animals exhibit a variety of behavioral changes as part of the normal aging process. Males and females differ in the baseline values of many stress markers. Inbred murine strains differ in almost every behavioral, sensory, motor, and physiological trait studied and each inbred strain may respond to stress differently. Similar behavioral differences in response to stress have been observed in primates. Genetically engineered phenotypes need to be considered when assessing stress and distress in transgenic and knockout animals. The maternal environment and rearing experiences of the offspring affect their future responses to stress and distress. Special physiological states, such as impending parturition, are defined by state-specific behaviors. Housing conditions may also modify species-specific behavioral patterns. Behavioral normalcy is further characterized by the absence of bizarre or atypical patterns of species-specific behavior. The presence of stereotypies usually implies suboptimal environments and possibly poor animal welfare.

The identification of species-typical behavior often comes from ethograms developed by researchers to describe the kinds of behavior that animals display in various settings (Bronson 1979; for more references see Additional References). While the use of species-typical behavior as a normative benchmark has considerable value (Latham and Mason 2004), it does have limitations. First, the full range of species-specific behaviors cannot be recreated (or allowed to be expressed) in the laboratory animal care facilities as some types of behavior observed in natural settings (e.g., severe aggression) are clearly undesirable from a laboratory management perspective. Second, species-typical behaviors are neither invariant nor universal, as both the frequency and the presence of such behaviors vary