Rapid advances in biotechnology have created new demands for tests of mouse behaviour having both high reliability and high throughput for mass screening. This paper discusses several statistical and psychological factors pertinent to replication of results in different laboratories, and it considers the question of which inbred strains are best for test standardization. In this context, the problem of absent corpus callosum in the 129 strains is addressed with data from a recent study of six diverse tests of behaviour, and it is shown that effects of absent corpus callosum are usually nonsignificant and/or very small. Whether any 129 substrain is to be included in the list of standard strains depends on the goal of the standardization--collecting diverse phenotypic data on most available strains by a few expert investigators (the gold standard) or refining behavioural tests in order to establish a normal range of behaviour that can be used to judge a wider range of strains or even an individual mouse.

The distinction between alcohol use (normative) and abuse (unfortunately common) implies dysregulation of motivation directed toward the drug. Genetic contributions to abuse risk are mediated through personality differences, other predispositions to drink excessively, and differences in sensitivity to the acute and chronic consequences of the drug. How to assess motivation in laboratory animals is not straightforward but risk factors for and consequences of alcohol abuse can be modeled with reasonable fidelity in laboratory rodents. Remarkably few rodent studies focus on the genetic contributions to alcohol's reinforcing value: almost all examine preferential drinking of unflavored alcohol over water. Such studies will likely never avoid the confounding role of taste preferences and most often yield intake levels insufficient to yield a pharmacologically significant blood alcohol level. Genotypes that avoid alcohol probably do so based on pre-ingestive sensory cues; however, post-ingestive consequences are also important. Thus, the quest for improved measures of reinforcing value continues. We have genetic differences aplenty, but still lack evidence that any genotype will readily self-administer alcohol to the devastating extent that many alcoholics will. Encouraging results that are emerging include improved behavioral methods for elevating alcohol intake and inferring alcohol reinforcement, as well as new genetic animal models. Several ingenious assays to index alcohol's motivational effects have been used extensively. Alcoholic drinking that attempts to prevent or to alleviate withdrawal symptoms has been modeled. Another characteristic of alcoholic drinking is its persistence despite abundant evidence to the drinker of the damaging effects of the excessive drinking on work, relationships, and/or health. Modeling such persistence in rodents has been uncommon to date. New genetic animal models include lines of mice selectively bred for chronic high drinking, and other bred for high binge-like drinking. We have a much more clear idea now about some important experiments remaining to be performed.

Temperature regulation during withdrawal from ethanol dependence was studied in mice. Dependence was induced utilizing ethanol vapor inhalation combined with injections of pyrazole, an alcohol dehydrogenase inhibitor. One control group received vehicle (0.9% NaCl); another received pyrazole (68.1 mg/kg) in vehicle. During withdrawal, mice were placed in a thermal gradient, with core temperature (Tc) and preferred temperature (Tpref) continuously recorded for 26 h. During the period of maximal withdrawal severity (4-10 h after removal from ethanol vapor), the withdrawal group was more active than controls. Withdrawal group Tc [36.4 +/- 0.1 (SE) degrees C] was similar to that of NaCl (36.2 +/- 0.1 degrees C) and pyrazole (36.3 +/- 0.2 degrees C) controls. Withdrawal group Tpref (30.5 +/- 0.5 degrees C) was significantly lower than either NaCl (33.5 +/- 0.3 degrees C) or pyrazole (32.9 +/- 0.5 degrees C) controls. Analysis of covariance with activity as covariate indicated that the difference between Tc and Tpref was greater for the withdrawal group and was due to more than increased activity. Mice withdrawing at constant temperature (29.5 degrees C) did not show Tc different from that of controls. These results support the conclusion that regulated body temperature is not altered during withdrawal. We propose that the lower Tpref of withdrawing mice represents a means of dissipating excess heat that is partly generated by the hypermetabolic state accompanying withdrawal from ethanol dependence.