Commentary on the utility of experimental social and learning models of alcoholism

This chapter reviews the status of models in scientific research generally and in alcohol research in particular. The reader's attention is drawn to both advantages and disadvantages of models. The authors concur with others that models should be judged primarily by a criterion of usefulness rather than that of truthfulness . The commentary then examines what animal models of alcoholism have purported to model and the variety of criteria offered to evaluate such models. The authors agree in part with Cicero that there is not now and perhaps never will be a "true animal model of alcoholism." Indeed, when the variety of expression of alcoholism in humans is considered, the utility of any single, comprehensive model of alcoholism (animal or human) is questionable. However, the authors note that models of alcohol-relevant phenomena need not replicate or map all the characteristics thought to be present in the human alcoholic. Finally, the authors review several examples of models of alcohol-related phenomena (presented in Chapters 1-4), noting that although such models may be limited, they appear useful in advancing our understanding of this complex and pressing problem area. The authors conclude that an effective strategy for alcohol investigators would be to develop parallel models for both animals and humans and to engage in collaborative research based on such models.