Model Types And Validity

Two principal types of animal models are prevalent in psychopharmacology. Assay models are used to screen drugs with unknown therapeutic potential and need not resemble anything seen in a psychiatric disorder. The validity of assay models is determined by their ability to predict that a drug reliably belongs to a therapeutic class. In rats, for example, passive avoidance deficits induced by olfactory bulbectomies are reversed by antidepressants but not by psychostimulants, neuroleptics, or anticholinergics (Song and Leonard 2005). New drugs with unknown therapeutic potential that reverse bulbectomy-induced deficits in rats are therefore considered to be possible antidepressants. Assay models also satisfy additional criteria, including ease of use, high throughput capacity, reproducibility, and economic concerns.

The second principal type of animal model simulates an aspect of interest in a psychiatric disorder. Simulation models are used to investigate the biology of psychiatric disorders or the mechanisms of action of psychotherapeutic drugs. In addition to the criterion of predictive validity described above for assay models, simulation models are often evaluated for two other aspects of validity.

Face validity refers to phenomenological similarities between the animal model and the human psychiatric condition. As originally proposed by McKinney and Bunney (1969), animal models of human mental illness have a high degree of face validity when the following criteria are met: the model is produced by etiological factors known to produce the human disorder, the model resembles the behavioral manifestations and symptoms of the human disorder, the model has an underlying physiology similar to the human disorder, and the model responds to therapeutic treatments known to be effective in human psychiatric patients. How these criteria are evaluated and established has been described in detail elsewhere (McKinney 2001; Weiss and Kilts 1998).

Construct validity refers to the theoretical rationale for linking a psychiatric disorder to an endpoint measured in the animal model. To establish construct validity, a theory for understanding a disorder is mapped or shown to be equivalent to an endpoint in the animal model (Sarter and Bruno 2002). Disease heterogeneity and related concerns that no single animal model can capture the complexities of an entire disorder have shifted attention away from modeling disorders as a whole (McKinney 2001; Insel 2007) to focus on psychiatric endophenotypes (Gould and Gottesman 2006). The endophenotype strategy presupposes that each disorder comprises behavioral, physiological, neuroanatomical, cellular, and molecular components that are more proximal to causal risk factors than are the actual disorders defined in DSM-IV-TR (American Psychiatric Association 2000; Arguello and Gogos 2006).

Psychiatric endophenotypes are conceptualized as mediating the link between genetic or environmental risk factors and the resulting disorder (Figure 6-1). Precise delineation of endophenotypes also serves to highlight the fact that certain features of psychiatric disorders—for example, diminished verbal recall, self-conscious emotions, delusions of control, impaired theory of mind, and suicidal ideation—are likely unique to humans. Many other endophenotypes are, however, amenable to modeling in animal research, as will be described in the following sections of this chapter.

FIGURE 6-1. Psychiatric endophenotypes mediate the link between causal risk factors and the resulting psychiatric disorders defined in DSM-IV-TR.

FIGURE 6-1. Psychiatric endophenotypes mediate the link between causal risk factors and the resulting psychiatric disorders defined in DSM-IV-TR.

Representative examples and nonexhaustive lists of risk factors, endophenotypes, and disorders are provided to illustrate this theoretical framework for psychiatry neuroscience and psychopharmacological research.

Features that confer a high degree of validity for simulation models are often poorly suited for animal assay models used in drug screening research. An example is the typical delay in response onset for conventional antidepressants. On the other hand, simulation models that achieve all three aspects of validity may be better suited to identify substantively new drugs that differ from those used to establish an assay model. Excessive reliance on assay models may increase the tendency to perpetuate the same side effects as those produced by known medications. The following sections selectively illustrate how animal models have advanced our understanding of the psychopharmacology and biology of depressive disorders.

Beat Depression Now

Beat Depression Now

Let me be up front. My intention is to sell you something. Normally, it's not wise to come out and say that. However, I can do so because I have such an incredible deal for you that you'd be crazy to pass on it.

Get My Free Ebook


Post a comment