Behavioral screening for cognition enhancers: from indiscriminate to valid testing: Part I.

Preclinical efforts to detect and characterize potential cognition enhancers appear to have been dominated by a strategy of demonstrating a wide variety of apparently beneficial behavioral effects with little attention given to the specific psychological mechanisms underlying behavioral enhancement. In particular, the question of whether or not behavioral facilitation is based on relevant mnemonic mechanisms and is independent of the stimulus properties and/or the motivational and attentional components of a task is not often considered. As a result, an overwhelming number of compounds have failed to produce the clinical effects predicted for them on the basis of preclinical research. The available data suggest that a more successful approach requires deductive research strategies rather than the indiscriminate accumulation of apparently beneficial effects in a variety of behavioral tasks and animal models. The first step towards such an approach is a systematic and rigorous evaluation of the different aspects of validity for the models most frequently used in preclinical research. It is concluded that a combination of good construct validity and good face validity represents a necessary condition for screening tests with predictive validity, and that the most popular paradigms fail to fulfil these criteria. Future screening programs for cognition enhancers will probably be characterized by a depreciation of "fast and dirty tests" in favor of approaches focussing on the validity of the effects of potential cognition enhancers.