Utility of ethological analysis to overcome locomotor confounds in elevated maze models of anxiety.

The elevated plus-maze is a commonly used model to identify putative anxiolytic and anxiogenic drugs. However, the validity of elevated plus-maze and other recently developed variants such as the elevated zero-maze has recently been questioned on the grounds that both the reference anxiolytic drug chlordiazepoxide and the psychostimulant d-amphetamine increase open arm exploration and stimulate locomotor activity. These findings suggest that measures of "anxiety" in the elevated maze cannot be adequately dissociated from simple changes in locomotor activity, which may confound the interpretation of results obtained using these models. A variety of approaches to assess drug effects on locomotor activity in the elevated maze have been suggested, including the use of total and closed arm entries, as well as supplementary tests such as exploration of the holeboard apparatus. However, all these approaches utilise the measurement of exploration in a novel environment, and as such, could potentially be influenced by either changes in anxiety or locomotor activity. Recently, it has been shown that ethological measures of "risk assessment", such as stretched-attend postures and head-dipping, are sensitive indicators of drug-effects in the elevated maze. The present study assessed the utility of ethological analysis in dissociating locomotor activity from "anxiety" by comparing the effects of d-amphetamine to those of chlordiazepoxide in the rat elevated zero-maze. The results showed that both chlordiazepoxide and d-amphetamine increase the amount of time spent in the open arms and reduce "risk assessment" without increasing line crossing or rearing. These results confirm that under certain test conditions, psychostimulants are capable of producing "false-positives" in elevated maze models, and that both traditional methods and the ethological measures used in this study fail to unequivocally dissociate drug effects on anxiety from effects on locomotor activity. Further studies using other species and different classes of psychostimulants are warranted to determine the generality of these findings.