RESUMO
The free exploratory paradigm has been proposed as a model of trait anxiety. By definition, trait anxiety does not vary from moment to moment, therefore any model used to evaluate it needs to give the same results in sequential trials. With this in mind, the test-retest reliability of the free-exploratory paradigm in rats was evaluated. The behaviour of drug-naive, adult, male, Wistar rats was measured in the free-exploratory apparatus on two occasions, either one week apart (Experiment I) or four weeks apart (Experiment II). The following parameters were evaluated: percentage of time spent in the novel compartment of the free exploratory box; percentage of novel compartment units visited; percentage of time rearing in the novel compartment; and attempts to enter the novel compartment. Subsequently, the intraclass correlation coefficient (ICC) and the kappa index (kappa) were calculated for each of these parameters. The "percentage of time spent in the novel compartment" was found to be stable over time. Analysis of this parameter in Experiment I showed high values of ICC (0.805) and kappa (0.716), indicating good reliability with a one-week inter-trial interval. In Experiment II, although the same parameter's kappa index (0.255) failed to show a reasonable test/retest concordance, the ICC value (0.760) still demonstrated a good test/retest correlation, suggesting that with a four-week inter-trial interval, the reliability may diminish, but still exists. Therefore, our results bring further support to the free-exploratory paradigm as an animal model of trait anxiety.