Differences in the behavioural profiles of dopamine D2 receptor antagonists (e.g., haloperidol vs. sulpiride) in a animal models of anxiety have prompted speculation concerning the importance of their relative affinities for D2-like receptor populations. In an initial attempt to investigate the involvement of D3 receptors in anxiety, the present study examined the effects of the preferential D3-receptor agonist, (+/-)7-OH-DPAT (0.01-10.0 mg/kg), on behaviours displayed by male mice in the elevated plus-maze paradigm. An ethological approach incorporating measurement of a range of defensive acts and postures in addition to conventional parameters was used to provide a comprehensive behavioural profile for the compound. Data analysis indicated a significant increase in percentage of open-arm entries at 10 mg/kg and an altered temporal distribution of behaviour at 1-10 mg/kg. Furthermore, risk-assessment measures (stretched attend postures, closed-arm returns) were dose dependently reduced by drug treatment. Although these behavioural changes would be consistent with anxiety reduction, such an interpretation is negated by dose-dependent decreases in all active behaviours (arm entries, rearing, and head-dipping) and by marked increases in entry latencies and nonexploratory behaviour at the highest dose tested. Overall, these effects are remarkably similar to those previously reported for quinpirole, suggesting either that D2 and D3 receptors exert similar behavioural control or that the agents employed are sufficiently potent at D2 receptors to prevent a resolution of D2 and D3 responses.