The Y-maze was used to assess spontaneous alternation behaviour in mice to examine possible interactions between the N-methyl-D-aspartate receptor channel blocker dizocilpine and purine receptor agonists and antagonists. Scopolamine reduced spontaneous alternation. Dizocilpine also produced a dose-dependent reduction in alternation scores, which was accompanied by an increase in locomotion. The selective A(1) adenosine receptor antagonist 8-cyclopentyl-1,3-dipropylxanthine (CPX) had no effect when administered alone, or in combination with scopolamine. However, when co-administered with dizocilpine, CPX reversed both the deficit in alternation behaviour and also the increase in locomotion induced by dizocilpine. The A(1) selective agonist N-6-cyclopentyladenosine (CPA) had no effect on either locomotion or alternation scores when administered alone, but in combination with scopolamine, CPA attenuated the scopolamine-induced deficit. CPA had no significant effect on the dizocilpine-induced deficit. The A(2) selective agonist N-6-[2-(3, 5-dimethoxyphenyl)-2(2-methylphenyl)-ethyl]adenosine (DPMA), had no effect on spontaneous alternation when administered alone, but did cause a depression of locomotion. DPMA had no significant effect when co-administered with scopolamine, but reversed the deficit in spontaneous alternation, and the increase in locomotion induced by dizocilpine. The A(2) selective antagonist 3,7-dimethyl-1-propargylxanthine (DMPX) had no effect when given alone or in combination with scopolamine, but when co-administered with dizocilpine, DMPX reversed the reduction in spontaneous alternation caused by dizocilpine. It is concluded that dizocilpine has a detrimental effect on spontaneous alternation which is mediated partly by A(1) and A(2) adenosine receptors.