A series of experiments was carried out in which the potency of the selective α-amino-3-hydroxy-5-methyl-4-isoxazole proprionate (AMPA)-receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline (NBQX) (10-100 mg/kg) on locomotor activity was investigated, in mice. NBQX reduced all forms of activity studied, but its potency to do so varied according to the conditions of the experiment. The smallest dose of NBQX significantly reducing spontaneous or cocaineinduced activity was 100 mg/kg. Mice that had been repeatedly treated with 16 mg/kg cocaine once per week, for 7 weeks, showed a sensitized locomotor response to a challenge dose of cocaine (16 mg/kg). NBQX reversed the sensitized response at 30 and 100 mg/kg. The pattern of results obtained leaves open the role that AMPA-receptors may have in the expression of behavioural sensitization. In two further experiments, mice were trained to self-administer cocaine (30 µg per reinforcer) via intravenous catheters, using an operant lever pressing technique. When the amount of cocaine per reinforcer was doubled (to 60 µg) or halved (to 15 µg) the mice adapted lever pressing rates to maintain some constancy of self-dosing (but not at 7.5 µg per reinforcer) and when saline was substituted for cocaine, response rates increased considerably (extinction bursting). NBQX (10 and 30 mg/kg) reduced the self-administration of 30 µg reinforcers of cocaine, but only during the first 30 min of the test session. There was no evidence that NBQX specifically antagonized the reinforcing effect of cocaine, as responding was similarly reduced on both the reinforced and the non-reinforced lever, nor did the response to NBQX mimic behaviour seen following changes in the concentration of the reinforcer. The results of the locomotor experiments and the self-administration experiments are discussed together, in terms of current hypotheses about glutamatergic mechanisms involved in motivation for drug.