Discrimination of a drug mixture in rats: role of training dose, and specificity.

Many drugs produce compound discriminative stimuli with at least two elements; in contrast, the present study examines discrimination of a mixture of two drugs and tests the role of training dose in, and the specificity of, such a discrimination. Rats discriminated a mixture of nicotine (0.2mg/kg s.c.) and midazolam (0.1mg/kg s.c.) from saline in a two-bar operant conditioning procedure with accuracy of at least 80%. Stimulus control was analyzed by testing each drug separately. Initially, stimulus control was mainly attributable to the midazolam. The doses of drugs used to maintain the discrimination were then altered. As the training dose of nicotine increased and that of midazolam decreased, the magnitudes of responses to the separate drugs were progressively reversed, until stimulus control was mainly attributable to nicotine. Thus, responses to the components of the compound stimulus were systematically related to the amounts of drugs in the mixtures used to maintain the discrimination, and there was some evidence that a strong stimulus produced by one drug may have overshadowed a weaker stimulus produced by a different agent. To test specificity, generalization to other drugs was examined. There was no generalization to amphetamine, morphine or quipazine, up to doses that reduced overall rates of responding. It follows that cues produced by mixtures of drugs may be as specific as those produced by single agents.