Effects of drug discrimination history on the generalization of pentobarbital to other drugs.

In pigeons trained to discriminate between pentobarbital and saline, pentobarbital, amobarbital and diazepam substituted for pentobarbital, whereas phencyclidine (PCP) substituted in part for pentobarbital and d-amphetamine, morphine and drug vehicles did not substitute. After morphine replaced pentobarbital as the training drug (group A), morphine, pentobarbital and diazepam substituted, PCP substituted in part, but not d-amphetamine, haloperidol and vehicles. After d-amphetamine replaced pentobarbital as the training drug (group B), d-amphetamine, pentobarbital and diazepam substituted, PCP substituted in part, but not haloperidol, morphine and vehicles. Next, morphine and d-amphetamine were reversed as training drugs for the two groups. In group A, morphine, d-amphetamine, pentobarbital and diazepam substituted, PCP substituted in part, but not haloperidol and vehicles. Similar effects were observed in group B. Next, birds in group A were reinforced for responses on the drug key (red key) after d-amphetamine and on the previous saline key (green key) after pentobarbital. In group B, morphine continued as the training drug for the red key, whereas responses on the green key were reinforced after pentobarbital. In group A, d-amphetamine, morphine, d-pentazocine and to some extent PCP, produced responding on the red key, whereas pentobarbital, diazepam, haloperidol and the vehicles produced responding on the green key. Similar results were obtained in group B. Finally, responses were reinforced on the green key after pentobarbital and on the red key after saline. Group B did not learn this discrimination. In group A, responding occurred on the red key after d-amphetamine, morphine, haloperidol and vehicles, in part after d-pentazocine, but not after pentobarbital, diazepam and PCP. These experiments show that drug stimuli can continue to exert stimulus control over behavior for long periods, even when training with several other drug stimuli intervenes between tests, and the experiments also show that through sequential training procedures multiple drugs can serve as discriminative stimuli for the same response, even when these drugs are from different phamacological classes.