A comparison of the stimulus effects of codeine in rhesus monkeys under the contingencies of a two lever discrimination task and a cross self-administration paradigm: tests of generalization to pentazocine, buprenorphine, tilidine, and different doses of codeine.

The stimulus effects of codeine were assessed in three monkeys trained to perform first under the contingencies of a cross self-administration paradigm and then under a two lever discrimination task. Codeine-trained monkeys generalized to pentazocine, buprenorphine, and codeine under both procedures in doses different from the training dose. Codeine-trained monkeys did not generalize to tilidine. These results indicate that monkeys do not behave in a qualitatively different way when presented with the study drugs under both contingencies. However, there were marked quantitative differences between the generalization effects of doses of pentazocine, buprenorphine, and codeine to doses other than that used in training between the two paradigms. Much higher doses of codeine and pentazocine, but not of buprenorphine, were necessary for inducing generalization effects in the two lever task than in the cross self-administration procedure. The possible reasons for these quantitative differences are discussed. It is concluded that the cross self-administration procedure is more sensitive for the assessment of opioid-like stimulus properties of drugs than the two lever discrimination task.