In Experiment I, rats \vcrc trained to press a bar to obtain food in the presence of a tone. A gradient of stimulus generalization was obtained by testing groups of .S' s to stimuli varying along a tonal dimension, with each 5 being tested at only one stimulus until it reached an extinction criterion. Following this phase, all 5s were shifted back to the original training tone, and the extinction procedure was continued. This second shift yielded a gradient of generalization of extinction. Experiment II held constant the number of S-f-responses made to the generalized stimuli, while Experiment III eliminated S-responses. Experiment IV eliminated all responding to the generalized stimuli while holding constant stimulus exposure time. A typical generalization of extinction gradient was obtained for all experiments with the exception of Experiment IV.

No abstract

Two experiments were reported in which rats were presented a serial conditioned stimulus procedure consisting of 2 components: Si followed by S 2 (Si/S 2 ). In both experiments the stimulus similarity of Si to 82 in terms of tonal frequency was systematically manipulated. Experiment 1 tested the effects of similarity of components in a standard shuttle-box avoidance situation. Experiment 2 employed a conditioned emotional response paradigm measuring the suppression of consummatory licking. The results of both experiments suggest that the amount of fear elicited by Si is a direct function of the stimulus similarity of Si to 82. These data support a generalization interpretation of fear transference.

No abstract

Four profoundly retarded male residents of a state institution learned to comply with the instructions "come here," "stay," and "sit." Acquisition occurred in a special therapy room and instructions were presented in a fixed sequence. All subjects learned to comply with the instructions in the training environment. Testing occurred in the subject's ward dayroom where instructions were presented in a random sequence. Despite dramatic changes in the stimulus conditions associated with the transfer to the testing situation the subjects performed close to their asymptotic level. Three experimentally naive subjects participated in a second experiment which was similar to the first, except that the number of acquisition sessions and the number of trials per session were reduced. The reduction in number of acquisition trials did not appear to affect performance in the testing environment. The possibility of terminating off-task behavior of institutionalized individuals by eliciting previously acquired instruction following behavior was discussed.