Shock for right and wrong responses during learning and extinction in human subjects.

Abstract

Seventy human subjects learned a pattern of 20 right-left choices in a temporal maze under 4 conditions: no-shock, shock for right responses, shock for wrong responses, and shock for both responses. Correct responses were reinforced by a tone of 512 cps, which was omitted during extinction. Learning under all 3 shock conditions was significantly faster than with no shock, and the shock-both group learned significantly faster than the shock-right group. The shock-right and shock-wrong groups did not differ significantly. These results are consistent with previously published studies using animal subjects. In extinction, the shock-learning groups were each divided into 2 subgroups, one extinguished without shock and one extinguished with shock continuing as it had in learning. Where the conditions for differential secondary reinforcement occur, extinction tends to be significantly delayed over non-shock extinction conditions. The fastest extinction was shown by the group shocked for both responses in learning, and then shocked for both responses in extinction. Results are discussed in terms of the anxiety hypothesis and Hull''s theoretical structure.