Abstract
Rats were treated with discriminative punishment in a 2 lever Skinner box, and their lever-pressings were reinforced by apple juice. In Experiment 1, responses to one lever (abbr. PUNs) were punished on a FR1 schedule in the presence of signal tone, but never to the other lever (abbr. UNPs). Punishment-component of this procedure was more distinct than CER-component, both of which composed discriminative punishment. PUNs were completely suppressed, but UNPs showed no conditioned suppression. Then, in Experiment 2, the rats were divided into two groups. PUNs of group V1 were punished on a VR10 schedule, while the V2 on a VR20 during the tone presentation. PUNs and UNPs of both groups showed little suppression. Finally, in Experiment 3, these two groups were yoked as to the electric-shock presentation. While in group V1 38% of the elecric-shocks received were response-noncontingent, in the V2 68%. UNPs of group V2 showed conditioned suppression, but not those of the V1. This suggests that the more indistinct the punishment-component is, the more easily the subjects acquire conditioned suppression against the signal stimulus.
