抄録
The present goldfish experiments in the shuttlebox were attempted to separate and test the feedback (informational) and safety-signal (fear inhibition) hypotheses, both of which may account for the reinforcing effects of the FS in avoidance conditioning. In Experiment I, four groups of goldfish (N=48) were trained with (or without) response contingent FS and with (or without) shock contingent FS in a Sidman avoidance schedule (S-S=R-S=30s). The shock contingent FS tended to facilitate acquisition only in the response measure. The response contingent FS, on the other hand, facilitated acquisition only in the US measure. In Experiment II, four groups of goldfish (N=48) were trained with (or without) response contingent FS and with a fixed (30 s) or variable (15-45 s) R-S interval in a Sidman avoidance schedule (S-S=30 s). The results showed that, in both response and US measures, the FS facilitated avoidance acquisition, whereas variable R-S deteriorated it. As a whole, the results of the present experiments were considered to support the safety-signal (fear inhibition) hypothesis.
