Note on the invariance of response latency in shuttlebox avoidance learning

Rats trained to avoid shock by running in a shuttlebox showed an orderly increase over trials in the probability of avoidance but no corresponding change over trials in the latency of the avoidance response. Both responding in the presence of shock (escapes) and responding prior to shock onset (avoidances) appeared to be stationary in time. Latencies were found to vary with different experimental conditions but in every case to be invariant over trials.     

Freezing as an avoidance response: Another look at the operant-respondent distinction

Groups of rats were trained with shock either contingent on freezing (punishment procedure) or contingent on not freezing (avoidance procedure). Although the different contingencies produced different levels of freezing behavior, these levels were attained immediately rather than over a number of trials. This result, together with the results of control rats, suggest that while freezing can be controlled by both punishment and avoidance procedures, in both cases the effects on freezing are due to elicitation rather than learning.