Abstract A series of three experiments was conducted to test the proposal of Cicala and Owen ( Learning and Motivation, 1976 , 7, 356–367) that warning signal termination reinforces avoidance learning by permitting fear to dissipate and that a feedback signal reinforces avoidance learning through conditioned inhibition of fear. A CER procedure with a second-by-second analysis was used to measure fear elicitation and fear reduction following classical conditioning or avoidance training. The results show that a feedback signal produces rapid and marked fear reduction, suggesting an inhibitory process. Although warning signal termination produced fear reduction, the slowness of this process suggests fear dissipation and not an inhibitory process. The combination of both events reduced fear no more effectively than a feedback signal alone. It is suggested that warning signal termination is effective in some avoidance learning situations not because it reduces fear, but possibly because it helps terminate an avoidance response that closely resembles a species specific defense reaction.