Is fear present following sustained asymptotic avoidance responding?

The present experiment provided a direct test between a "fear" and an "expectancy" interpretation of avoidance in extinction. The test was suggested by Seligman and Johnston (1973). Three groups of rats were given discrete-trial avoidance training in a modified shuttlebox apparatus until they reached a criterion of either 5, 25, or 50 consecutive avoidance responses with a latency of less than four seconds (the duration of the CS). Asymptotic responding was reached quickly and continued to be maintained until avoidance training was discontinued. A transfer test using a lick suppression paradigm followed avoidance training. Seligman and Jonhston predicted that following asymptotic performance, fear (as indexed by lick suppression) would extinguish quickly and not be present during sustained avoidance responding. On the other hand, fear theory argued that "fear" needed to be present for avoidance responding to occur. The data clearly supported fear theory. All groups displayed good suppression to the CS, including the group which experienced 50 consecutive short-latency avoidance responses. Additional implications of the data were discussed.