Second-order conditioning and overshadowing in the observational conditioning of fear in monkeys

Several studies have recently documented strong and persistent observational conditioning of snake fear in rhesus monkeys. The studies presented here explore the extent to which two well-known phenomenon of direct classical conditioning-second-order conditioning and overshadowing-also occur in the context of observational conditioning. In Experiment 1, rhesus monkeys that had previously acquired a fear of snakes through observational learning underwent six sessions of second-order conditioning in which a black-striped box, the second-order conditioned stimulus (CS), was paired with snake stimuli, the first-order CS. Two of three measures of fear indicated that small but significant amounts of fear were conditioned to the second-order CS. In Experiment 2, a modified overshadowing paradigm was employed to determine whether a fear-relevant CS such as a snake would overshadow a fear-irrelevant CS such as a flower. Observer monkeys underwent six sessions of observational conditioning during which they watched models behaving fearfully in the presence of a compound snake/flower stimulus. Observers acquired a fear of snakes but not a fear of flowers. Implications of these findings for understanding the origins of human fears and phobias are discussed.