Effects of Stimulus Content on Second‐Order Electrodermal Conditioning in Humans

Two experiments investigated human second‐order conditioning with electrodermal activity as the dependent variable. In Experiment 1 (N = 144), Group PP received first‐order pairings of CS1 and a white noise US of 100dB. In phase 2, Croup PP received pairings of CS2 and CS1, Group PR received first‐order pairings of CS1, and the US during phase 1, and random presentations of CS: and CS, during phase 2. Group RP received random presentatiuns of CS2, and the US during phase 1, and pairings of CS: and CS, during phase 2. Within each of Groups PP, PR. and RP. the CSs were pictures of snakes and spiders, beetles and lizards, or flowers and mushrooms. All CS durations were 5 s, the US duration was 1 s, and the interstimulus interval was 5 s. First‐order conditioning was clearly established in Groups PP and PR. and in phase 2. second‐order conditioning was evident in Group PP. However, neither first‐order nor second‐order conditioning was affected, in either acquisition or extinction, by fear‐relevance of the CS. Experiment 2 (N‐ 72) employed only groups who received stimulus pairings in both phases, and crossed fear‐relevance of the CSs (fear‐relevant vs. fear‐irrelevant) with three US conditions (shock after work‐up, noise after work‐up, and noise of lOOdB). Second‐order responding during extinction was greater in the fear‐relevant condition than in the fear‐irrelevant condition regardless of the type of US employed. The results are discussed in terms of the preparedness theory of the acquisition of phobias.