An extension of the preparedness hypothesis to the conditioning of male sexual responses

The current study examined the extension of Seligman's Preparedness Hypothesis to the classical conditioning of sexual arousal through the use of a repeated measures, single-subject design. Nine males were each given a maximum of 40 conditioning trials to a purportedly prepared sexual object (boot), a nonprepared culturally-determined sexual object (bra), and a nonprepared nonsexual object (cup). The boot was defined as a possibly prepared sexual stimulus based on a review of the fetishism literature. During independent sessions, each CS was paired with erotic movies in a ^CS0-70"^-US, 40-70" schedule. Experimental phases were Pretest, Baseline, Conditioning, Probe, Generalization, Extinction, and Spontaneous Recovery. Penile responses and self-report of arousal were used to assess sexual responses. Subjects were considered conditioned if they made CRs on five of six successive trials.

Of the nine subjects, six conditioned to one or more CSs. Five of these six conditioned to all CSs. One additional subject showed mixed evidence for conditioning. Physiological conditioning largely occurred below subjects' awareness levels. Although there was a tendency for the bra to elicit more CRs prior to conditioning and to condition in fewer trials, no significant differences were found among CSs in number of generalization responses, in trials to extinction during spontaneous recovery, or in self-reports of arousal. Significant differences were found in trials to extinction with the boot requiring the most trials. The boot may therefore be a prepared sexual stimulus as defined by its relative resistance to extinction. This relative resistance to extinction may account for its apparent special status as a sexual signal. The pattern of results found indicates that preparedness may not occur across learning parameters.

In a second experiment subjects estimated their arousal on a 7-point scale during and at the completion of the conditioning sessions' trials and while viewing 6}- and 9-minute segments of erotic stimuli. Subjective ratings and physiological responses were found to correlate highly especially when large and long-lasting. physiological responses had been elicited. These factors were more important than whether the rating was made during or after the presence of the erotic stimuli.

A parametric analysis of the Barlow strain gauge was also performed. Mean pen readings of strain gauge output based on multiple measurements of different penis-analogue diameters showed that the sensor was not fully linear. Use of penis-analogue diameters was found to markedly exaggerate penis size. These findings are problematic for statistics assuming linearity such as "percentage of full erection." Because of these difficulties, a regression line was computed using caliper measures of actual penis diameters and the corresponding pen reading. This regression line was used in the preparedness and correlational studies described above.