The Relation between Peak Response Magnitudes and Agreeement in Diagnoses Obtained from Two Different Phallometric Tests for Pedophilia Amy D. Lykins, James M. Cantor, Michael E. Kuban, Thomas Blak, Robert Dickey, Philip E. Klassen, and Ray Blanchard Centre for Addiction and Mental Health, Toronto, Ontario, Canada Department of Psychiatry, University of Toronto Contact Information: Amy_Lykins@camh.net INTRODUCTION Abstract Phallometric testing is widely considered the best psychophysiological procedure for assessing erotic preferences in men. Researchers have differed, however, on the necessity of setting some minimum criterion of penile response for ascertaining the interpretability of a phallometric test result. Proponents of a minimum criterion have generally based their view on the intuitive notion that “more is better” rather than any formal demonstration of this. The present study was conducted to investigate whether there is any empirical evidence for this intuitive notion, by examining the relations between magnitude of penile response and the agreement in diagnoses obtained in two test sessions using different laboratory stimuli. The results showed that examinees with inconsistent diagnoses responded less on both tests, and that examinees with inconsistent diagnoses responded less on the second test after controlling for their response on the first test. Results also indicated that at response levels below 1cc, diagnostic consistency was no better than chance, supporting the establishment of a minimum response level criterion. Participants 79 men referred to the Kurt Freund Laboratory because of paraphilic, criminal, or otherwise problematic sexual behavior, all participants completed both phallometric tests Mean age at first test: 34.05 years (SD = 12.03) Mean age at second test: 38.08 years (SD = 11.95) Phallometric testing Volumetric plethysmographymeasures penile blood volume change Test stimuli presented in discrete trials, each trial presented stimuli from one and only one class (e.g., prepubescent girls, pubescent boys, etc.) Walking Nudes Test Stimuli consisted of 14 s film clips of nude children and adults smiling and walking slowly toward the camera (but not engaging in any overtly sexual or flirtatious behavior) Nine categories: adult women and men, pubescent girls and boys, girls and boys 8-11 years, girls and boys 5-8 years, neutral stimuli Narratives-Slides Test Stimuli consisted of static images of adult, prepubescent, and prepubescent males and females Three views: full-length front, full-length rear, close-up genital Audiotaped narratives which described sexual interactions with individuals of the gender and age of visual stimuli presented Diagnoses from the Two Phallometric Tests Output Indices on the Two Tests as a Function of Diagnostic Consistency Percentage Deviation from Expected Output Index on the Narratives-Slides Test as a Function of Diagnostic Consistency Main Findings 27% of sample received inconsistent diagnoses Examinees with inconsistent diagnoses responded less on both tests Examinees with inconsistent diagnoses responded less on the newer test after controlling for their response on the older test Consistency not related to length of time between the first and second test administrations Recommendations Establish a generally agreed-upon minimum response criterion (1 cc volumetric phallometry, 2.5 mm increase circumferential) Increase penile tumescence in some manner (e.g., pharmaceutical agents, stronger sexual stimuli) Include tasks during testing to direct attention to sexual stimuli and reduce purposeful distraction Clinicians should address degree of confidence in clinical diagnosis in their reports as it relates to phallometric test results RESULTS DISCUSSION Walking Nudes Test Narratives-Slides Test Negative for pedo/hebephilia Positive for pedo/hebephilia 43 11 10 15 METHODS