~ Pergamon 0005-7967(95)00070-4 Behav. Res. Ther. Vol. 34, No. 4, pp. 389-394, 1996 Copyright© 1996ElsevierScienceLtd Printed in Great Britain. All rights reserved 0005-7967/96$!5.00+ 0.00 VIEWING TIME AS A MEASURE OF SEXUAL INTEREST AMONG CHILD MOLESTERS AND NORMAL HETEROSEXUAL MEN GRANT T. HARRIS *j, MARNIE E. RICE l, VERNON L. QUINSEY 2 and TERRY C. CHAPLIN l ~Mental Health Centre, 500 Church St, Penetanguishene, Ontario, Canada L9M 1G3 and 2Queen's University, Kingston, Ontario, Canada (Received 10 May 1995; in revised form 3 October 1995) Sununary--Although phallometric assessment is the best scientific method for measuring male sexual interest, it is intrusive and highly technical. We examined viewing time as an unobtrusive and technically simple measure of sexual preference and compared the discrimination obtained by viewing time measures with that obtained by phallometric measures. Slides of nude males and females of various ages were shown to child molesters and normal men while their viewing times were recorded. Subjects then rated the sexual attractiveness of the stimulus persons. Phallometric assessments using the same stimulus categories were also given to some of the Ss. Deviance scores calculated from the viewing time data significantly discriminated between the child molesters and the normals, although the discrimination achieved was less than that obtained using phallometric measures. Sexual attractiveness ratings did not differentiate the two groups. Among the normal men, viewing time and sexual attractiveness ratings were highly correlated; but the correlation was much lower for child molesters. Viewing time shows considerable promise as an unobtrusive measure of male sexual interest. INTRODUCTION Phallometric assessment has become the most scientifically accepted method of measuring male sexual interests (Quinsey, 1988; Quinsey & Earls, 1990). The ability of phallometric tests of age preferences to discriminate child molesters from other men is well established (Harris, Rice, Quinsey, Chaplin & Earls, 1992). Some studies have shown that when appropriate stimuli and scoring methods are used, there is little overlap between the responses of the two groups (Chaplin, Rice & Harris, 1995; Harris et al., 1992). Sexual interest in children, as measured by phallometric tests, has also been shown to predict sexual and violent recidivism among child molesters (Barbaree & Marshall, 1988; Malcolm, Andrews & Quinsey, 1993; Rice, Quinsey & Harris, 1991; Quinsey, Rice & Harris, 1995). A major problem with phallometric testing, however, is intrusiveness. A device must be attached to the penis to monitor changes in size. Phallometric tests require specialized equipment and considerable skill from the examiner. Despite its reliability and validity, it is unlikely that such a specialized and intrusive test could be widely used in screening men who apply to work with children. Also, because of its intrusiveness, the use of phallometric testing with children and adolescents raises many ethical questions. Although some investigators have used phallometric tests with adolescent sex offenders (Becker, Kaplan & Tenke, 1992), there are no normative data about the sexual preferences of children or adolescents. A less intrusive measure would be very useful for studying the sexual interests of adults, adolescents, and children. Another measure of sexual interest is self-reported sexual arousal or sexual attraction. Among adult male nonoffenders, there are close correspondences among sexual behavior, self-reported sexual interest, and phallometrically-determined sexual interest (Quinsey, Steinman, Bergerson & Holmes, 1975). However, self-reported sexual interests bear little relationship to phallometrically determined preferences among child molesters, whose self-reported sexual interests are similar to *Author for correspondence: Research Department, Mental Health Centre, 500 Church Street, Penetanguishene, Ontario, Canada L9M 1G3. 389