Ponseti J;Granert O;Jansen O;Wolff S;Mehdorn H;Bosinski H;Siebner H
Assessment of sexual orientation using the hemodynamic brain response to visual sexual stimuli
J Sex Med 2009, 6(6), , 1628-1634

INTRODUCTION: The assessment of sexual orientation is of importance to the diagnosis and treatment of sex offenders and paraphilic disorders. Phallometry is considered gold standard in objectifying sexual orientation, yet this measurement has been criticized because of its intrusiveness and limited reliability. AIM: To evaluate whether the spatial response pattern to sexual stimuli as revealed by a change in blood oxygen level-dependent (BOLD) signal can be used for individual classification of sexual orientation. METHODS: We used a preexisting functional MRI (fMRI) data set that had been acquired in a nonclinical sample of 12 heterosexual men and 14 homosexual men. During fMRI, participants were briefly exposed to pictures of same-sex and opposite-sex genitals. Data analysis involved four steps: (i) differences in the BOLD response to female and male sexual stimuli were calculated for each subject; (ii) these contrast images were entered into a group analysis to calculate whole-brain difference maps between homosexual and heterosexual participants; (iii) a single expression value was computed for each subject expressing its correspondence to the group result; and (iv) based on these expression values, Fisher's linear discriminant analysis and the kappa-nearest neighbor classification method were used to predict the sexual orientation of each subject. MEAN OUTCOME MEASURE: Sensitivity and specificity of the two classification methods in predicting individual sexual orientation. RESULTS: Both classification methods performed well in predicting individual sexual orientation with a mean accuracy of >85% (Fisher's linear discriminant analysis: 92% sensitivity, 85% specificity; kappa-nearest neighbor classification: 88% sensitivity, 92% specificity). CONCLUSION: Despite the small sample size, the functional response patterns of the brain to sexual stimuli contained sufficient information to predict individual sexual orientation with high accuracy. These results suggest that fMRI-based classification methods hold promise for the diagnosis of paraphilic disorders (e.g., pedophilia)