Diagnosing pedohebephilia is fraught with obstacles given the tabooed nature of this sexual preference. The viewing reaction time effect (VRT) provides a non-intrusive indirect measure of sexual interest in minors. In forensic populations, the ability of the difference between the latencies while viewing child and adult sexual stimuli (VRT index) to discern child sexual offenders from a range of control groups has been ascertained meta-analytically. Given that the effect has been studied almost exclusively in forensic samples, its dependence or independence on prior overt (deviant) sexual behavior remains unclear. The present study sought to examine the relationship of prior sexual and non-sexual behaviors with the VRT in a sample of 282 self-referring, help-seeking men with and without pedohebephilia with and without a history of prior child sexual offenses (CSO) or a use of child sexual abuse materials (CSAM) recruited outside a forensic context. We found that (1) the clinical diagnosis of pedohebephilia but not prior CSO or CSAM showed a significant association with the VRT index; (2) the discriminatory ability of the VRT index did not differ significantly between samples with and without a history of prior overt sexual behavior with children; (3) the VRT index correlated positively with a behavioral marker of pedohebephilia in a subsample of individuals with prior judicially detected or undetected overt sexual behavior with children; and (4) in the same subsample, the VRT index correlated positively with markers of sexual interests in minors or hypersexuality but not of antisociality. Equivalence testing failed to refute a potential effect of prior sexual behavior on the VRT index. Our study showed that the VRT may provide an unintrusive diagnostic tool for pedohebephilia. The effect of prior overt sexual behavior with children needs further examination.