Background and aims Several authors consider Internet-pornography-use disorder (IPD) as addictive disorder. One of the mechanisms that has been intensively studied in substance- and non-substance-use disorders is an enhanced attentional bias toward addiction-related cues. Attentional biases are described as cognitive processes of individual’s perception affected by the addiction-related cues caused by the conditioned incentive salience of the cue itself. It is assumed in the I-PACE model that in individuals prone to develop IPD symptoms implicit cognitions as well as cue-reactivity and craving arise and increase within the addiction process. Methods To investigate the role of attentional biases in the development of IPD, we investigated a sample of 174 male and female participants. Attentional bias was measured with the Visual Probe Task, in which participants had to react on arrows appearing after pornographic or neutral pictures. In addition, participants had to indicate their sexual arousal induced by pornographic pictures. Furthermore, tendencies toward IPD were measured using the short-Internetsex Addiction Test. Results The results of this study showed a relationship between attentional bias and symptom severity of IPD partially mediated by indicators for cue-reactivity and craving. While men and women generally differ in reaction times due to pornographic pictures, a moderated regression analysis revealed that attentional biases occur independently of sex in the context of IPD symptoms. Discussion The results support theoretical assumptions of the I-PACE model regarding the incentive salience of addiction-related cues and are consistent with studies addressing cue-reactivity and craving in substance-use disorders. Methods To investigate the role of attentional biases in the development of IPD, we investigated a sample of 174 male and female participants. Attentional bias was measured with the Visual Probe Task, in which participants had to react on arrows appearing after pornographic or neutral pictures. In addition, participants had to indicate their sexual arousal induced by pornographic pictures. Furthermore, tendencies toward IPD were measured using the short-Internetsex Addiction Test. Results The results of this study showed a relationship between attentional bias and symptom severity of IPD partially mediated by indicators for cue-reactivity and craving. While men and women generally differ in reaction times due to pornographic pictures, a moderated regression analysis revealed that attentional biases occur independently of sex in the context of IPD symptoms. Discussion The results support theoretical assumptions of the I-PACE model regarding the incentive salience of addiction-related cues and are consistent with studies addressing cue-reactivity and craving in substance-use disorders. Contrary to the general perception, pornography is not exclusively consumed by men, but receives growing attention by female users, even if the time of usage and choice of content differ from males’ consumption ( Daneback, Cooper, & Månsson, 2005 ; Ferree, 2003 ; Shaughnessy, Byers, & Walsh, 2011 ). Despite the limited number of studies on addictive behaviors in female users, there is an empirical evidence for similarities between male and female users ( Green, Carnes, Carnes, & Weinmann, 2012 ; Laier, Pekal, & Brand, 2014 ). The results for men and women are consistent with the previous findings, suggesting that sexual arousal and craving are main predictors for the development and maintenance of an IPD and are based on learned associations to internal and external cues ( Brand et al., 2011 ; Laier, Pawlikowski, Pekal, Schulte, & Brand, 2013 ). In addition, research suggests that attentional bias toward sexual stimuli can occur independently of sex. Kagerer et al. ( 2014 ) could show that male and female participants did not differ in reaction times in terms of attentional bias toward sexual cues. However, it is still unclear how an attentional bias interacts with symptoms of an IPD. Simple slopes for the second moderated regression analysis with s-IATsex-craving as dependent variable, the first moderated regression analysis with s-IATsex-craving as dependent variable, sex as independent variable, and maintained AB as moderator. No interaction was found and simple slopes did not differ significantly from zero For the mediation model, the latent variable “cue-reactivity and craving” was modeled by the craving measures sexual arousal t2, and the need to masturbate t2 since the delta scores for both measures were not significantly correlated with the severity of symptoms due to an IPD and attentional bias scores. ( 2016 ), which emphasizes an important role of implicit cognitions in the development and maintenance of Internet-use disorders including IPD. Implicit cognitions are considered as a result of interactions between indicators for sexual arousal and specific predispositions, for example, a high sexual excitability triggered by sexual cues and experiencing gratification while using Internet pornography. We could show that the subjective sexual arousal due to the presentation of pornographic cues as well as the sexual arousal and the need to masturbate afterward are related to indicators of attentional biases and partially mediates the effect of attentional bias on IPD. Sexual pictures might be associated with both drug- and non-drug rewards and therefore occur also in a non-clinical sample of men and women. However, it has to be noted that the selection of stimuli was not aligned for both sexes, but rather fits male user’s preference. Kagerer et al. ( 2014 ) argue that females’ attention to sexual cues is interfered when they get confronted with same-sex models as found in the study by Schimmack ( 2005 ). In terms of tendencies toward IPD, these sexual stimuli also became significant for female individuals.

Genetically identical twins can differ in their self-reported sexual orientations. However, whether the twins’ subjective reports reflect valid differences in their sexual orientations is unknown. Measures of sexual orientation, which are free of the limitations of self-report, include genital arousal and pupil dilation while viewing sexual stimuli depicting men or women. We examined these responses in 6 male twin pairs and 9 female twin pairs who reported discordant sexual orientations. Across measures, heterosexual male twins responded more strongly to women than to men. Their homosexual co-twins showed an opposite pattern. Heterosexual female twins responded equally to both sexes, whereas their homosexual co-twins responded somewhat more to women than men. These differences within pairs were similar to differences between unrelated heterosexual and homosexual males and females. Our study provides physiological evidence confirming twins’ discordant sexual orientations, thereby supporting the importance of the non-shared environment for the development of sexual orientation and sexual arousal. Measures of sexual orientation, which are free of the limitations of self-report, include genital arousal and pupil dilation while viewing sexual stimuli depicting men or women. We examined these responses in 6 male twin pairs and 9 female twin pairs who reported discordant sexual orientations. Across measures, heterosexual male twins responded more strongly to women than to men. Their homosexual co-twins showed an opposite pattern. Heterosexual female twins responded equally to both sexes, whereas their homosexual co-twins responded somewhat more to women than men. These differences within pairs were similar to differences between unrelated heterosexual and homosexual males and females. In general, sex and sexual orientation differences in sexual response are comparable for these measures. Heterosexual men respond substantially more to females than males, and homosexual men respond substantially more to males than females 5 – 8 ; thus, on average, genital arousal and pupil dilation patterns verify men’s reported sexual orientations. Women of all sexual orientations respond, on average, with some level of arousal to erotic stimuli representing either sex. However, this effect is especially common in heterosexual women, whereas homosexual women respond somewhat more to the same sex than other sex. This pattern can be seen in women’s genital responses and pupillary responses. We then regressed female twins’ genital arousal to the same sex or other sex onto their sexual orientation. Twin pairs were a random effect. Heterosexual female twins responded about equally to both sexes (the 95% confidence intervals of their average contrast score included the value of 0). In comparison, their homosexual co-twins showed somewhat more arousal to the same sex. Although in the predicted direction, this difference in sexual orientation was not significant, p = 0.11, β = 0.60 [−0.04, 0.91] (Fig. 3A ). Figure 3 Women’s genital arousal to sexual stimuli. In most cases, the 95% confidence intervals of heterosexual and homosexual twins’ average responses included the means of the corresponding heterosexual and homosexual unrelated groups (Figs 1 – 4 ). A set of regression analyses indicated that for male and female genital arousal, and male pupil dilation, twins and unrelated individuals did not significantly differ in the effect of sexual orientation on sexual response (0.39 < ps < 0.70, 0.03 < βs < 0.04, −0.20 < 95% CIs < 0.21). However, homosexual female twins dilated more strongly to the same sex than the other sex, as compared with unrelated homosexual women. The difference between twins and unrelated women in the effect of sexual orientation was significant, p = 0.04, β = 0.12 [0.01, 0.25]. We further note that unlike heterosexual female twins, unrelated heterosexual women dilated significantly more to the other sex than the same sex, p = 0.89, Mz = −0.04 [−0.48, 0.39], and p = 0.0008, Mz = −0.13 [−0.21, −0.06], respectively (Fig. 4A,B ). Discussion The present study suggests that identical twins with discordant self-reported sexual orientations differ in their patterns of physiological sexual arousal in a manner similar to unrelated heterosexual and homosexual individuals. Including or excluding the two bisexual females did not substantially affect results, and they were included in all analyses. For the sake of simplicity, we grouped bisexual women with lesbians into a “homosexual” group. Likewise, gay male twins were labelled “homosexual;” straight twins were labelled “heterosexual.” For the 6 male pairs, the mean age (SD) was 25.00 (9.19) and 100% were White. The mean age of the 9 female pairs was 26.22 (4.52) and 89% were White and 11% were Black. For males, predicted sexual orientation effects (standardized regression coefficients βs ) on genital arousal or pupil dilation to male or female stimuli fall between 0.70 and 0.90 6 . Thus, significant power of 0.80 can be achieved with a minimum of 5 heterosexual and 5 homosexual males. The sample of male twins met this minimum. For females, predicted sexual orientation effects fall between 0.20 and 0.30. Thus, significant power of 0.80 may only be achieved with a larger sample (a minimum of 80 women) than the recruited 9 heterosexual twins and 9 co-twins. For the weak effect, sexual orientation differences in female sexual arousal are often not significant, and a focus may be given on the expected difference in effect rather than on level of significance. For all twins, their average scores were approximately 1, suggesting monozygosity. Similar questions about zygosity are usually 95% accurate or higher, based on comparisons with blood group or DNA analyses 32 . Unrelated participants Unrelated participants took part in previously conducted studies and included 94 heterosexual men, 97 homosexual men, 77 heterosexual women, and 44 homosexual women with