Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Pupillary Response as an Age-Specific Measure of Sexual Interest

Pupillary Response as an Age-Specific Measure of Sexual Interest Arch Sex Behav (2016) 45:855–870 DOI 10.1007/s10508-015-0681-3 OR IG INA L PAP E R Pupillary Response as an Age-Specific Measure of Sexual Interest 1 1 1 • • Janice Attard-Johnson Markus Bindemann Caoilte O Ciardha Received: 4 July 2014 / Revised: 7 December 2015 / Accepted: 11 December 2015 / Published online: 8 February 2016 The Author(s) 2016. This article is published with open access at Springerlink.com. This article is published with open access at Springerlink.com Abstract In the visual processing of sexual content, pupil dila- 2002;Sell, 1997) and the assessment of unhealthy and inappro- tion is an indicator of arousal that has been linked to observers’ priate sexual desires in clinical and forensic settings (Gannon, sexual orientation. This study investigated whether this measure Ward, & Polaschek, 2004; Laws & O’Donohue, 2008). Experi- can be extended to determine age-specific sexual interest. In two mental psychology has contributed to this field by developing a experiments, the pupillary responses of heterosexual adults to number of assessment methods (e.g., Gress, 2005;Laws&Gress, images of males and females of different ages were related to 2004; Mokros, Dombert, Osterheider, Zappala `, & Santtila, 2010; self-reported sexual interest, sexual appeal to the stimuli, and a O Ciardha & Gormley, 2012, 2013). Of these, viewing time, child molestation proclivity scale. In both experiments, the pupils which reflects the duration for which particular content is studied, of male observers dilated to photographs of women but not men, is now a widely utilized measure of interest in sexually appetitive children, or neutral stimuli. These pupillary responses corre- materials (e.g., Lykins, Meana, & Strauss, 2008; Rupp & Wallen, sponded with observer’s self-reported sexual interests and their 2007). The viewing of visual content is also accompanied by sexual appeal ratings of the stimuli. Female observers showed automatic changes in observers’ pupil size (Bradley, Miccoli, pupil dilation to photographs of men and women but not chil- Escrig, & Lang, 2008), which appear to be particularly sensitive dren. In women, pupillary responses also correlated poorly with to sexual arousal (Bernick, Kling, & Borowitz, 1971). While this sexual appeal ratings of the stimuli. These experiments provide pupillary response was first explored 40 years ago with some ele- initial evidence that eye-tracking could be used as a measure of mentary methods (Hess, Seltzer, & Shlien, 1965), it has received sex-specific interest in male observers, and as an age-specific little attention since. In this study, we attempt to replicate those index in male and female observers. early findings with contemporary eye-tracking equipment to determine if it can be used to assess sexual interests. We not only Keywords Sexual interest  Eye-tracking  wish to explore whether increased pupil size can provide an index Pupillary response  Sexual appeal of adults’ sexual interest in other adults but also whether this index is age-specific. This addition might be important for clinical and forensic practice. Introduction Viewing time is a measure that is linked to a person’s inter- ests and motivations (Henderson, 2003; Isaacowitz, 2006). In The measurement of sexual arousal and observers’ sexual inter- relation to sexual interest, viewing time has been used to mea- ests is important for psychological research and practice. For sure interest in preferred over non-preferred figures. One way example, this is necessary to conduct research into sexual orien- for measuring viewing time in these paradigms is to record tation causes and consequences (Mustanski, Chivers, & Bailey, observers’ response times while they rate the sexual appeal of pictures of men and women (Gress, 2005; Gress, Anderson, & Laws, 2013; for reviews, see Akerman & Beech, 2012;Laws & & Janice Attard-Johnson Gress, 2004; Snowden, Craig, & Gray, 2011). In these studies, j.attard@kent.ac.uk longer response times for a specific stimulus type correspond to the reported sexual interest for that category (Quinsey, Ketset- School of Psychology, University of Kent, Canterbury zis, Earls, & Karamanoukian, 1996) and physiological measures CT2 7NP, UK 123 856 Arch Sex Behav (2016) 45:855–870 of sexual arousal (Abel, Huffman, Warberg, & Holland, 1998). In this study, we explore an alternative eye-tracking mea- For example, heterosexual male observers tend to make slower sure that might be more sensitive and not under top-down con- responses when rating pictures of women than of men (Israel & trol. The pupils respond automatically to external stimulation, Strassberg, 2009) and prepubescent children (Harris, Rice, Quin- such as changes in lighting conditions, by increasing (dilating) sey, & Chaplin, 1996; Quinsey et al., 1996). Female heterosexual or decreasing (constricting) in size. A similar pattern is also observers also show age preferences in these viewing time para- found as an arousal response to pleasant and unpleasant stimuli digms (Ebsworth & Lalumiere, 2012; Quinsey et al., 1996)but are (Bradley et al., 2008). This dilation has been linked to the acti- inconsistent in their responses to sexually preferred and non-pre- vation of the autonomic nervous system (Zuckerman, 1971)and ferred adults (Ebsworth & Lalumiere, 2012; Israel & Strassberg, appears to be impervious to top-down control. It has been shown, 2009; Lippa, Patterson, & Marelich, 2010; Quinsey et al., 1996). for example, that observers cannot enlarge or reduce pupil size at While the response time-based assessment of viewing time will in the absence of a visual stimulus (Laeng & Sulutvedt, 2014) is an indirect measure of sexual interest, it is possible to achieve or suppress pupil dilation (for a review, see Laeng, Sirois, & Gre- similar results more directly by tracking observers’ eye move- deba ¨ ck, 2012). These characteristics might make pupillary ments. During visual processing, eye gaze is directed towards response an ideal measure for the assessment of sexual interest. scene content that matches a viewer’s personal interest (Calvo & While this is an interesting possibility, the pupillary response Lang, 2004), including longer fixations on sexually preferred to sexual arousal has received little research attention. In an early human figures (Fromberger et al., 2012b; Hall, Hogue, & Guo, study, Hess et al. (1965) showed five hetero- and five homo- 2011; Rupp & Wallen, 2007; for a review, see Rupp & Wallen, sexual males images of nude men and women while filming 2008). Heterosexual male observers, for example, view women the observers’ eyes at a rate of two frames per second. Twenty for longer than men (Lykins et al., 2008). These viewing patterns measurements were obtained for each stimulus by manually also appear to correspond to the sexual content on display (Hall measuring pupil diameter at each frame of the video footage. et al., 2011; Rupp & Wallen, 2007; Suschinsky, Elias, & Krupp Despite this elementary approach, a clear pupillary response 2007). For example, male and female observers predominantly was found whereby all heterosexual males exhibited larger study the faces of fully clothed persons (Hewig, Trippe, Hecht, pupils to pictures of women than men. By contrast, all but one Straube, & Miltner, 2008). However, female observers increase of the homosexual males showed larger pupil responses to pic- fixations to the body in semi-clothed stimuli (Rupp & Wallen, tures of men than women. These promising results were re- 2007) and male observers show a corresponding shift to pictures examined shortly after with the addition of female observers of nude women (Nummenmaa, Hietanen, Santtila, & Hyo ¨ na ¨, (Scott, Well, Wood, & Morgan, 1967). Here, observers were pre- 2012). These data therefore indicate that eye movements are sen- sented with semi-nude and clothed images of men and women. sitive to adult observers’ sexual interest in other adults. Male observers demonstrated more pupil dilation to semi-nude Viewing patterns also appear to be age-specific. For exam- women than any other stimuli. Female observers did not show ple, male and female adult observers fixate on figures of their different pupil responses to semi-naked and clothed stimuli or preferred age (20-year olds) more than babies and 60-year olds male and female targets. However, a subsequent experiment (Hall et al., 2011). However, whereas non-paedophilic adult also recorded a pupil dilation effect in female observers that males preferentially fixate on pictures of adults over children, appeared to be related to sexual interest (Hamel, 1974). In this paedophilic males show the reverse pattern (Fromberger et al., study, female observers showed increases in pupil size that were 2012a;Frombergeret al., 2013). This indicates that eye move- directly related to the degree of nudity of pictures of male, but not ments are not only sensitive to adult observers’ sexual interest in of female, models. other adults, but can also distinguish between such interest in Despite these promising results, there have been no attempts adults and children. to replicate these findings until recently. Rieger and Savin-Wil- Despite these advantages, fixation behavior is an index of liams (2012) showed hetero-, homo-, and bisexual observers sex- sexual interest that is vulnerable to top-down control. Observers ually explicit videos, while pupillary responses were recorded could, for example, conceal their sexual interest by diverting with contemporary eye-tracking equipment. This study repli- attention to other visual content (Bindemann, Burton, Langton, cated the clear relationship between sexual orientation and Schweinberger, & Doherty, 2007). This limitation could be pupil dilation that Hess et al. (1965)had foundinmaleobservers. overcome by considering only the initial fixation to a stimulus However, similar to Scott et al. (1967), pupillary responses in display, which might reflect a covert and automatic orientation heterosexual female observers were comparable when viewing response to pre-attentively selected stimuli of sexual interest. In footage of men and women. In a subsequent experiment, Rieger line with this reasoning, heterosexual adult males tend to direct et al. (2015) extended these findings to show that pupillary more initial fixations at women than men (56 vs. 44 %) and responses to sexually explicit images reflect the sexual orien- young girls (57 vs. 43 %; see Fromberger et al., 2012b). How- tation of male observers, but not of heterosexual female obser- ever, the difference between these percentage fixations is not vers, similarly to genital arousal. These findings indicate that indicative of a sensitive measure of involuntary behavior. pupillary response is a useful alternative for measuring sexual 123 Arch Sex Behav (2016) 45:855–870 857 interest in male observers. In addition, the lack of specificity in stimulus categories. This manipulation can decrease image heterosexual female observers converges with a broad range of quality by reducing light–dark contrasts. A third version of these assessment methods (e.g., genital arousal, self-reported sexual scenes was therefore also included, in which image quality of the arousal and attraction, response time, and viewing time; Chi- original photographs was enhanced with graphics software. vers, 2005; Chivers, Rieger, Latty, & Baily, 2004; Ebsworth & Lalumie `re, 2012;Lippa, 2006, 2007, 2012; Lippa et al., 2010; Suschinsky, Lalumiere, & Chivers, 2009). This is an interest- Experiment 1 ing finding because it suggests that pupillary responses to sex- ual content are also consistent with more established measures Method in the literature. While few studies have focussed on pupil dilation as a mea- Participants sure of sexual interest for photographs of adults, there has been even less research on pupillary responses to persons of differ- A total of 44 students (22 male and 22 female) from the School ent ages. An early study compared these pupillary responses in of Psychology at the University of Kent participated in this study incarcerated male pedophiles and non-pedophiles to images of in return for a small payment or course credits. Participants com- nude women and immature girls (Atwood & Howell, 1971). pleted the Kinsey scale for the assessment of sexual orientation This experiment revealed greater pupil dilation in 90 % of non- as part of a pre-screen on our online recruitment system. This is a paedophilic observers to pictures of women, but a pupil constric- seven-point scale in which a score of ‘‘0’’ represents complete tion to the same pictures in 80 % of pedophiles. Conversely, heterosexuality and ‘‘6’’ complete homosexuality. Only partic- images of girls produced dilation in 90 % of pedophiles and a ipants who reported to be completely heterosexual (i.e., report- constriction or no change in 50 % of the non-pedophilic control ing ‘‘0’’ on the Kinsey scale) wereinvitedtotakepart(Kinsey, subjects. Pomeroy, & Martin, 1948; Kinsey, Pomeroy, Martin, & Gebhard, Up to now, there have been no documented attempts to repli- 1953). The mean age of participants was 21.8 years (SD = 4.2; cate these findings. This is surprising considering the potential range 18–35 years). All reported normal or corrected-to-normal applied value of such a measurement (e.g., the assessment of vision. child sex offenders). In this exploratory study, we investigated whether pupil dilation can provide an age-specific indication of Materials a person’s sexual interests. For this purpose, heterosexual male and female observers were presented with images of beach The stimuli consisted of natural beach scenes portraying men, scenes that contained semi-clothed adults and children, while women, and children (5 scenes for each of these four categories). their eye movements and pupil sizes were recorded. These scenes To determine the approximate age of these categories, ten obser- contained only a single person or no persons in the case of a set of vers (5 males, 5 females) estimated the age of the people in the comparison landscape beach scenes. We expected the different scenes in a pilot study. This revealed a mean age of 26.4 years person content of these scenes to draw attention depending on the (SD = 2.1) for men, 22.8 years (SD = 2.6) for women, 5.7 years sexual interests of the observers. For example, heterosexual male (SD = 1.1) for boys, and 4.7 years (SD = 1.4) for girls. The age observers were anticipated to fixate on women more frequently of the children therefore corresponds to stage 1 (prepubescent) than men (see Hewig et al., 2008;Lykinsetal., 2008;Rupp& of the Tanner stages of sexual development (see Tanner, 1978). Wallen, 2007). Of particular interest here was whether these Additionally, a set of control beach scenes without any person observers would also show an increase in pupil size to images content (5 scenes) was included, resulting in a total of 25 scenes. of sexually preferred adults in comparison with sexually non- People were portrayed in swim or leisure wear. All stimuli were preferred adults and children. purchased from an internet photograph database (www.most As a secondary aim, we also sought to examine how pupil- photos.com) and were selected to be of similar composition lary responses to people of sexual interest are affected by image and size, and to depict the persons in similar poses and with a luminance. The pupils constrict in response to light (i.e., increased comparable level of clothing (see Fig. 1). To confirm that these luminance) to protect the cells of the retina (Bergamin & Kardon, targets were of similar size, their percentage occupancy area in 2003; Ellis, 1981). If this differentially affects the stimulus cate- the scenes was calculated. This showed that all person categories gories in the current study, then this could influence the measure- occupied a similar amount of space in our scenes (mean = 7.1 %, ment of pupil responses as an index of sexual interest. In turn, it is SD = 3.4, range across person categories = 6.6–7.7 %; one- possible that the pupillary response to sexual content is clearer factor ANOVA: F(3, 19) = 0.14, p = 0.94). when luminance is controlled across different stimulus cate- In addition, three versions were created of each scene that gories. To explore this possibility, the original photographs of were identical in all aspects except for image quality. This the beach scenes were compared with alternative versions, in resulted in a total of 75 scene images. In the original quality which the mean luminance was equated across the different condition, the image quality of the downloaded photographs 123 858 Arch Sex Behav (2016) 45:855–870 Fig. 1 The stimuli of the original quality condition in Experiment 1 was retained. In the high-quality version, the images were pro- Table 1 Mean luminance, standard deviation, and the minimum and maximum luminance values of images within a stimulus category for the cessed by applying the ‘‘Auto Levels,’’ ‘‘Auto Contrast,’’ and original, high-quality, and luminance-controlled images for all scene ‘‘Auto Color’’functions in Adobe Photoshop CS3 to artificially conditions enhance the original photographs. Finally, to create a luminance- Mean SD Max Min controlled version of the stimuli, the photographs were divided into groups of five (one of each category) based on similar lumi- Original quality nance values and standard deviation. A mean luminance value Men 166 25 190 125 and standard deviation were calculated for each of the five groups. Women 160 29 200 125 Each photo within a group was then re-adjusted to obtain the mean Boys 169 42 218 111 luminance and standard deviation that matched the group value. Girls 190 35 224 133 Therefore, at least one image from each category (men, women, No-person 165 28 190 127 boys, girls, no-person landscapes) had precisely matched lumi- High quality nance values. This particular group-based approach was adopted Men 167 23 186 131 to avoid the extreme deviation from the natural luminance values Women 163 20 182 130 of individual scenes. This can occur when a single mean lumi- Boys 171 41 221 123 nance value is derived for large stimulus sets, which can result in Girls 184 38 211 122 some highly distorted and unnatural looking images. Table 1 No-person 152 16 180 143 shows the overall mean luminance values and standard devia- Luminance controlled tion for the different image categories for all scenes. Example Men 162 18 194 152 stimuli are shown in Fig. 2. Women 162 18 194 152 Two questionnaires were also included in the experiment. Boys 162 18 194 152 The first was a general information scale relating to sexual inter- Girls 162 18 194 152 est and instructed participants to select one or more of five appli- No-person 162 18 194 152 cable statements (‘‘no sexual interest in adults,’’‘‘strong sexual interest in female adults,’’‘‘some sexual interest in female adults,’’ participants completed the Interest in Child Molestation Scale to ‘‘some sexual interest in male adults,’’‘‘strong sexual interest in ensure that they were solely sexually interested in adults (Gannon male adults’’). This was included to confirm the sexual interests & O’Connor, 2011). This scale consists of five short scenarios that participants reported in the pre-screen. In addition, all 123 Arch Sex Behav (2016) 45:855–870 859 Fig. 2 Example stimuli of the original quality, high quality, and the luminance-controlled image conditions in Experiment 1 and the scrambled images in Experiment 2 that describe incidents of child molestation. In response to these were being recorded. Participants were kept naı ¨ve to the full pur- scenarios, participants have to rate their arousal, enjoyment, and pose of the experiment until the end. To fully understand obser- behavioral propensity to child sex abuse on 7-point Likert scales. vers’ natural interests in these scenes, a free-viewing paradigm This scale has high test–retest reliability (r = .94) and its sexual was used so as not to constrain spontaneous eye movement pat- arousal subscale correlates with the Implicit Association Test, terns. Thus, participants were instructed simply to‘‘view the which provides an indirect measure of child sexualization asso- scenes as you naturally would’’(for similar approaches, see Bin- ciations (see Gannon & O’Connor, 2011). demann, Scheepers, & Burton, 2009;Frombergeretal., 2012a, b, 2013;Halletal., 2011; Hewig et al., 2008; Lykins et al., 2008; Eye-Tracking Nummenmaa et al., 2012). Participants were seated in a quiet and windowless room The stimuli were displayed using SR-Research Experiment with consistent artificial lighting and positioned approximately Builder software (version 1.1.0) on a 21 color monitor, with 60 cm from the display monitor. The participants’ left eye was ascreen resolution of10249 768 pixels. Eye movements were tracked and calibrated using the standard Eyelink procedure. tracked using an SR-Research Eyelink II head-mounted eye- To calibrate the eye tracker, observers fixated an initial series tracking system. The Eyelink II was running at a 500 Hz sam- of nine target points on the display monitor. Their accuracy was pling rate, a spatial resolution of\0.01 of visual angle, a gaze then validated against a second series of nine fixation targets. position accuracy of\0.5, and a pupil size resolution of 0.1 % Calibration was repeated if poor measurement accuracy was of diameter. The Eyelink II system works by measuring corneal indicated. In the experiment, each trial began with a central fix- reflection and dark pupil with a video-based infrared-camera eye ation dot, which allowed for drift correction. This was followed by a gray screen display for 1000 ms, and then the stimulus dis- tracker, which computes the number of camera pixels that are occluded by participants’ pupils. In this system, the diameter of play for 5000 ms, followed by another gray screen for 1000 ms. the pupil is recorded as an integer that ranges from 400 to This display duration is similar to other studies with static images 16,000 units. The device incorporates eye and head tracking (e.g., Fromberger et al., 2012a, b, 2013; Hewig et al., 2008;Num- that automatically compensates for minor head movements. Dur- menmaa et al., 2012) and allows for approximately 15 fixations ing the recording of eye movements, participants are instructed to (based on an average fixation duration lasting 200–300 ms, see remain seated still but further immobilization (e.g., a chinrest) is Rayner, 1998), which is sufficient time to scan the entire scene. not required. This eye-tracking system is compatible with most Each participant viewed all 75 stimuli. These were pre- glasses and contact lenses. sented in a randomized order that was uniquely generated for each participant by the EyeLink software. Short breaks were Procedure inserted every 25 trials, after which the calibration procedure was repeated. On completion of the eye-tracking task, partic- Participants were invited to take part in an experiment on sex- ipants answered the general information scale relating to their ual interest and informed that they would be viewing images sexual interests and the Interest in Child Molestation Proclivity of males and females of different ages while their eye movements scale (see Gannon & O’Connor, 2011). 123 860 Arch Sex Behav (2016) 45:855–870 Results from the overall mean was then computed, using the formula: (mean pupil diameter for category 9 100)/overall pupil mean. Confirmation of Sexual Interests Accordingly, a score of 100 % indicates that the pupillary response to a stimulus category does not differ from the overall To ensure that participants were not sexually interested in chil- mean. Scores higher or lower than this value indicate compar- dren, responses on the Interest in Child Molestation Scale were atively larger or smaller pupil sizes (for similar approaches, see analyzed first. An overall interest score was calculated for each Dabbs, 1997; Laeng & Falkenberg, 2007). To simplify the expres- participant by combining responses across all subscales (i.e., sion of these patterns, these scores were then deducted from 100 arousal, enjoyment, behavioral propensity) (for similar analysis, so that no change in pupil size is indicated by zero and positive or see Gannon & O’Connor, 2011). This produced a total score negative scores reflect relatively larger (dilation) or smaller (con- where a minimum of 15 (low sexual interest in children) and a striction) pupil sizes in response to a stimulus category. maximum score of 105 (high sexual interest in children) are possible. The results here converge with those obtained in pre- Viewing Behavior vious studies with a sample of non-offending community males (Gannon & O’Connor, 2011), such that male observers scored a We first examined the viewing patterns that the persons in the mean of 18.1 (mode = 15,SD = 5.6, min = 15, max = 30) and scenes elicited in male and female observers. To examine this, 16.8 for female observers (mode = 15, SD = 5.6, min = 15, the percentage fixations to the ROIs were calculated for all stim- max = 41). However, an established cut-off point for this scale ulus categories (see Fig. 3). Overall, 63 % of fixations fell on the does not exist. We adopted a simple metric by considering only figures in the scenes (range 58 to 71 % across conditions), which individuals with scores on the lowest third of the scale (i.e., with indicates that the person content of the scenes was of most inter- scores between 15 and 45). All participants fell within this range. est. A 4 (category: men, women, boys, girls) 9 3(ROI: head, Sexual orientation was confirmed with the general informa- body, background) 9 2 (observer sex: male, female) mixed- tion scale that was administered following the eye-tracking task factor ANOVA revealed a three-way interaction, F(6, 252) = (see‘‘Materials’’section). In the 22 male observers, 19 reported 8.01, p\0.001, partial g = 0.16. To explore this interaction, ‘‘strong sexual interest in women’’ and three selected ‘‘some sex- two separate 4 (category: men, women, boys, girls) 9 3 (ROI: ual interest in women.’’Among the 22 females, 12 selected‘‘strong head, body, background) within-subjects ANOVAs were per- sexual interest in males’’ and 10 selected ‘‘some sexual interest in formed for male and female observers. males.’’ Participants reported no other sexual interests in this For male observers, this analysis showed no main effect of questionnaire. category,F(3, 63)= 0.32,p= 0.81, partialg = 0.02, but revealed a main effect of ROI, F(2, 42)= 4.54, p\0.05, partial g = 0.18, Data Analysis and an interaction between both factors, F(6, 126)= 34.22, p\ 0.001, partial g = 0.62. To explore this interaction, Bonferroni- For the analysis of the eye-tracking data, all eye movements adjusted pairwise comparisons of the stimulus categories were were pre-processed by merging fixations of less than 80 ms with conducted for each ROI. These comparisons show that more the preceding or following fixation if it fell within half a degree fixations were directed at the background of scenes containing of visual angle (for similar approaches, see e.g., Attard & Bin- boys, girls, and men (39 to 42 %) than scenes depicting women demann, 2013; Bindemann et al., 2009; Bindemann, Scheepers, (30 %), all ps\0.01. In addition, boys (31 %) and girls (32 %) Ferguson, & Burton, 2010). In addition, any fixations that fell received more fixations to the head than men (27 %) and women outside the dimensions of the display monitor or that were (22 %), all ps\0.01, and men’s heads were also fixated more obscured by blinking were excluded. To analyze attention to frequently than those of women, p\0.01. By contrast, male specific areas within the visual scenes, each image was then coded observers directed more fixations to the bodies (48 %) of to define three regions of interest (ROIs), which comprised the women and men (34 %) than those of boys (27 %) and girls head and body of the persons and the scene background. The (26 %), all ps\0.001, and more at women’s bodies than those mean percentage of fixations that fell on these ROIs was then of men, p\0.001. None of the other comparisons reached sig- calculated across observer groups (males, females) and stimulus nificance, all ps C 0.10. categories (men, women, boys, girls). The equivalent analysis for female observers showed no For the measure of main interest, observers’ pupillary main effect of category, F(3, 63) = 0.16, p = 0.92, partial g = responses were computed by taking the mean pupil diameter 0.008, but a main effect of ROI,F(2, 42) = 2.58,p\0.001, par- at each fixation, averaged across the duration of a stimulus tial g = 0.11, and an interaction between factors, F(6, 126)= display. These values were then used to compute an overall 8.45,p\0.001, partialg = 0.29. Bonferroni-adjusted pairwise mean, across all stimuli, for each participant. The percentage comparisons of the stimulus categories show that more fix- difference (i.e., an increase or decrease) in pupil diameter for each ations landed on the head region of boys and girls (both 34 %) stimulus category (men, women, boys, girls, no-person scenes) than women (22 %) and men (29 %), all ps\0.001, and on the 123 Arch Sex Behav (2016) 45:855–870 861 Fig. 3 Mean percentage fixations to the head and body of the target persons and the scene background for male and female observers in Experiment 1. Lines represent standard errors of the means heads of men than women, p\0.001. By contrast, more fix- However, an interaction between image quality and observer ations landed on women’s bodies (40 %) compared to boys sex was found,F(2, 84)= 3.36,p\0.05, partialg = 0.07. Bon- (29 %) and girls (31 %), both ps\0.01. No other comparisons ferroni-adjusted pairwise comparisons revealed only that female reached significance, all ps C 0.08. observers exhibited larger pupils than male observers during the Overall, this pattern suggests a clear interest, whereby hetero- viewing of luminance-controlled scenes, p\0.05. No other dif- sexual males and females fixate men and women more fre- ferences were significant, all ps C 0.09. An interaction between quently than children, but are particular biased towards the image quality and category was also found, F(8, 336) = 2.17, bodies of adult female targets. p\0.05, partial g = 0.05, as the no-person beach scenes eli- cited smaller pupils in the luminance-controlled than the high Pupillary Responses quality, p\0.01, and original quality conditions, p\0.05. No other differences between any of the person content scenes The measure of main interest is pupillary response, which was were found, all psC 0.16. Therefore, image quality was not ana- lyzed further. analyzed in two ways. In the first analysis, pupillary responses were compared for male and female observers across the stim- An interaction between category and observer sex was also ulus categories and image conditions. These data are illustrated present, F(4, 168) = 2.73, p\0.05, partial g = 0.06. Bonfer- in Fig. 4. A 3 (image quality: original, high, luminance-controlled) roni-adjusted pairwise comparisons revealed smaller pupils in 9 5 (category: men, women, boys, girls, no-person)9 2(observer male than female observers during the viewing of men, p\ sex: male, female) mixed-factor ANOVA revealed a main effect 0.01. Furthermore, in male observers, women elicited larger of category, F(4, 168)= 20.35, p\0.001, partial g = 0.33, but pupil sizes than men, boys, girls, and no-person scenes, all ps B not of quality, F(2, 84) = 1.75, p = 0.18, partial g = 0.04, or 0.001. For female observers, women elicited larger pupil sizes observer sex, F(1, 42) = 1.00, p = 0.32, partial g = 0.02. than boys, girls, and no-person scenes, all ps B 0.05, but not men, p = 0.26. In addition, pupil responses were larger for scenes depicting boys than girls, p\0.05. No other differences were observed, allpsC 0.06, and an interaction between the three factors was not found, F(8, 336)= 1.10, p= 0.36, partial g = 0.03. Overall, these results therefore reveal a dilation response in male observers that appears to be consistent with self-reported sex and age preferences. Female observers’ responses are also consistent with their age preferences, but do not correspond with their reported sexual interest in adult men. In the second analysis, this pattern is confirmed when pupi- llary responses are compared via one-samplet-tests (with alpha corrected at p\0.01 for multiple comparisons) with a baseline that reflects the mean pupil diameter across all stimuli (see‘‘Data Analysis’’section). This analysis shows that the pupils of male observers were larger than baseline during the viewing of women,t(21)= 5.43, p\0.001,d= 2.37, and smaller during the viewing of men, t(21)=-3.02, p= 0.006, d= 1.32, and girls, Fig. 4 Percentage pupillary change for all stimulus categories for male t(21)=-3.1, p = 0.005, d = 1.35. In addition, pupil size was and female observers in Experiment 1. Lines represent standard errors of unchanged from baseline in response to boys and no-person the means 123 862 Arch Sex Behav (2016) 45:855–870 scenes, both tsB-1.59, ps C 0.126, ds B 0.69. In female elicited a clear pupillary dilation that was not present during observers, pictures of men,t(21) = 1.49, p = 0.15, d = 0.65, the viewing of men and children. In female observers, pupil boys, t(21)=-0.12, p = 0.91, d = 0.05, and landscape beach dilation was also greatest when pictures of women were viewed. scenes (-1.53 %), t(21)=-2.19, p = 0.04, d = 0.96 did not In these participants, pupillary recordings therefore do not cor- elicit a change in pupil size from baseline. The pupils were respond to their self-reported sexual orientation. However, these enlarged to scenes with women, t(21) = 4.71, p\0.001, d = responses still appeared to be age-specific as the pupils remained 2.06, and smaller than baseline during the viewing of girls, unchanged or constricted during the viewing of children. t(21)=-4.33, p\0.001, d = 1.89. These results converge with a recent study that has shown a similar pattern of pupillary responses for heterosexual adult Individual Differences in Pupillary Responses males and females (Rieger & Savin-Williams, 2012). Exper- iment 1 extends these findings by demonstrating that such pupil- We also sought to explore whether pupillary responses can be lary responses are also age-specific. A question that arises, how- informative about the sexual interests of individual observers. ever, is whether these dilation effects could be attributed to a low- For this purpose, the difference in raw pupil size for specific level factor such as luminance. To explore this possibility, we image comparisons (e.g., scenes with men vs. women) was cal- also compared scene photographs in which contrast and color culated separately for each participant. These data show, for were enhanced with a set in which luminance and contrast were example, that all of the male observers (22/22) recorded larger equated. The results for these stimulus categories were highly pupil sizes during the viewing of women than men, and 91 % comparable, which suggests that pupillary responses for the dif- (20/22) of male observers displayed larger pupils in response to ferent person categories cannot be explained by general variation in luminance. women than girls. In addition, only 22 % (5/22) of these partic- ipants showed a greater pupillary response to men than boys. There is, however, a problem with the luminance adjust- With regard to female observers, 73 % (16/22) showed more ment that was employed in Experiment 1. While this manip- pupil dilation during the viewing of women than men. However, ulation was used to equate luminance across scenes, it does not 86 % (19/22) of this participant group also exhibited larger pupils control other low-level image aspects, such as color, which might in response to women than girls, and 59 % (13/22) recorded also affect pupillary responses (Kohn & Clynes, 1996; Lobato- larger pupils to men than boys. Rinco ´netal., 2014). Such information was not matched across stimulus categories in Experiment 1. Consequently, the possibil- Discussion ity remains that the results might reflect such image artifacts. A second explanation is also possible for the observed pupil- The purpose of this experiment was to explore whether pupil- lary responses. While we adjusted the mean luminance of the lary responses to the visual presentation of men and women scenes, we did not measure the sexual attractiveness of the tar- can provide an indication of a person’s sexual interests. More get figures. As a result, this might have been mismatched across specifically, we sought to determine whether this approach can categories. Considering that photographs of women elicited be extended to reveal age-specific sexual interests. We first looked more pupil dilation in both male and female observers, it is at fixation patterns on the person content in scenes. Male obser- conceivable, for example, that these pictures were generally vers showed a viewing preference for women over men and more sexually arousing than those of men. To investigate these children, which was characterized by a high number of fixa- possibilities, a second experiment was conducted. tions on women’s bodies. These results are consistent with other studies, which have shown that heterosexual male observers attend more to images of the opposite sex (Lykins, Meana, Experiment 2 &Strauss, 2006;Lykinsetal., 2008;Rupp& Wallen, 2007; Suschinsky et al., 2007) and that such preferential viewing In Experiment 2, a new condition was created, in which the behavior is also age-specific (Fromberger et al., 2012a, b, 2013; pixels of the luminance-controlled images were randomized. Hall et al., 2011). Female observers also recorded fewer fixa- These scrambled images are no longer recognizable as the orig- tions on the faces of women than men and children, but more on inal scenes but provide the same color content. If the pupillary women’s bodies than those of children. Consistent with previ- responses in Experiment 1 reflect a low-level color artifact, then ous research, heterosexual females therefore showed age-speci- the same pattern should persist with these scrambled scenes in fic viewing patterns but did not exhibit the same strong visual Experiment 2. The experiment also examined whether the pic- preferences to opposite-sex figures as men (Hall et al., 2011; Israel tures of men and women in Experiment 1 were matched in terms &Strassberg, 2009; Lykins et al., 2008; Rupp & Wallen, 2007). of their perceived attractiveness. For this purpose, two measures The data of main interest were the pupillary responses. In of attractiveness were employed. The first measured general heterosexual male observers, these responses were consistent sexual appeal and recorded how attractive observers thought the stimuli were to others (i.e., sexual appeal by‘‘societal standards’’; with their reported sexual interests. Thus, pictures of women 123 Arch Sex Behav (2016) 45:855–870 863 for similar approaches, see Lippa et al., 2010). The second mea- et al., 2010). For all four tasks, the stimulus sequence in each sure concerned the sexual appeal that these images personally block was generated randomly by the display software for each hold for the individual observer (see Ebsworth & Lalumiere, participant. As in Experiment 1, participants completed the same 2012; Hewig et al., 2008). If the pupillary responses in Exper- general information scale and the Interest in Child Molestation iment 1 reflect sexual arousal, then personal sexual appeal rat- proclivity scale on completion of the eye-tracking tasks. ings should correlate with pupillary responses in Experiment 2. Results Method Confirmation of Sexual Interests Participants Once again, the responses on the Interest in Child Molestation A total of 41 students (21 male) from the University of Kent Scale were analyzed first. One of the male participants pro- participated in this study in return for a small payment or course duced a score of 52. This is the only score that falls above the credits. The mean age was 19.5 years (SD = 2.0; range 18–31 lowest third (i.e., 45) of the Child Molestation Scale in Exper- years). All participants reported to be exclusively heterosexual iment 1 and 2. It also exceeds the mean score (41.4) of ped- on the Kinsey scale (Kinsey et al., 1948, 1953), which was com- ophiles that have self-reported sexual acts with children (Mitchell pleted as a pre-screen on our online recruitment system. None of & Galupo, 2015). This individual was therefore excluded from the participants had taken part in thefirstexperiment. Allreported further analysis. For the remaining participants, means of 20.8 normal or corrected-to-normal vision. (mode = 15,SD = 6.2, min = 15, max = 34) and 16.3 (mode = 15, SD = 2.4, min = 15, max = 23) were obtained for male and Materials female observers, respectively. To confirm that participants showed a sexual interest towards This experiment employed the same eye-tracking set-up with the opposite sex, their responses on the sexual interests’ ques- the luminance-controlled stimuli from Experiment 1. To assess tionnaire were also analyzed. Nineteen of the 20 males reported the contribution of color within each of these 25 images (com- ‘‘strong sexual interest in women’’ and one reported ‘‘some prising five men, women, boys, girls, and no-person scenes) to sexual interest in women.’’ For the females, 14 of 20 reported pupillary response, the pixels in each image were randomized. ‘‘strong sexual interest in males,’’while the remaining six par- The resulting images provide a‘‘scrambled’’condition in which ticipants reported‘‘some sexual interest in males.’’Participants the original image content is not discernible (see Fig. 2;for sim- reported no other sexual interests in this questionnaire. ilar approaches, see Jenkins, Lavie, & Driver, 2003; VanRullen, 2006). Data Analysis Procedure The eye-tracking data were processed as in Experiment 1. Note that pupillary responses are reported for both free-viewing tasks The experiment consisted of four blocks. In the first block, (Block 1 and 2) but not for the two ratings tasks. In the latter participants were shown the 25 scrambled scene images. This tasks, 5.9 (SD = 3.7) and 6.5 (SD = 4.3) fixations were recor- was followed, in the second block, by the 25 unscrambled ver- ded on average per trial but the mean number of fixations varied sions of these stimuli. Both blocks were free-viewing tasks. greatly across observers (from 1 to 38). Consequently, these Each trial therefore consisted of a drift correction, which was tasks did not provide reliable eye movement data for analysis. followed by a gray mask for 1000 ms. The scrambled/intact scene The eye fixations for the free-viewing task with the intact scenes stimuli were then presented for 5000 ms, followed by the gray (Block 2) were also analyzed and revealed a similar pattern to Experiment 1. For brevity, this analysis is not reported here but mask for a further 1000 ms. In both blocks, participants were simply instructed to view these images naturally. is available on request. These data are not meaningful for the In the remaining blocks, the intact scenes with the men (5 scrambled scene images in Block 1 and are therefore also omitted. images), women (5 images), and children (5 images each) from Block 2 were repeated. In Block 3, participants were asked to Pupillary Responses provide personal sexual attractiveness ratings for these people (i.e., based on how sexually attractive they themselves find these The data of main interest were the pupillary responses. As in images) using a Likert scale ranging from 1 (‘‘not at all sexually Experiment 1, the mean percentage change in pupil size was appealing to me’’) to 7 (‘‘extremely sexually appealing to me’’). calculated for male and female observers for the person cat- In Block 4, participants were then asked to evaluate the people in egories (see Fig. 5) and was analyzed in two ways. First, a 5 the scenes based on their sexual attractiveness by societal stan- (category: men, women, boys, girls, no-person) 9 2 (observer dards using the same scale (for similar methods, see, e.g., Lippa sex: male, female) mixed-factor ANOVA showed a main 123 864 Arch Sex Behav (2016) 45:855–870 Fig. 5 Percentage pupillary change for all stimulus categories for male and female observers in Experiment 2 for intact scenes (left graph) and scrambled scenes (right graph). Lines represent standard errors of the means effect of category, F(4, 152) = 32.16, p\0.001, partial g = stimulus categories in raw pupil diameter during the free-view- 0.46. Post hoc analysis revealed overall larger pupils during ing task (Block 2). These data show that 80 % (16/20) of the male the viewing of women compared to all other categories, all psB participants displayed larger pupils when viewing women than 0.001, and larger pupils to men than boys, girls and no-person men, 95 % (19/20) displayed larger pupils to women than girls, scenes, all ps B 0.01. No other differences were found, all ps C and 85 % (17/20) displayed larger pupils to men than boys. Of 0.34. A main effect of observer sex, F(1, 38) = 0.05, p = 0.82, the female observers, 65 % (13/20) recorded larger pupils to partial g = 0.001, and an interaction between factors, F(4, women than men, 90 % (18/20) displayed larger pupils to women 152)= 2.01, p = 0.96, partial g = 0.05, was not found. than girls, and 90 % (18/20) displayed larger pupils to men than For completeness, these responses were also analyzed with boys. one-samplet-tests (withalpha corrected atp\0.01 for multiple comparisons), by comparing the change in pupil size for each Personal Sexual Appeal Ratings stimulus category with a baseline of zero (see ‘‘Data Analysis’’ section). For male observers, this analysis revealed pupil dilation In the next step of the analysis, we explored the extent to which during the viewing of women, t(19)= 7.58, p\0.001, d= 3.48, personal sexual appeal judgements of the persons in the scenes and pupil constriction during the viewing of boys,t(19)=-4.40, relate to pupil responses in the free-viewing task. For this pur- p\0.001, d = 2.02 and no-person scenes, t(19)=-4.62, p\ pose, the mean sexual appeal ratings for each of the person cat- 0.001, d = 2.12. A change in pupil size was not detected in egories were analyzed first. A 4 (category: men, women, boys, response to images of men, t(19)= 1.26, p = 0.22, d= 0.58 and girls)9 2 (observer sex: male and female) mixed-factor ANOVA girls, t(19)=-1.23, p = 0.24, d = 0.56. of these data did not show a main effect of observer sex, F(1, In female observers, dilation was also observed in response 38)= 0.02,p= 0.88, partialg = 0.00, but revealed a main effect to pictures of women, t(19) = 7.25, p\0.001, d = 3.33. How- of category, F(3, 114)= 83.26, p\0.001, partial g = 0.69, and ever, in this case, dilation was also found for pictures of men, an interaction between factors, F(3, 114) = 87.53, p\0.001, t(19)= 3.30,p= 0.004,d= 1.51. In contrast, the pupils appeared partialg = 0.70. Bonferroni-corrected post hoc comparisons to be smaller than baseline during the viewing of boys, t(19)= showed that male observers rated women as more sexually -2.65, p= 0.02, d= 1.22, girls, t(19)=-2.05, p= 0.05, d= appealing (M = 5.4, SD = 0.9) than men (M = 1.6, SD = 0.8), 0.94, and the no-person scenes, t(19)=-2.25, p = 0.04, d = boys (M = 1.2, SD = 0.8), and girls (M = 1.2, SD = 0.7), all 1.03, but these changes were not significantly below zero (with ps\0.001. In contrast, female observers rated men as more alpha corrected at p\0.01 for multiple comparisons). sexually appealing (M = 4.3, SD = 1.40) than women (M = In summary, this analysis shows that male observers’ pupils 2.1, SD = 1.2), boys (M = 1.3, SD = 0.9), and girls (M = 1.5, dilate in response to pictures of women but not men or children. SD = 1.3), all ps\0.001. No other differences were found. Female observers show a dilation response to both men and Overall, these sexual appeal ratings therefore converge clearly women, but not to children. These results therefore replicate the with observer’s self-reported sexual interest in adults of the oppo- sex-specific effect in male observers and the age-specific pattern site sex. that was observed in male and female observers in Experiment 1. We next performed a correlation between the mean pupil- lary change (%) in the free-viewing task (Block 2) and the sex- Individual Differences in Pupillary Responses ual appeal ratings. This analysis combined the person categories As in Experiment 1, we also performed a simple analysis of When this analysis was performed within category groups, no individual performance, based on the differences between correlations between pupillary response and appeal ratings were found, 123 Arch Sex Behav (2016) 45:855–870 865 (men, women, boys, girls) but was performed separately for male Scrambled Scenes and female observers. The distribution of observers’ sexual appeal ratings was skewed. Therefore, non-parametric Spear- The pupillary responses to scrambled scenes were analyzed man’s correlations are reported. For male observers, a strong pos- next. As in the analysis of intact scenes, the mean pupillary responses for each category (men, women, boys, girls, no-per- itive correlation between pupil change and sexual appeal ratings was found, r (78) = 0.64, p\0.001. This correlation also per- son scenes) were transformed to measure mean percentage change (see Fig. 5). A 5 (category: men, women, boys, girls, no- sisted when only the adult targets (men and women) were con- sidered, r (38) = 0.58, p\0.001, which suggests that it reflects person) 9 2 (observer sex: male, female) mixed-factor ANOVA did not show a main effect of observer sex, F(1, 38) = 0.00, p= observers’ sexual interests in specific adults. For female obser- vers, the correlation across all person categories (men, women, 1.00, partial g = 0.001, or an interaction between factors, F(4, boys, girls) was weaker, r (78) = 0.28, p\0.01, and was not 152) = 0.97, p = 0.43, partial g = 0.03, but revealed a main reliable when the child categories were excluded from analysis, effect of category, F(4, 152)= 4.34, p\0.01, partial g = 0.10. r (38)=-0.22, p = 0.17. Overall, these data therefore suggest Post hoc Bonferroni comparisons showed that observers’ pupils that pupillary responses provide a good index of sexual interest were smaller while viewing scrambled images of boys than in male, but not female, observers. those of women, p\0.01, and no-person scenes, p\0.01. No other differences between categories were found, all psC 0.20. General Sexual Attractiveness Ratings Once again, these responses were also analyzed via a series of one-sample t-tests (with alpha corrected at p\0.01) to com- In Block 4, the subjects were asked to objectively rate the per- pare the change in pupil size to a baseline of zero (see ‘‘Data Analysis’’section). This analysis showed no change in pupil sons in the scenes on their sexual attractiveness based on how they thought the general population would respond. The mean size across categories in male observers, all ts B 2.23, ps C ratings were analyzed with a 4 (category: men, women, boys, 0.04, ds B 1.02. The pupils of female observers were smaller girls) by 2 (observer sex: male and female) ANOVA. This during the viewing of scrambled scenes of boys, t(19) = 3.46, analysis did not show a main effect of observer sex, F(1, 38) = p\0.01, but no other differences were found, all tsB 1.83, psC 0.45, p = 0.51, partialg = 0.01, but a main effect of category, 0.08,ds B 1.59. We also correlated pupil sizes for scrambled and F(3, 114) = 331.15, p\0.001, partial g = .90, and an interac- intact scenes. This revealed no relationship between these con- ditions in male and female observers, r(98) = 0.06, p = 0.58 and tion between factors, F(3, 114) = 2.96, p\0.035, partial g = 0.07. Bonferroni-corrected post hoc comparisons revealed that r(98)= 0.04, p = 0.72, respectively. These results therefore indicate that pupillary responses to intact scenes do not reflect male observers rated the women in scenes (M = 6.0, SD = 0.6) higher on sexual attractiveness than men (M = 4.8,SD = 1.02), low-level image artifacts, such as color. p\0.001. Both adult categories were also rated higher than boys (M = 1.4, SD = 0.9) and girls (M = 1.4, SD = 0.9), all ps\ Discussion 0.001. Female observers rated men (M = 5.6, SD = 1.0) and women (M = 5.7, SD = 1.1) more similarly (p = 1.00), and This experiment assessed further whether observers’ pupillary more sexually attractive than boys (M = 1.4, SD = 1.0) and responses reflect their sexual interest in a seen stimulus. For girls (M = 1.5, SD = 1.2), both ps\0.001. No other differences this purpose, we compared pupillary responses to pictures of were observed. men and women with personal sexual appeal ratings and gen- eral attractiveness ratings (by societal standards). The pupils of A non-parametric Spearman’s correlational analysis between these ratings and observers’ pupillary responses (% change), male observers dilated to pictures of women but not men or chil- dren. Female observers showed pupillary dilation to pictures of which combined the data from all person categories (men, women, boys, girls), revealed a correlation for male and female observers, women and men but not to children. This experiment therefore r (78) = 0.62, p\0.001 and r (78) = 0.55, p\0.001, respec- replicates the age-specific dilation effects in male and female s s tively. Similar to the previous analysis, we performed a second observers that were shown in Experiment 1, and also the sex- correlation for which the data for child targets were excluded. specific dilation effect in males. This correlation was not significant in male, r (38) = 0.29, p = The personal sexual appeal ratings support the notion that 0.08, or female observers, r (38) = 0.07, p = 0.67. these pupillary responses reflect the sexual interests of hetero- sexual male observers (Rieger et al., 2015; Rieger & Savin- Williams, 2012). For example, these observers rated the pho- tographs of women as much more sexually attractive than those Footnote 1 continued of men and children, and these ratings correlated strongly with all ps C 0.06. We attribute this to the low number of images in each pupillary responses. This was evident when data from all person stimulus category (five) and the low variance in sexual appeal ratings categories were combined, but also when the children were omit- within categories. For example, male observers’ mean sexual appeal ted from the analysis. This suggests that the pupillary responses rating for female figures was 5.36 with a standard deviation of only 0.89. 123 866 Arch Sex Behav (2016) 45:855–870 of male observers reflect the sexual interest that is triggered by the presentation of women but not during the viewing of men stimuli. and children. This suggests that these pupillary responses are In line with their reported sexual orientation, heterosexual linked to the sexual interest of these observers (i.e., females) female observers rated male targets as most sexually appeal- and are also age-specific (adults). In contrast, the pupils of hetero- ing, while women and children received low ratings. These sexual female observers dilated to images of women and men, ratings diverge from their pupillary responses, which indicate but not to children. In these observers, pupillary responses there- dilation to pictures of men and women. In addition, a correla- fore appear to be age-specific but do not correspond to self- tion between sexual appeal ratings and pupillary responses was reported gender interests. found, but this did not hold when child categories were exclu- In light of these different effects in male and female obser- ded from analysis. This pattern deviates from our findings with vers, a further experiment was conducted to explore more directly heterosexual male observers. It is interesting to note, however, whether pupillary responses are linked to observers’ sexual inter- that such discrepancies were also obtained for pupil dilation and est. For this purpose, we recorded pupillary responses to male and subjective arousal in a recent experiment (Rieger et al., 2015) female adults and children and also asked observers to rate these and are commonly observed in studies comparing self-reported target persons in terms of their sexual attractiveness. Two mea- and physiological measures of sexual arousal in heterosexual sures were utilized for this purpose, which sought to capture the women (Rieger et al., 2015; Suschinsky & Lalumiere, 2012; sexual attractiveness that these stimuli personally held for an Suschinsky et al., 2009; for a meta-analysis, see Chivers, Seto, observer as well as their general sexual attractiveness to others. Lalumiere, Laan, & Grimbos, 2010). The pupillary responses in this experiment replicated the sex- We also investigated whether the pupillary responses of and age-specific effect in male observers and the age-specific male and female observers might reflect differences in the gen- effect in female observers. This suggests, once again, that pupil- eralattractiveness of the stimulus categories, by measuring how lary response can provide a measure of sexual interest for male sexually attractive observers thought the stimuli were to others. but not female observers. Male observers rated children and adult males as less generally These findings received further support from the ratings tasks. attractive than adult females. However, the difference between The relationship between personal sexual appeal ratings and male and female stimuli was smaller than for the personal appeal pupillary responses was weak for females and driven by the ratings, indicating some adjustment. This difference was smal- age of the persons in the scenes. However, the ratings of male ler still in female observers, who perceived men and women to observers showed a clear preference for adult females and cor- be of similar general sexual attractiveness. Moreover, while the related well with pupillary response, which suggests that it reflects general attractiveness ratings correlated with pupillary respon- the sexual interests of the males in this study. By contrast, male ses, this did not hold for male or female observers when the child and female observers perceived the general sexual attractive- categories were excluded from analysis. This suggests that the ness of men and women to be more comparable and these rat- general sexual attractiveness of male and female adult stimuli ings did not correlate with pupillary response. Taken together, was not grossly mismatched in the current experiments, or that these findings suggest that pupillary responses reflect the per- this was the key determinant of pupillary responses. sonal sexual interests of male but not female observers, but are We also explored whether the pupillary pattern could arise age-specific in both groups. from low-level artifacts within the scene images (Kohn & Cly- The responses of male observers to images of women con- nes, 1969; Lobato-Rinco ´netal., 2014). To investigate this pos- verge with previous research, which has also shown an increase sibility, a control condition of scrambled images was included, in pupil size to such content (Hess et al., 1965; Rieger & Savin- which are no longer recognizable as coherent scenes but retain Williams, 2012;Riegeretal., 2015). Female observers recorded their color content. These scrambled scenes failed to produce pupil dilation in response to images of men in Experiment 2 but pupillary dilation that corresponds with responses to the intact also displayed larger pupils for images of women across both scenes. These findings therefore converge with the sexual appeal experiments. The reason for this is unclear. However, this absence and attractiveness ratings to indicate that the pupillary responses of sex-specific pupillary responses for female observers is also in this study are driven by the person content of the scenes. consistent with other paradigms in this field, such as viewing time studies (Israel & Strassberg, 2009; Lippa et al., 2010), as well as self-reports and physiological arousal (Chivers et al., General Discussion 2004, 2010; Steinman, Wincze, Sakheim, Barlow, & Mavis- sakalian, 1981; Suschinsky et al., 2009). For example, in these The study examined whether pupillary responses to photographs studies, women frequently show increased physiological arou- of people can provide an indication of an observer’s sexual inter- sal to images of both sexes (e.g., Chivers et al., 2004;Wincze& ests. We specifically sought to determine whether such responses Qualls, 1984) and weaker correlations than men with self-reported are sensitive to the age of targets. Experiment 1 showed that preference and sexual arousal (Chivers et al., 2004; Schmidt, pupils of heterosexual male observers dilated during the 1975). These findings indicate that women’s sexual interests 123 Arch Sex Behav (2016) 45:855–870 867 are organized differently to those of men (Lippa, 2006, 2007; still exists that observers can manipulate such responses Suschinsky et al., 2009) and may not be as strongly linked to voluntarily by avoiding person content in the visual field (Bin- arousal patterns (for a review, see Chivers, 2005). The current demann et al., 2007), or by causing pupil constriction through experiments suggest that this also applies to pupillary responses. focusing on high-luminance scene regions. Considering that par- It is noteworthy that our pupillary responses in males and ticipants in this study were naı ¨ve to the full purpose of the exper- females are also consistent with a small set of studies from the iment until the end, it is unlikely that such methods were adopted 1960s, which first assessed pupil dilation with an elementary to exert top-down control on pupillary responses. Nonetheless, video-frame analysis (Hess et al., 1965; Scott et al., 1967), and this is clearly another important avenue for further investigation. a recent study that verified these findings with contemporary We have also only been able to demonstrate pupillary eye-tracking equipment (Rieger & Savin-Williams, 2012). responses with male adult observers who are sexually inter- The current experiments extend this recent work by demonstrat- ested in other adults but not in children. We therefore acknowl- ing that such pupillary responses are also age-specific, whereby edge that further work with a pedophilic population and con- the pupils of non-pedophilic observers dilate to pictures of adults temporary eye-tracking equipment is required to determine fully but not children. This age-specific effect represents, in fact, the whether pupillary responses can detect such inappropriate sex- most consistent aspect of our results. ual interests. In future research, it would also be valuable to com- This is an important finding that raises the possibility that pare pupil dilation directly with other existing measures of deviant pupillary response could be used as a measure of deviant sex- sexual interest, such as Implicit Association Tests (Babchishin, ual interest in children in the assessment and rehabilitation of Nunes, & Herman, 2013), Stroop Tasks (O Ciardha & Gormley, offending populations (Gannon et al., 2004; Laws & O’Donohue, 2012; Price & Hanson, 2007), and Choice Reaction Time tasks 2008). To this point, it is notable that the lack of pupil dilation by (Mokros et al., 2010;Wright & Adams, 1994). This may serve to male observers during the viewing of boys and girls is consistent strengthen the validity and assessment value of these diagnostic with an old study that compared pedophilic and non-pedophilic measures, and would also help to establish the comparative males with a more elementary approach (Atwood & Howell, strength of a pupil dilation paradigm. 1971). In that study, pupillary response appeared to provide an index of age-specific sexual interests in 77 % of individual Conclusion observers. The current study also recorded larger pupillary responses to women than men in the majority of male observers This is the first study to show with contemporary eye-tracking (100 and 80 % of participants in Experiment 1 and 2, respec- equipment that pupillary responses provide a promising method tively), and to women than girls (91 and 95 % of participants in for measuring age-specific sexual interests. We have only been Experiment 1 and 2). able to demonstrate this with male adult observers who are sex- ually interested in other adults and not in children. We therefore acknowledge that further work is required to determine fully Limitations and Directions for Future Research whether pupillary responses can detect pedophilic sexual inter- ests. However, pupil dilation appears to be a highly promising This is an exploratory study with limitations. For example, we method for assessing such deviant sexual interests. This mea- sought to increase ecological validity by using images of beach sure seems to relate directly to observers’ sexual interest in scenes, as these provide a natural setting to display semi-nude other adults and genital arousal (Rieger et al., 2015). It is also people (i.e., wearing only beachwear) to enhance sexual arousal. an autonomic response that operates outside of conscious con- However, this approach also resulted in variation of the person trol (Laeng et al., 2012; Laeng & Sulutvedt, 2014). Conse- content in terms of body posture, facial expression, eye gaze of quently, pupil dilation might provide a more robust measure of the targets, and so forth. This could have affected eye fixations deviant sexual interest than current measures, which are prone around the scenes and pupillary responses (Birmingham, Bis- to social desirable responding and participant manipulation chof, & Kingstone, 2008). This couldbeaddressedinfuture (for a review, see Kalmus & Beech, 2005). Our data also sug- studies by using more controlled stimuli. As an alternative, such gest that pupillary response could be a sensitive measure at an experiments could compare pupillary responses of hetero-, individual level. This is an important characteristic for imple- homo-, and bisexual male observers. If pupillary response pro- mentation into forensic practice (Gannon et al., 2004). Consid- vides a robust measure of sexual interest, rather than reflecting ering the potential applied value of pupillary responses as a direct other factors within natural scenes, then this should reflect the measure of age-specific sexual interest in this context, further specific sexual interests of these different observer groups. research is warranted. A small set of studies have shown that the pupils appear to be resistant to top-down control, such that observers cannot Acknowledgments This work was supported by a research grant from willingly increase or decrease their pupil size (Laeng et al., the National Organisation for the Treatment of Sexual Abusers (NOTA) to Caoilte O Ciardha and Markus Bindemann. 2012; Laeng & Sulutvedt, 2014). However, the possibility 123 868 Arch Sex Behav (2016) 45:855–870 Open Access This article is distributed under the terms of the Creative arousal in men and women: A meta-analysis. Archives of Sexual Commons Attribution 4.0 International License (http://creativecommons. Behavior, 39, 5–56. doi:10.1007/s10508-009-9556-9. org/licenses/by/4.0/), which permits unrestricted use, distribution, and Dabbs, J. M. (1997). Testosterone and pupillary response to auditory reproduction in any medium, provided you give appropriate credit to the sexual stimuli. Physiology & Behavior, 62, 909–912. doi:10.1016/ original author(s) and the source, provide a link to the Creative Commons S0031-9384(97)00268-0. license, and indicate if changes were made. Ebsworth, M., & Lalumie `re, M. L. (2012). Viewing time as a measure of bisexual sexual interest. Archives of Sexual Behavior, 41, 161–172. doi:10.1007/s10508-012-9923-9. Ellis, C. J. K. (1981). The pupillary reflex in normal subjects.BritishJournal References of Ophtalmology, 65, 754–759. doi:10.1136/bjo.65.11.754. Fromberger, P., Jordan, K., Steinkrauss, H., von Herder, J., Stolpmann, G., Kro ¨ ner-Herwig, B., & Mu ¨ ller, J. L. (2013). Eye movements in Abel, G. G., Huffman, J., Warberg, B., & Holland, C. L. (1998). Visual pedophiles: Automatic and controlled attentional processes while reaction time and plethysmography as measures of sexual interest viewing prepubescent stimuli. Journal of Abnormal Psychology, in child molesters. Sexual Abuse: A Journal of Research and Treat- 122, 587–599. doi:10.1037/a0030659. ment, 10, 81–95. doi:10.1023/A:1022063214826. Fromberger, P., Jordan, K., Steinkrauss, H., von Herder, J., Witzel, J., Akerman, G., & Beech, A. R. (2012). A systematic review of measures of Stolpmann, G.,… Mu ¨ ller, J. L. (2012a). Diagnostic accuracy of eye deviant sexual interest and arousal.Psychiatry,PsychologyandLaw, movements in assessing pedophilia. Journal of Sexual Medicine, 9, 19, 118–143. doi:10.1080/13218719.2010.547161. Attard, J., & Bindemann, M. (2013). Establishing the duration of crimes: 1868–1882. doi:10.1111/j.1743-6109.2012.02754.x. An individual differences and eye tracking investigation into time Fromberger, P., Jordan, K., von Herder, J., Steinkrauss, H., Nemetschek, estimation. Applied Cognitive Psychology, 28, 215–225. doi:10.1002/ R., Stolpmann, G., & Mu ¨ ller, J. L. (2012b). Initial orienting towards acp.2983. sexually relevant stimuli: Preliminary evidence from eye move- Atwood, R. W., & Howell, R. J. (1971). Pupillometric and personality test ment measures. Archives of Sexual Behavior, 41, 919–928. doi:10. score differences of female aggressing pedophiliacs and normals. 1007/s10508-011-9816-3. Psychonomic Science, 22,5–7.doi:10.3758/BF03332522. Gannon, T. A., & O’Connor, A. (2011). The development of the Interest Babchishin, K. M., Nunes, K. L., & Hermann, C. A. (2013). The validity in Child Molestation Scale. Sexual Abuse: A Journal of Research of Implicit Association Test (IAT) measures of sexual attraction to and Treatment, 23, 474–493. doi:10.1177/1079063211412390. children: A meta-analysis. Archives of Sexual Behavior, 42, 487– Gannon, T. A., Ward, T., & Polaschek, D. L. L. (2004). Child sexual 499. doi:10.1007/s10508-012-0022-8. offenders. In M. Connolly (Ed.), Violence in society: New Zealand Bergamin, O., & Kardon, R. H. (2003). Latency of the pupil light reflex: perspectives (pp. 31–47). Christchurch: Te Awatea Press. Gress, C. L. (2005). Viewing time measures and sexual interest: Another Sample rate, stimulus intensity, and variation in normal subjects. piece of the puzzle. Journal of Sexual Aggression, 11, 117–125. Investigative Ophthalmology & Visual Science, 44, 1546–1554. doi:10.1080/13552600500063666. doi:10.1167/iovs.02-0468. Gress, C. L., Anderson, J. O., & Laws, D. R. (2013). Delays in attentional Bernick, N., Kling, A., & Borowitz, G. (1971). Physiological differen- processing when viewing sexual imagery: The development and tiation of sexual arousal and anxiety. Psychosomatic Medicine, 33, comparison of two measures. Legal and Criminological Psychol- 341–352. doi:10.1097/00006842-197107000-00004. ogy, 18, 66–82. doi:10.1111/j.2044-8333.2011.02032.x. Bindemann, M., Burton, A. M., Langton, S. R. H., Schweinberger, S. R., Hall, C., Hogue, T., & Guo, K. (2011). Differential gaze behavior towards & Doherty, M. J. (2007). The control of attention to faces. Journal sexually preferred and non-preferred human figures. Journal of Sex of Vision, 7(10):15, 1–8. doi:10.1167/7.10.15. Research, 48, 461–469. doi:10.1080/00224499.2010.521899. Bindemann, M., Scheepers, C., & Burton, A. M. (2009). Viewpoint and Hamel, R. F. (1974). Female subjective and pupillary reaction to nude center of gravity affect eye movements to human faces. Journal of male and female figures. Journal of Psychology, 87, 171–175. Vision, 9(2):7, 1–16. doi:10.1167/9.2.7. doi:10.1080/00223980.1974.9915687. Bindemann, M., Scheepers, C., Ferguson, H. J., & Burton, A. M. (2010). Harris, G. T., Rice, M. E., Quinsey, V. L., & Chaplin, T. C. (1996). Face, body, and centre of gravity mediate person detection in natural Viewing time as a measure of sexual interest among child molesters scenes. Journal of Experimental Psychology: Human Perception and normal heterosexual men. Behaviour Research and Therapy, and Performance, 36, 1477–1485. doi:10.1037/a0019057. Birmingham, E., Bischof, W. F., & Kingstone, A. (2008). Social attention 34, 389–394. doi:10.1016/0005-7967(95)00070-4. and real-world scenes: The roles of action, competition and social Henderson, J. M. (2003). Human gaze control during real-world scene content. The Quarterly Journal of Experimental Psychology, 61, perception. TrendsinCognitive Sciences, 7, 498–504. doi:10.1016/ 986–998. doi:10.1080/17470210701410375. j.tics.2003.09.006. Bradley, M. M., Miccoli, L., Escrig, M. A., & Lang, P. J. (2008). The pupil as Hess, E. H., Seltzer, A. L., & Shlien, J. M. (1965). Pupil response of a measure of emotional arousal and autonomic activation.Psychophys- hetero- and homosexual males to pictures of men and women: A iology, 45, 602–607. doi:10.1111/j.1469-8986.2008.00654.x. pilot study. Journal of Abnormal Psychology, 70, 165–168. doi:10. Calvo, M. G., & Lang, P. J. (2004). Gaze patterns when looking at emotional 1037/h0021978. pictures: Motivationally biased attention. Motivation and Emotion, 28, Hewig, J., Trippe, R. H., Hecht, H., Straube, T., & Miltner, W. H. (2008). 221–243. Gender differences for specific body regions when looking at men Chivers, M. L. (2005). A brief review and discussion of sex differences in and women. Journal of Nonverbal Behavior, 32, 67–78. doi:10.1007/ s10919-007-0043-5. the specificity of sexual arousal. Sexual and Relationship Therapy, Isaacowitz, D. M. (2006). Motivated gaze: The view from the gazer. Current 20, 377–390. doi:10.1080/14681990500238802. Directions in Psychological Science, 15, 68–72. doi:10.1111/j.0963- Chivers, M. L., Rieger, G., Latty, E., & Bailey, J. M. (2004). A sex difference 7214.2006.00409.x. in the specificity of sexual arousal. Psychological Science, 15, 736– Israel, E., & Strassberg, D. S. (2009). Viewing time as an objective measure 744. doi:10.1111/j.0956-7976.2004.00750.x. ` of sexual interest in heterosexual men and women. Archives of Sexual Chivers, M. L., Seto, M. C., Lalumiere, M. L., Laan, E., & Grimbos, T. Behavior, 38, 551–558. doi:10.1007/s10508-007-9246-4. (2010). Agreement of self-reported and genital measures of sexual 123 Arch Sex Behav (2016) 45:855–870 869 Jenkins, R., Lavie, N., & Driver, J. (2003). Ignoring famous faces: Cate- Mustanski, B., Chivers, M. L., & Bailey, J. M. (2002). A critical review of gory-specific dilution of distractor interference. Perception and Psy- recent biological research on human sexual orientation. Annual Review chophysics, 65, 298–309. doi:10.1037/a0020733. of Sex Research, 13, 89–140. doi:10.1080/10532528.2002.10559803. ¨ ¨ Kalmus, E., & Beech, A. R. (2005). Forensic assessment of sexual inter- Nummenmaa, L., Hietanen, J. K., Santtila, P., & Hyona, J. (2012). Gen- est: A review. Aggression and Violent Behavior, 10, 193–217. der and visibility of sexual cues influence eye movements while doi:10.1016/j.avb.2003.12.002. viewing faces and bodies. Archives of Sexual Behaviour, 41, 1439– Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual behavior 1451. doi:10.1007/s10508-012-9911-0. in the human male. Philadelphia: W. B. Saunders. O Ciardha, C., & Gormley, M. (2012). Using a pictorial-modified stroop Kinsey, A. C., Pomeroy, W. B., Martin, C. E., & Gebhard, P. H. (1953). task to explore the sexual interests of sexual offenders against Sexual behavior in the human female. Philadelphia: W. B. Saunders. children. Sexual Abuse: A Journal of Research and Treatment, 24, Kohn, M., & Clynes, M. (1996). Color dynamics of the pupil. Annals of 175–197. doi:10.1177/1079063211407079. the New York Academy of Sciences, 156, 931–950. doi:10.1111/j. O Ciardha, C., & Gormley, M. (2013). Measuring sexual interest using a 1749-6632.1969.tb14024.x. pictorial modified Stroop task, a pictorial Implicit Association Test Laeng, B., & Falkenberg, L. (2007). Women’s pupillary responses to sex- and a Choice Reaction Time task. Journal ofSexual Aggression, 19, ually significant others during the hormonal cycle. Hormones and 158–170. doi:10.1080/13552600.2012.677486. Behavior, 52, 520–530. doi:10.1016/j.yhbeh.2007.07.013. Price, S. A., & Hanson, K. R. (2007). A modified Stroop task with sexual Laeng, B., Sirois, S., & Gredeba ¨ck, G. (2012). Pupillometry: A window offenders: Replication of a study. Journal of Sexual Aggression, 13, to the preconscious? Perspectives on Psychological Science, 7, 18– 203–216. doi:10.1080/13552600701785505. 27. doi:10.1177/1745691611427305. Quinsey, V. L., Ketsetzis, M., Earls, C., & Karamanoukian, A. (1996). View- Laeng, B., & Sulutvedt, U. (2014). The eye pupil adjusts to imaginary ing time as a measure of sexual interest. Ethology and Sociobiology, 17, light. Psychological Science, 25, 188–197. doi:10.1177/0956797 341–354. doi:10.1016/S0162-3095(96)00060-X. 613503556. Rayner, K. (1998). Eye movements in reading and information process- Laws, D. R., & Gress, C. L. (2004). Seeing things differently: The view- ing: 20 years of research. Psychological Bulletin, 124, 372–422. ing time alternative to penile plethysmography. Legal and Crimi- doi:10.1037/0033-2909.124.3.372. nological Psychology, 9, 183–196. doi:10.1348/1355325041719 Rieger, G., Cash, B. M., Merrill, S. M., Jones-Rounds, J., Dharmavaram, 338. S. M., & Savin-Williams, R. C. (2015). Sexual arousal: The corre- Laws, D. R., & O’Donohue, W. T. (Eds.). (2008). Sexual deviancy: Theory, spondence of eyes and genitals. Biological Psychology, 104, 56–64. assessment, and treatment (2nd ed.). New York: The Guildford Press. doi:10.1016/j.biopsycho.2014.11.009. Lippa, R. A. (2006). Is high sex drive associated with increased sexual Rieger, G., & Savin-Williams, R. C. (2012). The eyes have it: Sex and attraction to both sexes? It depends on whether you are male or sexual orientation differences in pupil dilation patterns. PLoS One, female. Psychological Science, 17, 46–52. doi:10.1111/j.1467- 7, e40256. doi:10.1371/journal.pone.0040256. 9280.2005.01663.x. Rupp, H. A., & Wallen, K. (2007). Sex differences in viewing sexual Lippa, R. A. (2007). The relation between sex drive and sexual attraction stimuli: An eye-tracking study in men and women. Hormones and to men and women: A cross-national study of heterosexual, bisexual Behavior, 51, 524–533. doi:10.1016/j.yhbeh.2007.01.008. and homosexual men and women. Archives of Sexual Behavior, 36, Rupp, H. A., & Wallen, K. (2008). Sex differences in response to visual 209–222. doi:10.1007/s10508-006-9146-z. sexual stimuli: A review. Archives of Sexual Behavior, 37, 206– Lippa, R. A. (2012). Effects of sex and sexual orientation on self- 218. doi:10.1007/s10508-007-9217-9. reported attraction and viewing times to images of men and women: Schmidt, G. (1975). Male-female differences in sexual arousal and behavior Testing for category specificity. Archives of Sexual Behavior, 41, during and after exposure to sexual explicit stimuli. Archives of Sexual 149–160. doi:10.1007/s10508-011-9898-y. Behavior, 4, 353–365. doi:10.1007/BF01541721. Lippa, R. A., Patterson, T. M., & Marelich, W. D. (2010). Looking at and Scott, T. R., Wells, W. H., Wood, D. Z., & Morgan, D. I. (1967). Pupillary longing for male and female‘‘swimsuit models’’: Men are much more response and sexual interest re-examined. Journal of Clinical Psy- category specific than women.SocialPsychologicalandPersonality chology, 23, 433–438. doi:10.1002/1097-4679(196710). Science, 1, 238–245. doi:10.1177/1948550609359814. Sell, R. L. (1997). Defining and measuring sexual orientation: A review. Lobato-Rinco ´ n, L.-L., del Cabanillas-Campos, M. C., Bonnin-Arias, C., Archives of Sexual Behaviour, 26, 643–658. doi:10.1023/A:10245 Chamorro-Gutie ´rrez, E., Murciano-Cespedosa, A., & Sa ´nchez- 28427013. Ramos Roda, C. (2014). Pupillary behavior in relation to wave- Snowden, R. J., Craig, R. L., & Gray, N. S. (2011). Indirect behavioral length and age. Frontiers in Human Neuroscience, 8, 1–8. doi:10. measures of cognition among sexual offenders. Journal of Sex 3389/fnhum.2014.00221. Research, 48, 192–217. doi:10.1080/00224499.2011.557750. Lykins, A. D., Meana, M., & Strauss, G. P. (2006). Detection of differ Steinman,D.L., Wincze,J.P., Sakheim, B. A.,Barlow, D. H.,& ential viewing patterns to erotic and non-erotic stimuli using eye- Mavissakalian, M. (1981). A comparison of male and female patterns tracking methodology. Archives of Sexual Behavior, 35, 219–228. of sexual arousal. Archives of Sexual Behavior, 10, 529–547. doi:10. doi:10.1007/s10508-006-9065-z. 1007/BF01541588. Lykins, A. D., Meana, M., & Strauss, G. P. (2008). Sex differences in Suschinsky, K. D., Elias, L. J., & Krupp, D. B. (2007). Looking for Ms. visual attention to erotic and non-erotic stimuli. Archives of Sexual Right: Allocating attention to facilitate mate choice decisions. Behavior, 37, 219–228. doi:10.1007/s10508-007-9208-x. Evolutionary Psychology, 5, 428–441. Mitchell, R. C., & Galupo, M. P. (2015). Interest in child molestation among Suschinsky, K. D., & Lalumiere, M. L. (2012). Is sexual concordance a community sample of men sexually attracted to children. Journal of related to awareness of physiological states? Archives of Sexual Sexual Aggression.doi:10.1080/13552600.2015.1056263. Behavior, 41, 199–208. doi:10.1007/s10508-012-9931-9. Mokros, A., Dombert, B., Osterheider, M., Zappala `, A., & Santtila, P. Suschinsky, K. D., Lalumie `re, M. L., & Chivers, M. L. (2009). Sex differences (2010). Assessment of pedophilic sexual interest with attentional in patterns of genital sexual arousal: Measurement artifacts or true choice reaction time task. Archives of Sexual Behavior, 39, 1081– phenomena? Archives of Sexual Behavior, 38, 559–573. doi:10.1007/ 1090. doi:10.1007/s10508-009-9530-6. s10508-008-9339-8. 123 870 Arch Sex Behav (2016) 45:855–870 Tanner, J. (1978). Fetus into man: Physical growth from conception to Wright, L. W., & Adams, H. E. (1994). Assessment of sexual preference maturity. Cambridge, MA: Harvard University Press. using a choice-reaction time task. Journal of Psychopathology and VanRullen, R. (2006). One second glance: Still no high-level pop-out Behavioral Assessment, 16, 221–231. doi:10.1007/BF02229209. effect for faces. Vision Research, 46, 3017–3027. doi:10.1016/j. Zuckerman, M. (1971). Physiological measures of sexual arousal in the visres.2005.07.009. human. Psychological Bulletin, 75, 297–329. doi:10.1037/h0030923. Wincze, J. P., & Qualls, C. B. (1984). A comparison of structural patterns of sexual arousal in male and female homosexuals. Archives of Sexual Behaviour, 13, 361–370. doi:10.1007/BF01541908. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Archives of Sexual Behavior Pubmed Central

Pupillary Response as an Age-Specific Measure of Sexual Interest

Archives of Sexual Behavior , Volume 45 – Feb 8, 2016

Loading next page...
 
/lp/pubmed-central/pupillary-response-as-an-age-specific-measure-of-sexual-interest-VSFfpDnosB

References (83)

Publisher
Pubmed Central
Copyright
© The Author(s) 2016
ISSN
0004-0002
eISSN
1573-2800
DOI
10.1007/s10508-015-0681-3
Publisher site
See Article on Publisher Site

Abstract

Arch Sex Behav (2016) 45:855–870 DOI 10.1007/s10508-015-0681-3 OR IG INA L PAP E R Pupillary Response as an Age-Specific Measure of Sexual Interest 1 1 1 • • Janice Attard-Johnson Markus Bindemann Caoilte O Ciardha Received: 4 July 2014 / Revised: 7 December 2015 / Accepted: 11 December 2015 / Published online: 8 February 2016 The Author(s) 2016. This article is published with open access at Springerlink.com. This article is published with open access at Springerlink.com Abstract In the visual processing of sexual content, pupil dila- 2002;Sell, 1997) and the assessment of unhealthy and inappro- tion is an indicator of arousal that has been linked to observers’ priate sexual desires in clinical and forensic settings (Gannon, sexual orientation. This study investigated whether this measure Ward, & Polaschek, 2004; Laws & O’Donohue, 2008). Experi- can be extended to determine age-specific sexual interest. In two mental psychology has contributed to this field by developing a experiments, the pupillary responses of heterosexual adults to number of assessment methods (e.g., Gress, 2005;Laws&Gress, images of males and females of different ages were related to 2004; Mokros, Dombert, Osterheider, Zappala `, & Santtila, 2010; self-reported sexual interest, sexual appeal to the stimuli, and a O Ciardha & Gormley, 2012, 2013). Of these, viewing time, child molestation proclivity scale. In both experiments, the pupils which reflects the duration for which particular content is studied, of male observers dilated to photographs of women but not men, is now a widely utilized measure of interest in sexually appetitive children, or neutral stimuli. These pupillary responses corre- materials (e.g., Lykins, Meana, & Strauss, 2008; Rupp & Wallen, sponded with observer’s self-reported sexual interests and their 2007). The viewing of visual content is also accompanied by sexual appeal ratings of the stimuli. Female observers showed automatic changes in observers’ pupil size (Bradley, Miccoli, pupil dilation to photographs of men and women but not chil- Escrig, & Lang, 2008), which appear to be particularly sensitive dren. In women, pupillary responses also correlated poorly with to sexual arousal (Bernick, Kling, & Borowitz, 1971). While this sexual appeal ratings of the stimuli. These experiments provide pupillary response was first explored 40 years ago with some ele- initial evidence that eye-tracking could be used as a measure of mentary methods (Hess, Seltzer, & Shlien, 1965), it has received sex-specific interest in male observers, and as an age-specific little attention since. In this study, we attempt to replicate those index in male and female observers. early findings with contemporary eye-tracking equipment to determine if it can be used to assess sexual interests. We not only Keywords Sexual interest  Eye-tracking  wish to explore whether increased pupil size can provide an index Pupillary response  Sexual appeal of adults’ sexual interest in other adults but also whether this index is age-specific. This addition might be important for clinical and forensic practice. Introduction Viewing time is a measure that is linked to a person’s inter- ests and motivations (Henderson, 2003; Isaacowitz, 2006). In The measurement of sexual arousal and observers’ sexual inter- relation to sexual interest, viewing time has been used to mea- ests is important for psychological research and practice. For sure interest in preferred over non-preferred figures. One way example, this is necessary to conduct research into sexual orien- for measuring viewing time in these paradigms is to record tation causes and consequences (Mustanski, Chivers, & Bailey, observers’ response times while they rate the sexual appeal of pictures of men and women (Gress, 2005; Gress, Anderson, & Laws, 2013; for reviews, see Akerman & Beech, 2012;Laws & & Janice Attard-Johnson Gress, 2004; Snowden, Craig, & Gray, 2011). In these studies, j.attard@kent.ac.uk longer response times for a specific stimulus type correspond to the reported sexual interest for that category (Quinsey, Ketset- School of Psychology, University of Kent, Canterbury zis, Earls, & Karamanoukian, 1996) and physiological measures CT2 7NP, UK 123 856 Arch Sex Behav (2016) 45:855–870 of sexual arousal (Abel, Huffman, Warberg, & Holland, 1998). In this study, we explore an alternative eye-tracking mea- For example, heterosexual male observers tend to make slower sure that might be more sensitive and not under top-down con- responses when rating pictures of women than of men (Israel & trol. The pupils respond automatically to external stimulation, Strassberg, 2009) and prepubescent children (Harris, Rice, Quin- such as changes in lighting conditions, by increasing (dilating) sey, & Chaplin, 1996; Quinsey et al., 1996). Female heterosexual or decreasing (constricting) in size. A similar pattern is also observers also show age preferences in these viewing time para- found as an arousal response to pleasant and unpleasant stimuli digms (Ebsworth & Lalumiere, 2012; Quinsey et al., 1996)but are (Bradley et al., 2008). This dilation has been linked to the acti- inconsistent in their responses to sexually preferred and non-pre- vation of the autonomic nervous system (Zuckerman, 1971)and ferred adults (Ebsworth & Lalumiere, 2012; Israel & Strassberg, appears to be impervious to top-down control. It has been shown, 2009; Lippa, Patterson, & Marelich, 2010; Quinsey et al., 1996). for example, that observers cannot enlarge or reduce pupil size at While the response time-based assessment of viewing time will in the absence of a visual stimulus (Laeng & Sulutvedt, 2014) is an indirect measure of sexual interest, it is possible to achieve or suppress pupil dilation (for a review, see Laeng, Sirois, & Gre- similar results more directly by tracking observers’ eye move- deba ¨ ck, 2012). These characteristics might make pupillary ments. During visual processing, eye gaze is directed towards response an ideal measure for the assessment of sexual interest. scene content that matches a viewer’s personal interest (Calvo & While this is an interesting possibility, the pupillary response Lang, 2004), including longer fixations on sexually preferred to sexual arousal has received little research attention. In an early human figures (Fromberger et al., 2012b; Hall, Hogue, & Guo, study, Hess et al. (1965) showed five hetero- and five homo- 2011; Rupp & Wallen, 2007; for a review, see Rupp & Wallen, sexual males images of nude men and women while filming 2008). Heterosexual male observers, for example, view women the observers’ eyes at a rate of two frames per second. Twenty for longer than men (Lykins et al., 2008). These viewing patterns measurements were obtained for each stimulus by manually also appear to correspond to the sexual content on display (Hall measuring pupil diameter at each frame of the video footage. et al., 2011; Rupp & Wallen, 2007; Suschinsky, Elias, & Krupp Despite this elementary approach, a clear pupillary response 2007). For example, male and female observers predominantly was found whereby all heterosexual males exhibited larger study the faces of fully clothed persons (Hewig, Trippe, Hecht, pupils to pictures of women than men. By contrast, all but one Straube, & Miltner, 2008). However, female observers increase of the homosexual males showed larger pupil responses to pic- fixations to the body in semi-clothed stimuli (Rupp & Wallen, tures of men than women. These promising results were re- 2007) and male observers show a corresponding shift to pictures examined shortly after with the addition of female observers of nude women (Nummenmaa, Hietanen, Santtila, & Hyo ¨ na ¨, (Scott, Well, Wood, & Morgan, 1967). Here, observers were pre- 2012). These data therefore indicate that eye movements are sen- sented with semi-nude and clothed images of men and women. sitive to adult observers’ sexual interest in other adults. Male observers demonstrated more pupil dilation to semi-nude Viewing patterns also appear to be age-specific. For exam- women than any other stimuli. Female observers did not show ple, male and female adult observers fixate on figures of their different pupil responses to semi-naked and clothed stimuli or preferred age (20-year olds) more than babies and 60-year olds male and female targets. However, a subsequent experiment (Hall et al., 2011). However, whereas non-paedophilic adult also recorded a pupil dilation effect in female observers that males preferentially fixate on pictures of adults over children, appeared to be related to sexual interest (Hamel, 1974). In this paedophilic males show the reverse pattern (Fromberger et al., study, female observers showed increases in pupil size that were 2012a;Frombergeret al., 2013). This indicates that eye move- directly related to the degree of nudity of pictures of male, but not ments are not only sensitive to adult observers’ sexual interest in of female, models. other adults, but can also distinguish between such interest in Despite these promising results, there have been no attempts adults and children. to replicate these findings until recently. Rieger and Savin-Wil- Despite these advantages, fixation behavior is an index of liams (2012) showed hetero-, homo-, and bisexual observers sex- sexual interest that is vulnerable to top-down control. Observers ually explicit videos, while pupillary responses were recorded could, for example, conceal their sexual interest by diverting with contemporary eye-tracking equipment. This study repli- attention to other visual content (Bindemann, Burton, Langton, cated the clear relationship between sexual orientation and Schweinberger, & Doherty, 2007). This limitation could be pupil dilation that Hess et al. (1965)had foundinmaleobservers. overcome by considering only the initial fixation to a stimulus However, similar to Scott et al. (1967), pupillary responses in display, which might reflect a covert and automatic orientation heterosexual female observers were comparable when viewing response to pre-attentively selected stimuli of sexual interest. In footage of men and women. In a subsequent experiment, Rieger line with this reasoning, heterosexual adult males tend to direct et al. (2015) extended these findings to show that pupillary more initial fixations at women than men (56 vs. 44 %) and responses to sexually explicit images reflect the sexual orien- young girls (57 vs. 43 %; see Fromberger et al., 2012b). How- tation of male observers, but not of heterosexual female obser- ever, the difference between these percentage fixations is not vers, similarly to genital arousal. These findings indicate that indicative of a sensitive measure of involuntary behavior. pupillary response is a useful alternative for measuring sexual 123 Arch Sex Behav (2016) 45:855–870 857 interest in male observers. In addition, the lack of specificity in stimulus categories. This manipulation can decrease image heterosexual female observers converges with a broad range of quality by reducing light–dark contrasts. A third version of these assessment methods (e.g., genital arousal, self-reported sexual scenes was therefore also included, in which image quality of the arousal and attraction, response time, and viewing time; Chi- original photographs was enhanced with graphics software. vers, 2005; Chivers, Rieger, Latty, & Baily, 2004; Ebsworth & Lalumie `re, 2012;Lippa, 2006, 2007, 2012; Lippa et al., 2010; Suschinsky, Lalumiere, & Chivers, 2009). This is an interest- Experiment 1 ing finding because it suggests that pupillary responses to sex- ual content are also consistent with more established measures Method in the literature. While few studies have focussed on pupil dilation as a mea- Participants sure of sexual interest for photographs of adults, there has been even less research on pupillary responses to persons of differ- A total of 44 students (22 male and 22 female) from the School ent ages. An early study compared these pupillary responses in of Psychology at the University of Kent participated in this study incarcerated male pedophiles and non-pedophiles to images of in return for a small payment or course credits. Participants com- nude women and immature girls (Atwood & Howell, 1971). pleted the Kinsey scale for the assessment of sexual orientation This experiment revealed greater pupil dilation in 90 % of non- as part of a pre-screen on our online recruitment system. This is a paedophilic observers to pictures of women, but a pupil constric- seven-point scale in which a score of ‘‘0’’ represents complete tion to the same pictures in 80 % of pedophiles. Conversely, heterosexuality and ‘‘6’’ complete homosexuality. Only partic- images of girls produced dilation in 90 % of pedophiles and a ipants who reported to be completely heterosexual (i.e., report- constriction or no change in 50 % of the non-pedophilic control ing ‘‘0’’ on the Kinsey scale) wereinvitedtotakepart(Kinsey, subjects. Pomeroy, & Martin, 1948; Kinsey, Pomeroy, Martin, & Gebhard, Up to now, there have been no documented attempts to repli- 1953). The mean age of participants was 21.8 years (SD = 4.2; cate these findings. This is surprising considering the potential range 18–35 years). All reported normal or corrected-to-normal applied value of such a measurement (e.g., the assessment of vision. child sex offenders). In this exploratory study, we investigated whether pupil dilation can provide an age-specific indication of Materials a person’s sexual interests. For this purpose, heterosexual male and female observers were presented with images of beach The stimuli consisted of natural beach scenes portraying men, scenes that contained semi-clothed adults and children, while women, and children (5 scenes for each of these four categories). their eye movements and pupil sizes were recorded. These scenes To determine the approximate age of these categories, ten obser- contained only a single person or no persons in the case of a set of vers (5 males, 5 females) estimated the age of the people in the comparison landscape beach scenes. We expected the different scenes in a pilot study. This revealed a mean age of 26.4 years person content of these scenes to draw attention depending on the (SD = 2.1) for men, 22.8 years (SD = 2.6) for women, 5.7 years sexual interests of the observers. For example, heterosexual male (SD = 1.1) for boys, and 4.7 years (SD = 1.4) for girls. The age observers were anticipated to fixate on women more frequently of the children therefore corresponds to stage 1 (prepubescent) than men (see Hewig et al., 2008;Lykinsetal., 2008;Rupp& of the Tanner stages of sexual development (see Tanner, 1978). Wallen, 2007). Of particular interest here was whether these Additionally, a set of control beach scenes without any person observers would also show an increase in pupil size to images content (5 scenes) was included, resulting in a total of 25 scenes. of sexually preferred adults in comparison with sexually non- People were portrayed in swim or leisure wear. All stimuli were preferred adults and children. purchased from an internet photograph database (www.most As a secondary aim, we also sought to examine how pupil- photos.com) and were selected to be of similar composition lary responses to people of sexual interest are affected by image and size, and to depict the persons in similar poses and with a luminance. The pupils constrict in response to light (i.e., increased comparable level of clothing (see Fig. 1). To confirm that these luminance) to protect the cells of the retina (Bergamin & Kardon, targets were of similar size, their percentage occupancy area in 2003; Ellis, 1981). If this differentially affects the stimulus cate- the scenes was calculated. This showed that all person categories gories in the current study, then this could influence the measure- occupied a similar amount of space in our scenes (mean = 7.1 %, ment of pupil responses as an index of sexual interest. In turn, it is SD = 3.4, range across person categories = 6.6–7.7 %; one- possible that the pupillary response to sexual content is clearer factor ANOVA: F(3, 19) = 0.14, p = 0.94). when luminance is controlled across different stimulus cate- In addition, three versions were created of each scene that gories. To explore this possibility, the original photographs of were identical in all aspects except for image quality. This the beach scenes were compared with alternative versions, in resulted in a total of 75 scene images. In the original quality which the mean luminance was equated across the different condition, the image quality of the downloaded photographs 123 858 Arch Sex Behav (2016) 45:855–870 Fig. 1 The stimuli of the original quality condition in Experiment 1 was retained. In the high-quality version, the images were pro- Table 1 Mean luminance, standard deviation, and the minimum and maximum luminance values of images within a stimulus category for the cessed by applying the ‘‘Auto Levels,’’ ‘‘Auto Contrast,’’ and original, high-quality, and luminance-controlled images for all scene ‘‘Auto Color’’functions in Adobe Photoshop CS3 to artificially conditions enhance the original photographs. Finally, to create a luminance- Mean SD Max Min controlled version of the stimuli, the photographs were divided into groups of five (one of each category) based on similar lumi- Original quality nance values and standard deviation. A mean luminance value Men 166 25 190 125 and standard deviation were calculated for each of the five groups. Women 160 29 200 125 Each photo within a group was then re-adjusted to obtain the mean Boys 169 42 218 111 luminance and standard deviation that matched the group value. Girls 190 35 224 133 Therefore, at least one image from each category (men, women, No-person 165 28 190 127 boys, girls, no-person landscapes) had precisely matched lumi- High quality nance values. This particular group-based approach was adopted Men 167 23 186 131 to avoid the extreme deviation from the natural luminance values Women 163 20 182 130 of individual scenes. This can occur when a single mean lumi- Boys 171 41 221 123 nance value is derived for large stimulus sets, which can result in Girls 184 38 211 122 some highly distorted and unnatural looking images. Table 1 No-person 152 16 180 143 shows the overall mean luminance values and standard devia- Luminance controlled tion for the different image categories for all scenes. Example Men 162 18 194 152 stimuli are shown in Fig. 2. Women 162 18 194 152 Two questionnaires were also included in the experiment. Boys 162 18 194 152 The first was a general information scale relating to sexual inter- Girls 162 18 194 152 est and instructed participants to select one or more of five appli- No-person 162 18 194 152 cable statements (‘‘no sexual interest in adults,’’‘‘strong sexual interest in female adults,’’‘‘some sexual interest in female adults,’’ participants completed the Interest in Child Molestation Scale to ‘‘some sexual interest in male adults,’’‘‘strong sexual interest in ensure that they were solely sexually interested in adults (Gannon male adults’’). This was included to confirm the sexual interests & O’Connor, 2011). This scale consists of five short scenarios that participants reported in the pre-screen. In addition, all 123 Arch Sex Behav (2016) 45:855–870 859 Fig. 2 Example stimuli of the original quality, high quality, and the luminance-controlled image conditions in Experiment 1 and the scrambled images in Experiment 2 that describe incidents of child molestation. In response to these were being recorded. Participants were kept naı ¨ve to the full pur- scenarios, participants have to rate their arousal, enjoyment, and pose of the experiment until the end. To fully understand obser- behavioral propensity to child sex abuse on 7-point Likert scales. vers’ natural interests in these scenes, a free-viewing paradigm This scale has high test–retest reliability (r = .94) and its sexual was used so as not to constrain spontaneous eye movement pat- arousal subscale correlates with the Implicit Association Test, terns. Thus, participants were instructed simply to‘‘view the which provides an indirect measure of child sexualization asso- scenes as you naturally would’’(for similar approaches, see Bin- ciations (see Gannon & O’Connor, 2011). demann, Scheepers, & Burton, 2009;Frombergeretal., 2012a, b, 2013;Halletal., 2011; Hewig et al., 2008; Lykins et al., 2008; Eye-Tracking Nummenmaa et al., 2012). Participants were seated in a quiet and windowless room The stimuli were displayed using SR-Research Experiment with consistent artificial lighting and positioned approximately Builder software (version 1.1.0) on a 21 color monitor, with 60 cm from the display monitor. The participants’ left eye was ascreen resolution of10249 768 pixels. Eye movements were tracked and calibrated using the standard Eyelink procedure. tracked using an SR-Research Eyelink II head-mounted eye- To calibrate the eye tracker, observers fixated an initial series tracking system. The Eyelink II was running at a 500 Hz sam- of nine target points on the display monitor. Their accuracy was pling rate, a spatial resolution of\0.01 of visual angle, a gaze then validated against a second series of nine fixation targets. position accuracy of\0.5, and a pupil size resolution of 0.1 % Calibration was repeated if poor measurement accuracy was of diameter. The Eyelink II system works by measuring corneal indicated. In the experiment, each trial began with a central fix- reflection and dark pupil with a video-based infrared-camera eye ation dot, which allowed for drift correction. This was followed by a gray screen display for 1000 ms, and then the stimulus dis- tracker, which computes the number of camera pixels that are occluded by participants’ pupils. In this system, the diameter of play for 5000 ms, followed by another gray screen for 1000 ms. the pupil is recorded as an integer that ranges from 400 to This display duration is similar to other studies with static images 16,000 units. The device incorporates eye and head tracking (e.g., Fromberger et al., 2012a, b, 2013; Hewig et al., 2008;Num- that automatically compensates for minor head movements. Dur- menmaa et al., 2012) and allows for approximately 15 fixations ing the recording of eye movements, participants are instructed to (based on an average fixation duration lasting 200–300 ms, see remain seated still but further immobilization (e.g., a chinrest) is Rayner, 1998), which is sufficient time to scan the entire scene. not required. This eye-tracking system is compatible with most Each participant viewed all 75 stimuli. These were pre- glasses and contact lenses. sented in a randomized order that was uniquely generated for each participant by the EyeLink software. Short breaks were Procedure inserted every 25 trials, after which the calibration procedure was repeated. On completion of the eye-tracking task, partic- Participants were invited to take part in an experiment on sex- ipants answered the general information scale relating to their ual interest and informed that they would be viewing images sexual interests and the Interest in Child Molestation Proclivity of males and females of different ages while their eye movements scale (see Gannon & O’Connor, 2011). 123 860 Arch Sex Behav (2016) 45:855–870 Results from the overall mean was then computed, using the formula: (mean pupil diameter for category 9 100)/overall pupil mean. Confirmation of Sexual Interests Accordingly, a score of 100 % indicates that the pupillary response to a stimulus category does not differ from the overall To ensure that participants were not sexually interested in chil- mean. Scores higher or lower than this value indicate compar- dren, responses on the Interest in Child Molestation Scale were atively larger or smaller pupil sizes (for similar approaches, see analyzed first. An overall interest score was calculated for each Dabbs, 1997; Laeng & Falkenberg, 2007). To simplify the expres- participant by combining responses across all subscales (i.e., sion of these patterns, these scores were then deducted from 100 arousal, enjoyment, behavioral propensity) (for similar analysis, so that no change in pupil size is indicated by zero and positive or see Gannon & O’Connor, 2011). This produced a total score negative scores reflect relatively larger (dilation) or smaller (con- where a minimum of 15 (low sexual interest in children) and a striction) pupil sizes in response to a stimulus category. maximum score of 105 (high sexual interest in children) are possible. The results here converge with those obtained in pre- Viewing Behavior vious studies with a sample of non-offending community males (Gannon & O’Connor, 2011), such that male observers scored a We first examined the viewing patterns that the persons in the mean of 18.1 (mode = 15,SD = 5.6, min = 15, max = 30) and scenes elicited in male and female observers. To examine this, 16.8 for female observers (mode = 15, SD = 5.6, min = 15, the percentage fixations to the ROIs were calculated for all stim- max = 41). However, an established cut-off point for this scale ulus categories (see Fig. 3). Overall, 63 % of fixations fell on the does not exist. We adopted a simple metric by considering only figures in the scenes (range 58 to 71 % across conditions), which individuals with scores on the lowest third of the scale (i.e., with indicates that the person content of the scenes was of most inter- scores between 15 and 45). All participants fell within this range. est. A 4 (category: men, women, boys, girls) 9 3(ROI: head, Sexual orientation was confirmed with the general informa- body, background) 9 2 (observer sex: male, female) mixed- tion scale that was administered following the eye-tracking task factor ANOVA revealed a three-way interaction, F(6, 252) = (see‘‘Materials’’section). In the 22 male observers, 19 reported 8.01, p\0.001, partial g = 0.16. To explore this interaction, ‘‘strong sexual interest in women’’ and three selected ‘‘some sex- two separate 4 (category: men, women, boys, girls) 9 3 (ROI: ual interest in women.’’Among the 22 females, 12 selected‘‘strong head, body, background) within-subjects ANOVAs were per- sexual interest in males’’ and 10 selected ‘‘some sexual interest in formed for male and female observers. males.’’ Participants reported no other sexual interests in this For male observers, this analysis showed no main effect of questionnaire. category,F(3, 63)= 0.32,p= 0.81, partialg = 0.02, but revealed a main effect of ROI, F(2, 42)= 4.54, p\0.05, partial g = 0.18, Data Analysis and an interaction between both factors, F(6, 126)= 34.22, p\ 0.001, partial g = 0.62. To explore this interaction, Bonferroni- For the analysis of the eye-tracking data, all eye movements adjusted pairwise comparisons of the stimulus categories were were pre-processed by merging fixations of less than 80 ms with conducted for each ROI. These comparisons show that more the preceding or following fixation if it fell within half a degree fixations were directed at the background of scenes containing of visual angle (for similar approaches, see e.g., Attard & Bin- boys, girls, and men (39 to 42 %) than scenes depicting women demann, 2013; Bindemann et al., 2009; Bindemann, Scheepers, (30 %), all ps\0.01. In addition, boys (31 %) and girls (32 %) Ferguson, & Burton, 2010). In addition, any fixations that fell received more fixations to the head than men (27 %) and women outside the dimensions of the display monitor or that were (22 %), all ps\0.01, and men’s heads were also fixated more obscured by blinking were excluded. To analyze attention to frequently than those of women, p\0.01. By contrast, male specific areas within the visual scenes, each image was then coded observers directed more fixations to the bodies (48 %) of to define three regions of interest (ROIs), which comprised the women and men (34 %) than those of boys (27 %) and girls head and body of the persons and the scene background. The (26 %), all ps\0.001, and more at women’s bodies than those mean percentage of fixations that fell on these ROIs was then of men, p\0.001. None of the other comparisons reached sig- calculated across observer groups (males, females) and stimulus nificance, all ps C 0.10. categories (men, women, boys, girls). The equivalent analysis for female observers showed no For the measure of main interest, observers’ pupillary main effect of category, F(3, 63) = 0.16, p = 0.92, partial g = responses were computed by taking the mean pupil diameter 0.008, but a main effect of ROI,F(2, 42) = 2.58,p\0.001, par- at each fixation, averaged across the duration of a stimulus tial g = 0.11, and an interaction between factors, F(6, 126)= display. These values were then used to compute an overall 8.45,p\0.001, partialg = 0.29. Bonferroni-adjusted pairwise mean, across all stimuli, for each participant. The percentage comparisons of the stimulus categories show that more fix- difference (i.e., an increase or decrease) in pupil diameter for each ations landed on the head region of boys and girls (both 34 %) stimulus category (men, women, boys, girls, no-person scenes) than women (22 %) and men (29 %), all ps\0.001, and on the 123 Arch Sex Behav (2016) 45:855–870 861 Fig. 3 Mean percentage fixations to the head and body of the target persons and the scene background for male and female observers in Experiment 1. Lines represent standard errors of the means heads of men than women, p\0.001. By contrast, more fix- However, an interaction between image quality and observer ations landed on women’s bodies (40 %) compared to boys sex was found,F(2, 84)= 3.36,p\0.05, partialg = 0.07. Bon- (29 %) and girls (31 %), both ps\0.01. No other comparisons ferroni-adjusted pairwise comparisons revealed only that female reached significance, all ps C 0.08. observers exhibited larger pupils than male observers during the Overall, this pattern suggests a clear interest, whereby hetero- viewing of luminance-controlled scenes, p\0.05. No other dif- sexual males and females fixate men and women more fre- ferences were significant, all ps C 0.09. An interaction between quently than children, but are particular biased towards the image quality and category was also found, F(8, 336) = 2.17, bodies of adult female targets. p\0.05, partial g = 0.05, as the no-person beach scenes eli- cited smaller pupils in the luminance-controlled than the high Pupillary Responses quality, p\0.01, and original quality conditions, p\0.05. No other differences between any of the person content scenes The measure of main interest is pupillary response, which was were found, all psC 0.16. Therefore, image quality was not ana- lyzed further. analyzed in two ways. In the first analysis, pupillary responses were compared for male and female observers across the stim- An interaction between category and observer sex was also ulus categories and image conditions. These data are illustrated present, F(4, 168) = 2.73, p\0.05, partial g = 0.06. Bonfer- in Fig. 4. A 3 (image quality: original, high, luminance-controlled) roni-adjusted pairwise comparisons revealed smaller pupils in 9 5 (category: men, women, boys, girls, no-person)9 2(observer male than female observers during the viewing of men, p\ sex: male, female) mixed-factor ANOVA revealed a main effect 0.01. Furthermore, in male observers, women elicited larger of category, F(4, 168)= 20.35, p\0.001, partial g = 0.33, but pupil sizes than men, boys, girls, and no-person scenes, all ps B not of quality, F(2, 84) = 1.75, p = 0.18, partial g = 0.04, or 0.001. For female observers, women elicited larger pupil sizes observer sex, F(1, 42) = 1.00, p = 0.32, partial g = 0.02. than boys, girls, and no-person scenes, all ps B 0.05, but not men, p = 0.26. In addition, pupil responses were larger for scenes depicting boys than girls, p\0.05. No other differences were observed, allpsC 0.06, and an interaction between the three factors was not found, F(8, 336)= 1.10, p= 0.36, partial g = 0.03. Overall, these results therefore reveal a dilation response in male observers that appears to be consistent with self-reported sex and age preferences. Female observers’ responses are also consistent with their age preferences, but do not correspond with their reported sexual interest in adult men. In the second analysis, this pattern is confirmed when pupi- llary responses are compared via one-samplet-tests (with alpha corrected at p\0.01 for multiple comparisons) with a baseline that reflects the mean pupil diameter across all stimuli (see‘‘Data Analysis’’section). This analysis shows that the pupils of male observers were larger than baseline during the viewing of women,t(21)= 5.43, p\0.001,d= 2.37, and smaller during the viewing of men, t(21)=-3.02, p= 0.006, d= 1.32, and girls, Fig. 4 Percentage pupillary change for all stimulus categories for male t(21)=-3.1, p = 0.005, d = 1.35. In addition, pupil size was and female observers in Experiment 1. Lines represent standard errors of unchanged from baseline in response to boys and no-person the means 123 862 Arch Sex Behav (2016) 45:855–870 scenes, both tsB-1.59, ps C 0.126, ds B 0.69. In female elicited a clear pupillary dilation that was not present during observers, pictures of men,t(21) = 1.49, p = 0.15, d = 0.65, the viewing of men and children. In female observers, pupil boys, t(21)=-0.12, p = 0.91, d = 0.05, and landscape beach dilation was also greatest when pictures of women were viewed. scenes (-1.53 %), t(21)=-2.19, p = 0.04, d = 0.96 did not In these participants, pupillary recordings therefore do not cor- elicit a change in pupil size from baseline. The pupils were respond to their self-reported sexual orientation. However, these enlarged to scenes with women, t(21) = 4.71, p\0.001, d = responses still appeared to be age-specific as the pupils remained 2.06, and smaller than baseline during the viewing of girls, unchanged or constricted during the viewing of children. t(21)=-4.33, p\0.001, d = 1.89. These results converge with a recent study that has shown a similar pattern of pupillary responses for heterosexual adult Individual Differences in Pupillary Responses males and females (Rieger & Savin-Williams, 2012). Exper- iment 1 extends these findings by demonstrating that such pupil- We also sought to explore whether pupillary responses can be lary responses are also age-specific. A question that arises, how- informative about the sexual interests of individual observers. ever, is whether these dilation effects could be attributed to a low- For this purpose, the difference in raw pupil size for specific level factor such as luminance. To explore this possibility, we image comparisons (e.g., scenes with men vs. women) was cal- also compared scene photographs in which contrast and color culated separately for each participant. These data show, for were enhanced with a set in which luminance and contrast were example, that all of the male observers (22/22) recorded larger equated. The results for these stimulus categories were highly pupil sizes during the viewing of women than men, and 91 % comparable, which suggests that pupillary responses for the dif- (20/22) of male observers displayed larger pupils in response to ferent person categories cannot be explained by general variation in luminance. women than girls. In addition, only 22 % (5/22) of these partic- ipants showed a greater pupillary response to men than boys. There is, however, a problem with the luminance adjust- With regard to female observers, 73 % (16/22) showed more ment that was employed in Experiment 1. While this manip- pupil dilation during the viewing of women than men. However, ulation was used to equate luminance across scenes, it does not 86 % (19/22) of this participant group also exhibited larger pupils control other low-level image aspects, such as color, which might in response to women than girls, and 59 % (13/22) recorded also affect pupillary responses (Kohn & Clynes, 1996; Lobato- larger pupils to men than boys. Rinco ´netal., 2014). Such information was not matched across stimulus categories in Experiment 1. Consequently, the possibil- Discussion ity remains that the results might reflect such image artifacts. A second explanation is also possible for the observed pupil- The purpose of this experiment was to explore whether pupil- lary responses. While we adjusted the mean luminance of the lary responses to the visual presentation of men and women scenes, we did not measure the sexual attractiveness of the tar- can provide an indication of a person’s sexual interests. More get figures. As a result, this might have been mismatched across specifically, we sought to determine whether this approach can categories. Considering that photographs of women elicited be extended to reveal age-specific sexual interests. We first looked more pupil dilation in both male and female observers, it is at fixation patterns on the person content in scenes. Male obser- conceivable, for example, that these pictures were generally vers showed a viewing preference for women over men and more sexually arousing than those of men. To investigate these children, which was characterized by a high number of fixa- possibilities, a second experiment was conducted. tions on women’s bodies. These results are consistent with other studies, which have shown that heterosexual male observers attend more to images of the opposite sex (Lykins, Meana, Experiment 2 &Strauss, 2006;Lykinsetal., 2008;Rupp& Wallen, 2007; Suschinsky et al., 2007) and that such preferential viewing In Experiment 2, a new condition was created, in which the behavior is also age-specific (Fromberger et al., 2012a, b, 2013; pixels of the luminance-controlled images were randomized. Hall et al., 2011). Female observers also recorded fewer fixa- These scrambled images are no longer recognizable as the orig- tions on the faces of women than men and children, but more on inal scenes but provide the same color content. If the pupillary women’s bodies than those of children. Consistent with previ- responses in Experiment 1 reflect a low-level color artifact, then ous research, heterosexual females therefore showed age-speci- the same pattern should persist with these scrambled scenes in fic viewing patterns but did not exhibit the same strong visual Experiment 2. The experiment also examined whether the pic- preferences to opposite-sex figures as men (Hall et al., 2011; Israel tures of men and women in Experiment 1 were matched in terms &Strassberg, 2009; Lykins et al., 2008; Rupp & Wallen, 2007). of their perceived attractiveness. For this purpose, two measures The data of main interest were the pupillary responses. In of attractiveness were employed. The first measured general heterosexual male observers, these responses were consistent sexual appeal and recorded how attractive observers thought the stimuli were to others (i.e., sexual appeal by‘‘societal standards’’; with their reported sexual interests. Thus, pictures of women 123 Arch Sex Behav (2016) 45:855–870 863 for similar approaches, see Lippa et al., 2010). The second mea- et al., 2010). For all four tasks, the stimulus sequence in each sure concerned the sexual appeal that these images personally block was generated randomly by the display software for each hold for the individual observer (see Ebsworth & Lalumiere, participant. As in Experiment 1, participants completed the same 2012; Hewig et al., 2008). If the pupillary responses in Exper- general information scale and the Interest in Child Molestation iment 1 reflect sexual arousal, then personal sexual appeal rat- proclivity scale on completion of the eye-tracking tasks. ings should correlate with pupillary responses in Experiment 2. Results Method Confirmation of Sexual Interests Participants Once again, the responses on the Interest in Child Molestation A total of 41 students (21 male) from the University of Kent Scale were analyzed first. One of the male participants pro- participated in this study in return for a small payment or course duced a score of 52. This is the only score that falls above the credits. The mean age was 19.5 years (SD = 2.0; range 18–31 lowest third (i.e., 45) of the Child Molestation Scale in Exper- years). All participants reported to be exclusively heterosexual iment 1 and 2. It also exceeds the mean score (41.4) of ped- on the Kinsey scale (Kinsey et al., 1948, 1953), which was com- ophiles that have self-reported sexual acts with children (Mitchell pleted as a pre-screen on our online recruitment system. None of & Galupo, 2015). This individual was therefore excluded from the participants had taken part in thefirstexperiment. Allreported further analysis. For the remaining participants, means of 20.8 normal or corrected-to-normal vision. (mode = 15,SD = 6.2, min = 15, max = 34) and 16.3 (mode = 15, SD = 2.4, min = 15, max = 23) were obtained for male and Materials female observers, respectively. To confirm that participants showed a sexual interest towards This experiment employed the same eye-tracking set-up with the opposite sex, their responses on the sexual interests’ ques- the luminance-controlled stimuli from Experiment 1. To assess tionnaire were also analyzed. Nineteen of the 20 males reported the contribution of color within each of these 25 images (com- ‘‘strong sexual interest in women’’ and one reported ‘‘some prising five men, women, boys, girls, and no-person scenes) to sexual interest in women.’’ For the females, 14 of 20 reported pupillary response, the pixels in each image were randomized. ‘‘strong sexual interest in males,’’while the remaining six par- The resulting images provide a‘‘scrambled’’condition in which ticipants reported‘‘some sexual interest in males.’’Participants the original image content is not discernible (see Fig. 2;for sim- reported no other sexual interests in this questionnaire. ilar approaches, see Jenkins, Lavie, & Driver, 2003; VanRullen, 2006). Data Analysis Procedure The eye-tracking data were processed as in Experiment 1. Note that pupillary responses are reported for both free-viewing tasks The experiment consisted of four blocks. In the first block, (Block 1 and 2) but not for the two ratings tasks. In the latter participants were shown the 25 scrambled scene images. This tasks, 5.9 (SD = 3.7) and 6.5 (SD = 4.3) fixations were recor- was followed, in the second block, by the 25 unscrambled ver- ded on average per trial but the mean number of fixations varied sions of these stimuli. Both blocks were free-viewing tasks. greatly across observers (from 1 to 38). Consequently, these Each trial therefore consisted of a drift correction, which was tasks did not provide reliable eye movement data for analysis. followed by a gray mask for 1000 ms. The scrambled/intact scene The eye fixations for the free-viewing task with the intact scenes stimuli were then presented for 5000 ms, followed by the gray (Block 2) were also analyzed and revealed a similar pattern to Experiment 1. For brevity, this analysis is not reported here but mask for a further 1000 ms. In both blocks, participants were simply instructed to view these images naturally. is available on request. These data are not meaningful for the In the remaining blocks, the intact scenes with the men (5 scrambled scene images in Block 1 and are therefore also omitted. images), women (5 images), and children (5 images each) from Block 2 were repeated. In Block 3, participants were asked to Pupillary Responses provide personal sexual attractiveness ratings for these people (i.e., based on how sexually attractive they themselves find these The data of main interest were the pupillary responses. As in images) using a Likert scale ranging from 1 (‘‘not at all sexually Experiment 1, the mean percentage change in pupil size was appealing to me’’) to 7 (‘‘extremely sexually appealing to me’’). calculated for male and female observers for the person cat- In Block 4, participants were then asked to evaluate the people in egories (see Fig. 5) and was analyzed in two ways. First, a 5 the scenes based on their sexual attractiveness by societal stan- (category: men, women, boys, girls, no-person) 9 2 (observer dards using the same scale (for similar methods, see, e.g., Lippa sex: male, female) mixed-factor ANOVA showed a main 123 864 Arch Sex Behav (2016) 45:855–870 Fig. 5 Percentage pupillary change for all stimulus categories for male and female observers in Experiment 2 for intact scenes (left graph) and scrambled scenes (right graph). Lines represent standard errors of the means effect of category, F(4, 152) = 32.16, p\0.001, partial g = stimulus categories in raw pupil diameter during the free-view- 0.46. Post hoc analysis revealed overall larger pupils during ing task (Block 2). These data show that 80 % (16/20) of the male the viewing of women compared to all other categories, all psB participants displayed larger pupils when viewing women than 0.001, and larger pupils to men than boys, girls and no-person men, 95 % (19/20) displayed larger pupils to women than girls, scenes, all ps B 0.01. No other differences were found, all ps C and 85 % (17/20) displayed larger pupils to men than boys. Of 0.34. A main effect of observer sex, F(1, 38) = 0.05, p = 0.82, the female observers, 65 % (13/20) recorded larger pupils to partial g = 0.001, and an interaction between factors, F(4, women than men, 90 % (18/20) displayed larger pupils to women 152)= 2.01, p = 0.96, partial g = 0.05, was not found. than girls, and 90 % (18/20) displayed larger pupils to men than For completeness, these responses were also analyzed with boys. one-samplet-tests (withalpha corrected atp\0.01 for multiple comparisons), by comparing the change in pupil size for each Personal Sexual Appeal Ratings stimulus category with a baseline of zero (see ‘‘Data Analysis’’ section). For male observers, this analysis revealed pupil dilation In the next step of the analysis, we explored the extent to which during the viewing of women, t(19)= 7.58, p\0.001, d= 3.48, personal sexual appeal judgements of the persons in the scenes and pupil constriction during the viewing of boys,t(19)=-4.40, relate to pupil responses in the free-viewing task. For this pur- p\0.001, d = 2.02 and no-person scenes, t(19)=-4.62, p\ pose, the mean sexual appeal ratings for each of the person cat- 0.001, d = 2.12. A change in pupil size was not detected in egories were analyzed first. A 4 (category: men, women, boys, response to images of men, t(19)= 1.26, p = 0.22, d= 0.58 and girls)9 2 (observer sex: male and female) mixed-factor ANOVA girls, t(19)=-1.23, p = 0.24, d = 0.56. of these data did not show a main effect of observer sex, F(1, In female observers, dilation was also observed in response 38)= 0.02,p= 0.88, partialg = 0.00, but revealed a main effect to pictures of women, t(19) = 7.25, p\0.001, d = 3.33. How- of category, F(3, 114)= 83.26, p\0.001, partial g = 0.69, and ever, in this case, dilation was also found for pictures of men, an interaction between factors, F(3, 114) = 87.53, p\0.001, t(19)= 3.30,p= 0.004,d= 1.51. In contrast, the pupils appeared partialg = 0.70. Bonferroni-corrected post hoc comparisons to be smaller than baseline during the viewing of boys, t(19)= showed that male observers rated women as more sexually -2.65, p= 0.02, d= 1.22, girls, t(19)=-2.05, p= 0.05, d= appealing (M = 5.4, SD = 0.9) than men (M = 1.6, SD = 0.8), 0.94, and the no-person scenes, t(19)=-2.25, p = 0.04, d = boys (M = 1.2, SD = 0.8), and girls (M = 1.2, SD = 0.7), all 1.03, but these changes were not significantly below zero (with ps\0.001. In contrast, female observers rated men as more alpha corrected at p\0.01 for multiple comparisons). sexually appealing (M = 4.3, SD = 1.40) than women (M = In summary, this analysis shows that male observers’ pupils 2.1, SD = 1.2), boys (M = 1.3, SD = 0.9), and girls (M = 1.5, dilate in response to pictures of women but not men or children. SD = 1.3), all ps\0.001. No other differences were found. Female observers show a dilation response to both men and Overall, these sexual appeal ratings therefore converge clearly women, but not to children. These results therefore replicate the with observer’s self-reported sexual interest in adults of the oppo- sex-specific effect in male observers and the age-specific pattern site sex. that was observed in male and female observers in Experiment 1. We next performed a correlation between the mean pupil- lary change (%) in the free-viewing task (Block 2) and the sex- Individual Differences in Pupillary Responses ual appeal ratings. This analysis combined the person categories As in Experiment 1, we also performed a simple analysis of When this analysis was performed within category groups, no individual performance, based on the differences between correlations between pupillary response and appeal ratings were found, 123 Arch Sex Behav (2016) 45:855–870 865 (men, women, boys, girls) but was performed separately for male Scrambled Scenes and female observers. The distribution of observers’ sexual appeal ratings was skewed. Therefore, non-parametric Spear- The pupillary responses to scrambled scenes were analyzed man’s correlations are reported. For male observers, a strong pos- next. As in the analysis of intact scenes, the mean pupillary responses for each category (men, women, boys, girls, no-per- itive correlation between pupil change and sexual appeal ratings was found, r (78) = 0.64, p\0.001. This correlation also per- son scenes) were transformed to measure mean percentage change (see Fig. 5). A 5 (category: men, women, boys, girls, no- sisted when only the adult targets (men and women) were con- sidered, r (38) = 0.58, p\0.001, which suggests that it reflects person) 9 2 (observer sex: male, female) mixed-factor ANOVA did not show a main effect of observer sex, F(1, 38) = 0.00, p= observers’ sexual interests in specific adults. For female obser- vers, the correlation across all person categories (men, women, 1.00, partial g = 0.001, or an interaction between factors, F(4, boys, girls) was weaker, r (78) = 0.28, p\0.01, and was not 152) = 0.97, p = 0.43, partial g = 0.03, but revealed a main reliable when the child categories were excluded from analysis, effect of category, F(4, 152)= 4.34, p\0.01, partial g = 0.10. r (38)=-0.22, p = 0.17. Overall, these data therefore suggest Post hoc Bonferroni comparisons showed that observers’ pupils that pupillary responses provide a good index of sexual interest were smaller while viewing scrambled images of boys than in male, but not female, observers. those of women, p\0.01, and no-person scenes, p\0.01. No other differences between categories were found, all psC 0.20. General Sexual Attractiveness Ratings Once again, these responses were also analyzed via a series of one-sample t-tests (with alpha corrected at p\0.01) to com- In Block 4, the subjects were asked to objectively rate the per- pare the change in pupil size to a baseline of zero (see ‘‘Data Analysis’’section). This analysis showed no change in pupil sons in the scenes on their sexual attractiveness based on how they thought the general population would respond. The mean size across categories in male observers, all ts B 2.23, ps C ratings were analyzed with a 4 (category: men, women, boys, 0.04, ds B 1.02. The pupils of female observers were smaller girls) by 2 (observer sex: male and female) ANOVA. This during the viewing of scrambled scenes of boys, t(19) = 3.46, analysis did not show a main effect of observer sex, F(1, 38) = p\0.01, but no other differences were found, all tsB 1.83, psC 0.45, p = 0.51, partialg = 0.01, but a main effect of category, 0.08,ds B 1.59. We also correlated pupil sizes for scrambled and F(3, 114) = 331.15, p\0.001, partial g = .90, and an interac- intact scenes. This revealed no relationship between these con- ditions in male and female observers, r(98) = 0.06, p = 0.58 and tion between factors, F(3, 114) = 2.96, p\0.035, partial g = 0.07. Bonferroni-corrected post hoc comparisons revealed that r(98)= 0.04, p = 0.72, respectively. These results therefore indicate that pupillary responses to intact scenes do not reflect male observers rated the women in scenes (M = 6.0, SD = 0.6) higher on sexual attractiveness than men (M = 4.8,SD = 1.02), low-level image artifacts, such as color. p\0.001. Both adult categories were also rated higher than boys (M = 1.4, SD = 0.9) and girls (M = 1.4, SD = 0.9), all ps\ Discussion 0.001. Female observers rated men (M = 5.6, SD = 1.0) and women (M = 5.7, SD = 1.1) more similarly (p = 1.00), and This experiment assessed further whether observers’ pupillary more sexually attractive than boys (M = 1.4, SD = 1.0) and responses reflect their sexual interest in a seen stimulus. For girls (M = 1.5, SD = 1.2), both ps\0.001. No other differences this purpose, we compared pupillary responses to pictures of were observed. men and women with personal sexual appeal ratings and gen- eral attractiveness ratings (by societal standards). The pupils of A non-parametric Spearman’s correlational analysis between these ratings and observers’ pupillary responses (% change), male observers dilated to pictures of women but not men or chil- dren. Female observers showed pupillary dilation to pictures of which combined the data from all person categories (men, women, boys, girls), revealed a correlation for male and female observers, women and men but not to children. This experiment therefore r (78) = 0.62, p\0.001 and r (78) = 0.55, p\0.001, respec- replicates the age-specific dilation effects in male and female s s tively. Similar to the previous analysis, we performed a second observers that were shown in Experiment 1, and also the sex- correlation for which the data for child targets were excluded. specific dilation effect in males. This correlation was not significant in male, r (38) = 0.29, p = The personal sexual appeal ratings support the notion that 0.08, or female observers, r (38) = 0.07, p = 0.67. these pupillary responses reflect the sexual interests of hetero- sexual male observers (Rieger et al., 2015; Rieger & Savin- Williams, 2012). For example, these observers rated the pho- tographs of women as much more sexually attractive than those Footnote 1 continued of men and children, and these ratings correlated strongly with all ps C 0.06. We attribute this to the low number of images in each pupillary responses. This was evident when data from all person stimulus category (five) and the low variance in sexual appeal ratings categories were combined, but also when the children were omit- within categories. For example, male observers’ mean sexual appeal ted from the analysis. This suggests that the pupillary responses rating for female figures was 5.36 with a standard deviation of only 0.89. 123 866 Arch Sex Behav (2016) 45:855–870 of male observers reflect the sexual interest that is triggered by the presentation of women but not during the viewing of men stimuli. and children. This suggests that these pupillary responses are In line with their reported sexual orientation, heterosexual linked to the sexual interest of these observers (i.e., females) female observers rated male targets as most sexually appeal- and are also age-specific (adults). In contrast, the pupils of hetero- ing, while women and children received low ratings. These sexual female observers dilated to images of women and men, ratings diverge from their pupillary responses, which indicate but not to children. In these observers, pupillary responses there- dilation to pictures of men and women. In addition, a correla- fore appear to be age-specific but do not correspond to self- tion between sexual appeal ratings and pupillary responses was reported gender interests. found, but this did not hold when child categories were exclu- In light of these different effects in male and female obser- ded from analysis. This pattern deviates from our findings with vers, a further experiment was conducted to explore more directly heterosexual male observers. It is interesting to note, however, whether pupillary responses are linked to observers’ sexual inter- that such discrepancies were also obtained for pupil dilation and est. For this purpose, we recorded pupillary responses to male and subjective arousal in a recent experiment (Rieger et al., 2015) female adults and children and also asked observers to rate these and are commonly observed in studies comparing self-reported target persons in terms of their sexual attractiveness. Two mea- and physiological measures of sexual arousal in heterosexual sures were utilized for this purpose, which sought to capture the women (Rieger et al., 2015; Suschinsky & Lalumiere, 2012; sexual attractiveness that these stimuli personally held for an Suschinsky et al., 2009; for a meta-analysis, see Chivers, Seto, observer as well as their general sexual attractiveness to others. Lalumiere, Laan, & Grimbos, 2010). The pupillary responses in this experiment replicated the sex- We also investigated whether the pupillary responses of and age-specific effect in male observers and the age-specific male and female observers might reflect differences in the gen- effect in female observers. This suggests, once again, that pupil- eralattractiveness of the stimulus categories, by measuring how lary response can provide a measure of sexual interest for male sexually attractive observers thought the stimuli were to others. but not female observers. Male observers rated children and adult males as less generally These findings received further support from the ratings tasks. attractive than adult females. However, the difference between The relationship between personal sexual appeal ratings and male and female stimuli was smaller than for the personal appeal pupillary responses was weak for females and driven by the ratings, indicating some adjustment. This difference was smal- age of the persons in the scenes. However, the ratings of male ler still in female observers, who perceived men and women to observers showed a clear preference for adult females and cor- be of similar general sexual attractiveness. Moreover, while the related well with pupillary response, which suggests that it reflects general attractiveness ratings correlated with pupillary respon- the sexual interests of the males in this study. By contrast, male ses, this did not hold for male or female observers when the child and female observers perceived the general sexual attractive- categories were excluded from analysis. This suggests that the ness of men and women to be more comparable and these rat- general sexual attractiveness of male and female adult stimuli ings did not correlate with pupillary response. Taken together, was not grossly mismatched in the current experiments, or that these findings suggest that pupillary responses reflect the per- this was the key determinant of pupillary responses. sonal sexual interests of male but not female observers, but are We also explored whether the pupillary pattern could arise age-specific in both groups. from low-level artifacts within the scene images (Kohn & Cly- The responses of male observers to images of women con- nes, 1969; Lobato-Rinco ´netal., 2014). To investigate this pos- verge with previous research, which has also shown an increase sibility, a control condition of scrambled images was included, in pupil size to such content (Hess et al., 1965; Rieger & Savin- which are no longer recognizable as coherent scenes but retain Williams, 2012;Riegeretal., 2015). Female observers recorded their color content. These scrambled scenes failed to produce pupil dilation in response to images of men in Experiment 2 but pupillary dilation that corresponds with responses to the intact also displayed larger pupils for images of women across both scenes. These findings therefore converge with the sexual appeal experiments. The reason for this is unclear. However, this absence and attractiveness ratings to indicate that the pupillary responses of sex-specific pupillary responses for female observers is also in this study are driven by the person content of the scenes. consistent with other paradigms in this field, such as viewing time studies (Israel & Strassberg, 2009; Lippa et al., 2010), as well as self-reports and physiological arousal (Chivers et al., General Discussion 2004, 2010; Steinman, Wincze, Sakheim, Barlow, & Mavis- sakalian, 1981; Suschinsky et al., 2009). For example, in these The study examined whether pupillary responses to photographs studies, women frequently show increased physiological arou- of people can provide an indication of an observer’s sexual inter- sal to images of both sexes (e.g., Chivers et al., 2004;Wincze& ests. We specifically sought to determine whether such responses Qualls, 1984) and weaker correlations than men with self-reported are sensitive to the age of targets. Experiment 1 showed that preference and sexual arousal (Chivers et al., 2004; Schmidt, pupils of heterosexual male observers dilated during the 1975). These findings indicate that women’s sexual interests 123 Arch Sex Behav (2016) 45:855–870 867 are organized differently to those of men (Lippa, 2006, 2007; still exists that observers can manipulate such responses Suschinsky et al., 2009) and may not be as strongly linked to voluntarily by avoiding person content in the visual field (Bin- arousal patterns (for a review, see Chivers, 2005). The current demann et al., 2007), or by causing pupil constriction through experiments suggest that this also applies to pupillary responses. focusing on high-luminance scene regions. Considering that par- It is noteworthy that our pupillary responses in males and ticipants in this study were naı ¨ve to the full purpose of the exper- females are also consistent with a small set of studies from the iment until the end, it is unlikely that such methods were adopted 1960s, which first assessed pupil dilation with an elementary to exert top-down control on pupillary responses. Nonetheless, video-frame analysis (Hess et al., 1965; Scott et al., 1967), and this is clearly another important avenue for further investigation. a recent study that verified these findings with contemporary We have also only been able to demonstrate pupillary eye-tracking equipment (Rieger & Savin-Williams, 2012). responses with male adult observers who are sexually inter- The current experiments extend this recent work by demonstrat- ested in other adults but not in children. We therefore acknowl- ing that such pupillary responses are also age-specific, whereby edge that further work with a pedophilic population and con- the pupils of non-pedophilic observers dilate to pictures of adults temporary eye-tracking equipment is required to determine fully but not children. This age-specific effect represents, in fact, the whether pupillary responses can detect such inappropriate sex- most consistent aspect of our results. ual interests. In future research, it would also be valuable to com- This is an important finding that raises the possibility that pare pupil dilation directly with other existing measures of deviant pupillary response could be used as a measure of deviant sex- sexual interest, such as Implicit Association Tests (Babchishin, ual interest in children in the assessment and rehabilitation of Nunes, & Herman, 2013), Stroop Tasks (O Ciardha & Gormley, offending populations (Gannon et al., 2004; Laws & O’Donohue, 2012; Price & Hanson, 2007), and Choice Reaction Time tasks 2008). To this point, it is notable that the lack of pupil dilation by (Mokros et al., 2010;Wright & Adams, 1994). This may serve to male observers during the viewing of boys and girls is consistent strengthen the validity and assessment value of these diagnostic with an old study that compared pedophilic and non-pedophilic measures, and would also help to establish the comparative males with a more elementary approach (Atwood & Howell, strength of a pupil dilation paradigm. 1971). In that study, pupillary response appeared to provide an index of age-specific sexual interests in 77 % of individual Conclusion observers. The current study also recorded larger pupillary responses to women than men in the majority of male observers This is the first study to show with contemporary eye-tracking (100 and 80 % of participants in Experiment 1 and 2, respec- equipment that pupillary responses provide a promising method tively), and to women than girls (91 and 95 % of participants in for measuring age-specific sexual interests. We have only been Experiment 1 and 2). able to demonstrate this with male adult observers who are sex- ually interested in other adults and not in children. We therefore acknowledge that further work is required to determine fully Limitations and Directions for Future Research whether pupillary responses can detect pedophilic sexual inter- ests. However, pupil dilation appears to be a highly promising This is an exploratory study with limitations. For example, we method for assessing such deviant sexual interests. This mea- sought to increase ecological validity by using images of beach sure seems to relate directly to observers’ sexual interest in scenes, as these provide a natural setting to display semi-nude other adults and genital arousal (Rieger et al., 2015). It is also people (i.e., wearing only beachwear) to enhance sexual arousal. an autonomic response that operates outside of conscious con- However, this approach also resulted in variation of the person trol (Laeng et al., 2012; Laeng & Sulutvedt, 2014). Conse- content in terms of body posture, facial expression, eye gaze of quently, pupil dilation might provide a more robust measure of the targets, and so forth. This could have affected eye fixations deviant sexual interest than current measures, which are prone around the scenes and pupillary responses (Birmingham, Bis- to social desirable responding and participant manipulation chof, & Kingstone, 2008). This couldbeaddressedinfuture (for a review, see Kalmus & Beech, 2005). Our data also sug- studies by using more controlled stimuli. As an alternative, such gest that pupillary response could be a sensitive measure at an experiments could compare pupillary responses of hetero-, individual level. This is an important characteristic for imple- homo-, and bisexual male observers. If pupillary response pro- mentation into forensic practice (Gannon et al., 2004). Consid- vides a robust measure of sexual interest, rather than reflecting ering the potential applied value of pupillary responses as a direct other factors within natural scenes, then this should reflect the measure of age-specific sexual interest in this context, further specific sexual interests of these different observer groups. research is warranted. A small set of studies have shown that the pupils appear to be resistant to top-down control, such that observers cannot Acknowledgments This work was supported by a research grant from willingly increase or decrease their pupil size (Laeng et al., the National Organisation for the Treatment of Sexual Abusers (NOTA) to Caoilte O Ciardha and Markus Bindemann. 2012; Laeng & Sulutvedt, 2014). However, the possibility 123 868 Arch Sex Behav (2016) 45:855–870 Open Access This article is distributed under the terms of the Creative arousal in men and women: A meta-analysis. Archives of Sexual Commons Attribution 4.0 International License (http://creativecommons. Behavior, 39, 5–56. doi:10.1007/s10508-009-9556-9. org/licenses/by/4.0/), which permits unrestricted use, distribution, and Dabbs, J. M. (1997). Testosterone and pupillary response to auditory reproduction in any medium, provided you give appropriate credit to the sexual stimuli. Physiology & Behavior, 62, 909–912. doi:10.1016/ original author(s) and the source, provide a link to the Creative Commons S0031-9384(97)00268-0. license, and indicate if changes were made. Ebsworth, M., & Lalumie `re, M. L. (2012). Viewing time as a measure of bisexual sexual interest. Archives of Sexual Behavior, 41, 161–172. doi:10.1007/s10508-012-9923-9. Ellis, C. J. K. (1981). The pupillary reflex in normal subjects.BritishJournal References of Ophtalmology, 65, 754–759. doi:10.1136/bjo.65.11.754. Fromberger, P., Jordan, K., Steinkrauss, H., von Herder, J., Stolpmann, G., Kro ¨ ner-Herwig, B., & Mu ¨ ller, J. L. (2013). Eye movements in Abel, G. G., Huffman, J., Warberg, B., & Holland, C. L. (1998). Visual pedophiles: Automatic and controlled attentional processes while reaction time and plethysmography as measures of sexual interest viewing prepubescent stimuli. Journal of Abnormal Psychology, in child molesters. Sexual Abuse: A Journal of Research and Treat- 122, 587–599. doi:10.1037/a0030659. ment, 10, 81–95. doi:10.1023/A:1022063214826. Fromberger, P., Jordan, K., Steinkrauss, H., von Herder, J., Witzel, J., Akerman, G., & Beech, A. R. (2012). A systematic review of measures of Stolpmann, G.,… Mu ¨ ller, J. L. (2012a). Diagnostic accuracy of eye deviant sexual interest and arousal.Psychiatry,PsychologyandLaw, movements in assessing pedophilia. Journal of Sexual Medicine, 9, 19, 118–143. doi:10.1080/13218719.2010.547161. Attard, J., & Bindemann, M. (2013). Establishing the duration of crimes: 1868–1882. doi:10.1111/j.1743-6109.2012.02754.x. An individual differences and eye tracking investigation into time Fromberger, P., Jordan, K., von Herder, J., Steinkrauss, H., Nemetschek, estimation. Applied Cognitive Psychology, 28, 215–225. doi:10.1002/ R., Stolpmann, G., & Mu ¨ ller, J. L. (2012b). Initial orienting towards acp.2983. sexually relevant stimuli: Preliminary evidence from eye move- Atwood, R. W., & Howell, R. J. (1971). Pupillometric and personality test ment measures. Archives of Sexual Behavior, 41, 919–928. doi:10. score differences of female aggressing pedophiliacs and normals. 1007/s10508-011-9816-3. Psychonomic Science, 22,5–7.doi:10.3758/BF03332522. Gannon, T. A., & O’Connor, A. (2011). The development of the Interest Babchishin, K. M., Nunes, K. L., & Hermann, C. A. (2013). The validity in Child Molestation Scale. Sexual Abuse: A Journal of Research of Implicit Association Test (IAT) measures of sexual attraction to and Treatment, 23, 474–493. doi:10.1177/1079063211412390. children: A meta-analysis. Archives of Sexual Behavior, 42, 487– Gannon, T. A., Ward, T., & Polaschek, D. L. L. (2004). Child sexual 499. doi:10.1007/s10508-012-0022-8. offenders. In M. Connolly (Ed.), Violence in society: New Zealand Bergamin, O., & Kardon, R. H. (2003). Latency of the pupil light reflex: perspectives (pp. 31–47). Christchurch: Te Awatea Press. Gress, C. L. (2005). Viewing time measures and sexual interest: Another Sample rate, stimulus intensity, and variation in normal subjects. piece of the puzzle. Journal of Sexual Aggression, 11, 117–125. Investigative Ophthalmology & Visual Science, 44, 1546–1554. doi:10.1080/13552600500063666. doi:10.1167/iovs.02-0468. Gress, C. L., Anderson, J. O., & Laws, D. R. (2013). Delays in attentional Bernick, N., Kling, A., & Borowitz, G. (1971). Physiological differen- processing when viewing sexual imagery: The development and tiation of sexual arousal and anxiety. Psychosomatic Medicine, 33, comparison of two measures. Legal and Criminological Psychol- 341–352. doi:10.1097/00006842-197107000-00004. ogy, 18, 66–82. doi:10.1111/j.2044-8333.2011.02032.x. Bindemann, M., Burton, A. M., Langton, S. R. H., Schweinberger, S. R., Hall, C., Hogue, T., & Guo, K. (2011). Differential gaze behavior towards & Doherty, M. J. (2007). The control of attention to faces. Journal sexually preferred and non-preferred human figures. Journal of Sex of Vision, 7(10):15, 1–8. doi:10.1167/7.10.15. Research, 48, 461–469. doi:10.1080/00224499.2010.521899. Bindemann, M., Scheepers, C., & Burton, A. M. (2009). Viewpoint and Hamel, R. F. (1974). Female subjective and pupillary reaction to nude center of gravity affect eye movements to human faces. Journal of male and female figures. Journal of Psychology, 87, 171–175. Vision, 9(2):7, 1–16. doi:10.1167/9.2.7. doi:10.1080/00223980.1974.9915687. Bindemann, M., Scheepers, C., Ferguson, H. J., & Burton, A. M. (2010). Harris, G. T., Rice, M. E., Quinsey, V. L., & Chaplin, T. C. (1996). Face, body, and centre of gravity mediate person detection in natural Viewing time as a measure of sexual interest among child molesters scenes. Journal of Experimental Psychology: Human Perception and normal heterosexual men. Behaviour Research and Therapy, and Performance, 36, 1477–1485. doi:10.1037/a0019057. Birmingham, E., Bischof, W. F., & Kingstone, A. (2008). Social attention 34, 389–394. doi:10.1016/0005-7967(95)00070-4. and real-world scenes: The roles of action, competition and social Henderson, J. M. (2003). Human gaze control during real-world scene content. The Quarterly Journal of Experimental Psychology, 61, perception. TrendsinCognitive Sciences, 7, 498–504. doi:10.1016/ 986–998. doi:10.1080/17470210701410375. j.tics.2003.09.006. Bradley, M. M., Miccoli, L., Escrig, M. A., & Lang, P. J. (2008). The pupil as Hess, E. H., Seltzer, A. L., & Shlien, J. M. (1965). Pupil response of a measure of emotional arousal and autonomic activation.Psychophys- hetero- and homosexual males to pictures of men and women: A iology, 45, 602–607. doi:10.1111/j.1469-8986.2008.00654.x. pilot study. Journal of Abnormal Psychology, 70, 165–168. doi:10. Calvo, M. G., & Lang, P. J. (2004). Gaze patterns when looking at emotional 1037/h0021978. pictures: Motivationally biased attention. Motivation and Emotion, 28, Hewig, J., Trippe, R. H., Hecht, H., Straube, T., & Miltner, W. H. (2008). 221–243. Gender differences for specific body regions when looking at men Chivers, M. L. (2005). A brief review and discussion of sex differences in and women. Journal of Nonverbal Behavior, 32, 67–78. doi:10.1007/ s10919-007-0043-5. the specificity of sexual arousal. Sexual and Relationship Therapy, Isaacowitz, D. M. (2006). Motivated gaze: The view from the gazer. Current 20, 377–390. doi:10.1080/14681990500238802. Directions in Psychological Science, 15, 68–72. doi:10.1111/j.0963- Chivers, M. L., Rieger, G., Latty, E., & Bailey, J. M. (2004). A sex difference 7214.2006.00409.x. in the specificity of sexual arousal. Psychological Science, 15, 736– Israel, E., & Strassberg, D. S. (2009). Viewing time as an objective measure 744. doi:10.1111/j.0956-7976.2004.00750.x. ` of sexual interest in heterosexual men and women. Archives of Sexual Chivers, M. L., Seto, M. C., Lalumiere, M. L., Laan, E., & Grimbos, T. Behavior, 38, 551–558. doi:10.1007/s10508-007-9246-4. (2010). Agreement of self-reported and genital measures of sexual 123 Arch Sex Behav (2016) 45:855–870 869 Jenkins, R., Lavie, N., & Driver, J. (2003). Ignoring famous faces: Cate- Mustanski, B., Chivers, M. L., & Bailey, J. M. (2002). A critical review of gory-specific dilution of distractor interference. Perception and Psy- recent biological research on human sexual orientation. Annual Review chophysics, 65, 298–309. doi:10.1037/a0020733. of Sex Research, 13, 89–140. doi:10.1080/10532528.2002.10559803. ¨ ¨ Kalmus, E., & Beech, A. R. (2005). Forensic assessment of sexual inter- Nummenmaa, L., Hietanen, J. K., Santtila, P., & Hyona, J. (2012). Gen- est: A review. Aggression and Violent Behavior, 10, 193–217. der and visibility of sexual cues influence eye movements while doi:10.1016/j.avb.2003.12.002. viewing faces and bodies. Archives of Sexual Behaviour, 41, 1439– Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual behavior 1451. doi:10.1007/s10508-012-9911-0. in the human male. Philadelphia: W. B. Saunders. O Ciardha, C., & Gormley, M. (2012). Using a pictorial-modified stroop Kinsey, A. C., Pomeroy, W. B., Martin, C. E., & Gebhard, P. H. (1953). task to explore the sexual interests of sexual offenders against Sexual behavior in the human female. Philadelphia: W. B. Saunders. children. Sexual Abuse: A Journal of Research and Treatment, 24, Kohn, M., & Clynes, M. (1996). Color dynamics of the pupil. Annals of 175–197. doi:10.1177/1079063211407079. the New York Academy of Sciences, 156, 931–950. doi:10.1111/j. O Ciardha, C., & Gormley, M. (2013). Measuring sexual interest using a 1749-6632.1969.tb14024.x. pictorial modified Stroop task, a pictorial Implicit Association Test Laeng, B., & Falkenberg, L. (2007). Women’s pupillary responses to sex- and a Choice Reaction Time task. Journal ofSexual Aggression, 19, ually significant others during the hormonal cycle. Hormones and 158–170. doi:10.1080/13552600.2012.677486. Behavior, 52, 520–530. doi:10.1016/j.yhbeh.2007.07.013. Price, S. A., & Hanson, K. R. (2007). A modified Stroop task with sexual Laeng, B., Sirois, S., & Gredeba ¨ck, G. (2012). Pupillometry: A window offenders: Replication of a study. Journal of Sexual Aggression, 13, to the preconscious? Perspectives on Psychological Science, 7, 18– 203–216. doi:10.1080/13552600701785505. 27. doi:10.1177/1745691611427305. Quinsey, V. L., Ketsetzis, M., Earls, C., & Karamanoukian, A. (1996). View- Laeng, B., & Sulutvedt, U. (2014). The eye pupil adjusts to imaginary ing time as a measure of sexual interest. Ethology and Sociobiology, 17, light. Psychological Science, 25, 188–197. doi:10.1177/0956797 341–354. doi:10.1016/S0162-3095(96)00060-X. 613503556. Rayner, K. (1998). Eye movements in reading and information process- Laws, D. R., & Gress, C. L. (2004). Seeing things differently: The view- ing: 20 years of research. Psychological Bulletin, 124, 372–422. ing time alternative to penile plethysmography. Legal and Crimi- doi:10.1037/0033-2909.124.3.372. nological Psychology, 9, 183–196. doi:10.1348/1355325041719 Rieger, G., Cash, B. M., Merrill, S. M., Jones-Rounds, J., Dharmavaram, 338. S. M., & Savin-Williams, R. C. (2015). Sexual arousal: The corre- Laws, D. R., & O’Donohue, W. T. (Eds.). (2008). Sexual deviancy: Theory, spondence of eyes and genitals. Biological Psychology, 104, 56–64. assessment, and treatment (2nd ed.). New York: The Guildford Press. doi:10.1016/j.biopsycho.2014.11.009. Lippa, R. A. (2006). Is high sex drive associated with increased sexual Rieger, G., & Savin-Williams, R. C. (2012). The eyes have it: Sex and attraction to both sexes? It depends on whether you are male or sexual orientation differences in pupil dilation patterns. PLoS One, female. Psychological Science, 17, 46–52. doi:10.1111/j.1467- 7, e40256. doi:10.1371/journal.pone.0040256. 9280.2005.01663.x. Rupp, H. A., & Wallen, K. (2007). Sex differences in viewing sexual Lippa, R. A. (2007). The relation between sex drive and sexual attraction stimuli: An eye-tracking study in men and women. Hormones and to men and women: A cross-national study of heterosexual, bisexual Behavior, 51, 524–533. doi:10.1016/j.yhbeh.2007.01.008. and homosexual men and women. Archives of Sexual Behavior, 36, Rupp, H. A., & Wallen, K. (2008). Sex differences in response to visual 209–222. doi:10.1007/s10508-006-9146-z. sexual stimuli: A review. Archives of Sexual Behavior, 37, 206– Lippa, R. A. (2012). Effects of sex and sexual orientation on self- 218. doi:10.1007/s10508-007-9217-9. reported attraction and viewing times to images of men and women: Schmidt, G. (1975). Male-female differences in sexual arousal and behavior Testing for category specificity. Archives of Sexual Behavior, 41, during and after exposure to sexual explicit stimuli. Archives of Sexual 149–160. doi:10.1007/s10508-011-9898-y. Behavior, 4, 353–365. doi:10.1007/BF01541721. Lippa, R. A., Patterson, T. M., & Marelich, W. D. (2010). Looking at and Scott, T. R., Wells, W. H., Wood, D. Z., & Morgan, D. I. (1967). Pupillary longing for male and female‘‘swimsuit models’’: Men are much more response and sexual interest re-examined. Journal of Clinical Psy- category specific than women.SocialPsychologicalandPersonality chology, 23, 433–438. doi:10.1002/1097-4679(196710). Science, 1, 238–245. doi:10.1177/1948550609359814. Sell, R. L. (1997). Defining and measuring sexual orientation: A review. Lobato-Rinco ´ n, L.-L., del Cabanillas-Campos, M. C., Bonnin-Arias, C., Archives of Sexual Behaviour, 26, 643–658. doi:10.1023/A:10245 Chamorro-Gutie ´rrez, E., Murciano-Cespedosa, A., & Sa ´nchez- 28427013. Ramos Roda, C. (2014). Pupillary behavior in relation to wave- Snowden, R. J., Craig, R. L., & Gray, N. S. (2011). Indirect behavioral length and age. Frontiers in Human Neuroscience, 8, 1–8. doi:10. measures of cognition among sexual offenders. Journal of Sex 3389/fnhum.2014.00221. Research, 48, 192–217. doi:10.1080/00224499.2011.557750. Lykins, A. D., Meana, M., & Strauss, G. P. (2006). Detection of differ Steinman,D.L., Wincze,J.P., Sakheim, B. A.,Barlow, D. H.,& ential viewing patterns to erotic and non-erotic stimuli using eye- Mavissakalian, M. (1981). A comparison of male and female patterns tracking methodology. Archives of Sexual Behavior, 35, 219–228. of sexual arousal. Archives of Sexual Behavior, 10, 529–547. doi:10. doi:10.1007/s10508-006-9065-z. 1007/BF01541588. Lykins, A. D., Meana, M., & Strauss, G. P. (2008). Sex differences in Suschinsky, K. D., Elias, L. J., & Krupp, D. B. (2007). Looking for Ms. visual attention to erotic and non-erotic stimuli. Archives of Sexual Right: Allocating attention to facilitate mate choice decisions. Behavior, 37, 219–228. doi:10.1007/s10508-007-9208-x. Evolutionary Psychology, 5, 428–441. Mitchell, R. C., & Galupo, M. P. (2015). Interest in child molestation among Suschinsky, K. D., & Lalumiere, M. L. (2012). Is sexual concordance a community sample of men sexually attracted to children. Journal of related to awareness of physiological states? Archives of Sexual Sexual Aggression.doi:10.1080/13552600.2015.1056263. Behavior, 41, 199–208. doi:10.1007/s10508-012-9931-9. Mokros, A., Dombert, B., Osterheider, M., Zappala `, A., & Santtila, P. Suschinsky, K. D., Lalumie `re, M. L., & Chivers, M. L. (2009). Sex differences (2010). Assessment of pedophilic sexual interest with attentional in patterns of genital sexual arousal: Measurement artifacts or true choice reaction time task. Archives of Sexual Behavior, 39, 1081– phenomena? Archives of Sexual Behavior, 38, 559–573. doi:10.1007/ 1090. doi:10.1007/s10508-009-9530-6. s10508-008-9339-8. 123 870 Arch Sex Behav (2016) 45:855–870 Tanner, J. (1978). Fetus into man: Physical growth from conception to Wright, L. W., & Adams, H. E. (1994). Assessment of sexual preference maturity. Cambridge, MA: Harvard University Press. using a choice-reaction time task. Journal of Psychopathology and VanRullen, R. (2006). One second glance: Still no high-level pop-out Behavioral Assessment, 16, 221–231. doi:10.1007/BF02229209. effect for faces. Vision Research, 46, 3017–3027. doi:10.1016/j. Zuckerman, M. (1971). Physiological measures of sexual arousal in the visres.2005.07.009. human. Psychological Bulletin, 75, 297–329. doi:10.1037/h0030923. Wincze, J. P., & Qualls, C. B. (1984). A comparison of structural patterns of sexual arousal in male and female homosexuals. Archives of Sexual Behaviour, 13, 361–370. doi:10.1007/BF01541908.

Journal

Archives of Sexual BehaviorPubmed Central

Published: Feb 8, 2016

There are no references for this article.