C 2006) Archives of Sexual Behavior (! DOI: 10.1007/s10508-005-9000-8
Pictorial Cues and Sexual Desire: An Experimental Approach Helen M. Conaglen, Ph.D.1,3 and Ian M. Evans, Ph.D.2 Received February 23, 2004; revision received June 28, 2005; accepted September 14, 2005
The cognitive processing of sexual and non-sexual pictorial stimuli was examined to see whether picture rating and recognition tasks have potential utility as a means of assessing levels of sexual desire. Previous research has revealed slower responding to sexual compared to neutral semantic cues in persons with lower self-reported sexual desire. The present study investigated whether sexual pictorial cues evoked a similarly slower responding in people reporting low sexual desire compared to other individuals. A total of 136 participants completed two self-report measures of sexual desire (the Hurlbert Index of Sexual Desire and the Sexual Desire Inventory) before carrying out tasks involving affective ratings and recognition memory for pictorial stimuli. Participants were classified into relatively low, average, and higher groups on the basis of their scores on the desire measures. The stimuli were selected from the International Affective Picture System, and the tasks included (1) rating the valence, arousal, and sense of dominance or control for each picture, (2) recognition of previously seen images, and (3) a second rating of pictures viewed earlier. Level of sexual desire did not influence responding in the male participants. Female participants with lower sexual desire rated sexual images less pleasant and less arousing than the other participants, and completed picture recognition tasks more quickly. Sexual desire levels significantly influenced the interest ratings women gave to sexual pictures. We also found sexual content induced delays. These delays were not significantly different among desire groups. The variation in responding linked to levels of sexual desire in women suggests that more investigation of this methodology in a clinically diagnosed population might contribute to an understanding of low desire, and help design interventions addressing distress due to lack of sexual desire. KEY WORDS: sex; information processing; sexual desire; sexual content induced delay.
INTRODUCTION Sexual desire has been the focus of ongoing debate over ways of conceptualizing the construct for clinical and research purposes (see, e.g., Basson, 2001, 2002; Beck, 1995; Heiman, 2001; Kaplan, 1977; Leiblum & Rosen, 1988; Levine, 2003; Schnarch, 1991). No one current definition appears to suit all needs, but it can be useful to view desire as a “subjective feeling state that may be triggered by both internal and external cues, 1 Clinical
Research Laboratory, Psychology Department, University of Waikato, Hamilton, New Zealand. 2 School of Psychology, Massey University, Palmerston North, New Zealand. 3 To whom correspondence should be addressed at The Psychology Centre, P.O. Box 4423, Hamilton, New Zealand; e-mail:
[email protected].
which may or may not result in overt sexual behavior” (Leiblum & Rosen, 1988, p. 5). This definition allows for both psychological and physiological components to be involved in any responses detected, but for the psychological to be specific to desire, and the physiological to be associated with arousal. Inevitably, therefore, biases putatively accompany subjective methods of measuring aspects of sexual desire, especially on questionnaires. Opinions range from the view that self-report data are suspect to the alternative suggestion that the reverse is the case (for a review, see Catania, Gibson, Chitwood, & Coates, 1990). In any event, self-report remains the methodology that is most heavily relied upon in research on sexual desire. In an explanation of how sexual dysfunctions develop and are maintained, McCabe (1991) has suggested that
C 2006 Springer Science+Business Media, Inc. 0004-0002/06 !
Conaglen and Evans various factors may contribute to a person’s evaluation of a sexual event. Factors proposed include intergenerational, individual, and relationship elements that interact to determine the way in which a sexual event is experienced. Bass and Walen (1986) suggested that, in the sexual arena, an inability to detect stimuli, incorrect labeling of them, or misattribution of them, might all significantly impair performance. The detection of a stimulus or cue that may or may not be sexual requires that the individual pay attention to the stimulus, appraise it, process the information received, and decide whether or not to respond.
Emotional Responding to Pictures and Words In the sexual arena, as in other domains, the components of an emotional response to a stimulus can be described in terms of the semantic (or content) knowledge about that stimulus, the response or efferent outflow information (or arousal), and the meaning given to the stimulus derived from a synthesis of the content and affect information. These components, termed action dispositions by Lang (1979), determine the individual’s overt response to the stimulus. Lang’s model of emotion also describes affective evaluation along three dimensions: valence, arousal, and dominance. These capture most variation in emotional meaning reported by individuals. In other words, people’s experience of emotion can be recorded in a three dimensional plot of the pleasure (or displeasure) involved, the arousal (or lack of it) experienced, and the sense of control or domination generated in that emotional event. Lang’s explanation of the connections between emotion and cognition provides a model of how both interact to produce response dispositions in individuals. Lang has generated the International Affective Picture System (IAPS; Lang, Bradley, & Cuthbert, 1996) by assessing emotional responding to pictorial stimuli. The system consists of an extensive series of over 400 pictures of people, places, and objects representing different affective possibilities. Each picture has been assigned specific combinations of valence (or pleasure), arousal, and dominance (or control) values during an ongoing process of rating by participants in laboratory research (Lang, 1995; Lang, Bradley, & Cuthbert, 1995). These values have been shown to hold for different cultures and the stimuli form a system useful for investigating responding in varying groups of people (Lang, 1995). The previous ratings of all the stimuli allow the choice of stimuli matched with respect to valence and arousal,
thus generating a potential test of the influence of sexual, compared to non-sexual, themes. A number of studies have shown that the salience of an emotional stimulus will affect the manner in which a person responds to that stimulus. One example of how such responses can be generated and objectively quantified is the emotional Stroop test, which allows responding to be investigated objectively, since the emotional bias is inferred from interference in cognitive processing. In this task words, which vary according to the theme of the psychopathology being investigated, appear in different colors and participants are asked to name the colors in which the words are printed. Responses are usually slower when the word is associated with concerns relevant to the person’s clinical condition. The emotional Stroop test has been applied in various areas of psychopathology (see, e.g., Williams, Mathews, & McLeod, 1996), including general anxiety disorder, panic, phobias, obsessivecompulsive disorder, posttraumatic stress disorder, and depression. Studies of sexuality related issues are less common (but see Sipski, Alexander, & Rosen, 1977, where the Stroop Color-Word Test was used as a distractor task in a study of female sexual arousal). We could find no such studies specifically relating to low sexual desire. Other examples of protocols allowing objective measurement of responding to different types of stimuli include verbal tasks, in which participants record their decision about whether a letter-string is a real word or not as rapidly as possible. Studies using such lexical decision making tasks have found sex differences in response to a range of words or sentences (e.g., Geer, 1996; Geer & Bellard, 1996; Geer, Judice, & Jackson, 1994; Geer & Manguno-Mire, 1996; Geer & McGlone, 1990; Geer & Melton, 1997). A recurring pattern in these studies is a phenomenon called Sexual Content Induced Delay (SCID). SCID refers to the apparent delay in responding to sexual as distinct from non-sexual or neutral words. While the studies cited have investigated gender differences, none indicated whether participants were experiencing normal sexual function. The potential for a person’s responses to be different if they were experiencing sexual desire problems might suggest these procedures could be used as a means of investigating aspects of such problems. Because desire can be construed as the outcome of evaluating stimuli in order to determine a suitable response, it has been suggested that individuals with varying levels of sexual desire might respond to sexual stimuli differently (Conaglen, 2004). Specifically, it might be expected that persons with relatively low levels of sexual interest or desire would respond more slowly to sexual words than other people.
Pictorial Cues and Sexual Desire In a partial replication of Geer and Ballard’s (1996) study, Conaglen (2004) examined whether the SCID phenomenon was expressed differentially in people who varied in levels of self-reported sexual desire. It was found that subjects with relatively low self-reported sexual desire showed a tendency to respond more slowly to sexual than to neutral words. One suggested explanation for this finding was that those with lower sexual desire might have deemed sexual stimuli as threatening or having a more negative valence, and thus took longer to respond. This accords with the findings that responses on Stroop tests by people with varying emotional difficulties were slower to words related to those difficulties (Williams, Mathews, & MacLeod, 1996). Conaglen (2004) suggested that further examination of the responding of those with varying levels of desire with respect to stimuli in another modality (e.g., pictures) might help to better understand the relation between responding to external sexual cues and relative level of sexual desire.
Cognitive Processing in Different Modalities When comparing responses to different types of stimuli, one needs to anticipate the way people respond cognitively to specific tasks involving words and pictures. In the literature describing cognitive appraisal of cues with respect to implicit and explicit memory, Mori and Graf (1996) differentiated between unitizing and elaborating processes in cognition. Unitizing is the cognitive activity by which separate components of an item are combined so they function as one unit (e.g., letters of a word). An example of this unitizing process in an experimental context is decision-making about whether letters on a screen form a real word or not (as required in semantic tasks; see Conaglen, 2004). Elaborating is the cognitive activity in which one item or event is associated with other items or events. In the present study, a picture recognition task required participants to decide whether they had seen the picture before or not, a query that required them to elaborate, or associate, the item with another item previously viewed. Implicit memory tests involve unitizing activity and explicit tests depend primarily on elaborative processing. In the study examining decisionmaking about sexual and non-sexual words implicit memory was involved, whereas the task requiring recognition of sexual and non-sexual pictures involved explicit aspects of memory. Spiering, Everaerd, and Janssen (2003) considered the involvement of implicit and explicit memory in responses to primed picture tasks. They suggested that, in accordance with Zajonc’s (1984) model of the importance
of affect, sexual meaning can be activated implicitly, that is, an automatic physiological response can be generated by the emotional content of images. Spiering et al. further described subjective responding as being based on explicit or declarative memory and involving a conscious elaboration of emotional information. In the first of two priming studies, they found responding to sexual stimuli was affected by sexual primes, but responding to plant stimuli was not affected by plant primes. They then sought to examine whether these differences relied on the implicit or the explicit processing of primes, by using the same stimuli in a protocol that allowed conscious processing of the primes. In this second study, they found that sexual primes influenced decisions about sexual targets, slowing them down (Spiering, Everaerd, & Elzinga, 2002). These studies have been followed by an investigation of the conscious processing of sexual information, also in a primed situation, by students presumed to be experiencing normal sexual function (Spiering, Everaerd, & Laan, 2004). This study found that priming contributed to an effect similar to SCID using pictorial stimuli when a cognitive task was involved. There were no significant differences between men and women on this task, but when an affective task, presumed to involve explicit or conscious memory was undertaken, gender differences were found in participants’ ratings.
Processing Affected by Fear Emotion processing investigations using valence and arousal ratings from the IAPS have linked variations in responding to slides to some clinical conditions. Studies have shown that people with phobias respond to their feared stimuli with a more potentiated startle response and avoided those stimuli more than controls (Hamm, Cuthbert, Globisch, & Vaitl, 1997; Tolin, Lohr, Lee, & Sawchuk, 1999). The fearful individuals’ valence and arousal ratings of the pictures of feared objects were also consistent with their phobias and significantly different from those of the controls. While most studies have utilized psychophysiology involving startle probes, heart rate, and skin conductance monitoring, others have established that the emotional rating of pictures corresponds with psychophysiological findings (Greenwald, Cook, & Lang, 1989). Given this correspondence, it would seem feasible to use such ratings in lieu of more intrusive types of psychophysiological measurement. To date, such studies have not reported on sexually related images and responding in people with sexual difficulties. One exception is a report that tied responding to the IAPS stimuli with emotion probe investigations to the
Conaglen and Evans erotophilic/erotophobic scales of the Sexual Opinion Survey (Haerich & Khoury, 1994). In this study, erotophilic persons responded to nude pictures as though those slides were appetitive stimuli, while the erotophobic participants reacted as though the slides were aversive. Low sexual desire difficulties have been characterized as the inhibition of responses to sexual stimuli or, in more extreme cases, as phobic responses to things sexual. If either of these conceptualizations is appropriate, one should be able to show differential responding in persons with lower levels of sexual desire. If inhibition were the more useful explanation, responses in persons with low sexual desire would be faster than in others. If phobic mechanisms are involved, one would expect more extreme low valence and high arousal ratings in those with lower levels of sexual desire (more fearful of sexual stimuli), responses similar to those found in the studies of animal fearful and spider phobic subjects discussed by Hamm et al. (1997). Participants in that study also showed a diminishing of arousal when repeated exposure to threat pictures was experienced. The present study sought to develop further the concepts from our study of responding to semantic stimuli by people with differing levels of self-reported sexual desire (Conaglen, 2004). In that study, the slower responding of persons with lower levels of self-reported sexual desire was thought to be explained by their view of sexual words as more negative than the contrast words, and thus their response times were somewhat slower. In the present study, we investigated whether variation in participants’ responding to different types of pictorial cues would be associated with their levels of sexual desire. Using a grouping based on self-reported levels of sexual desire, we investigated several questions: (1) Do relatively low desire persons rate the valence, arousal, and dominance for varying types of pictures differently than others? Specifically, does the low desire group give the sexual pictures lower valence, higher arousal, and lower dominance ratings than the other groups? (2) When pictures are viewed for a second time, is there a change in the valence, arousal, or dominance ratings, and how is this related to desire levels? We would expect that valence and arousal ratings would be less extreme in all picture types when pictures were rated a second time, particularly if those pictures were salient to the person’s fears. (3) Do people with relatively low levels of sexual desire complete a picture recognition task more slowly than other groups? Having found slower reactions by those with lower sexual desire levels to sexual word tasks in our previous study, a finding in this direction would support the idea that attentional bias relating to sexual pictures, viewed as threatening, might cause a slower response in lower desire
people. (4) Given the differences in SCID found in a previous study using semantic stimuli (Conaglen, 2004), we also sought to understand if similar patterns of greater delay by those with low sexual desire would be found with pictorial stimuli. (5) Are the interest ratings given sexual pictures by people with low sexual desire lower than those assigned by the other participants in the study?
METHOD Participants A total of 54 men and 82 women completed the study, which had been approved by the institutional ethics committee. Just over half of the participants were volunteers from within the university department in which the study was conducted, while the rest were from the local community. We attempted to include a wider range of ages in this study than is usually the case in university studies, because of the age ranges at which low sexual desire is said to be a problem clinically (Donahey & Carroll, 1993). Our sample ranged in age from 17 to 61 years (M = 30.2; SD = 11.6). The mean age of the men was 30.3 years (SD = 12.2) and the mean age of the women was 30.2 years (SD = 11.2). Because facility with a language potentially affects questionnaire completion, only data from participants with English as their first language were analyzed. Most participants categorized themselves as Pakeha/European with respect to ethnicity (n = 110), 13 described their ethnicity as Mixed, and 10 as Maori, while 3 endorsed the “Other” category. All participants were aware the study involved sexual material and gave written informed consent prior to participation. The volunteers from the student population (n = 70) received course credit for completing the protocol. Other participants (n = 66) who responded to advertisements were offered a gas (petrol) voucher as a token of thanks for their time. Pictorial Stimuli Color slides representing four types of images were chosen from the IAPS (Lang et al., 1996). Those chosen were 19 “high valence” sexual slides, 19 “high valence” non-sexual slides, 11 neutral valence slides, and 12 “low valence” unpleasant slides (slide numbers given in Appendix). “High valence” was the more enjoyable end of this spectrum. The high valence slides were chosen so that the U.S. ratings for each were closely matched. Sexual slides were distinguished from non-sexual in the
Pictorial Cues and Sexual Desire laboratory by fellow researchers prior to the calibration studies carried out by the first author. These studies (Conaglen, 1999) established the correspondence of New Zealand student ratings for the images with U.S. student ratings published by Lang et al. (1995), and confirmed the categorization of the slides. A further selection comprised of pictures from each of these groups was chosen for the first part of the protocol, a slide-rating task. These 30 slides were chosen by selecting every second picture when valence ratings were ranked numerically, thus balancing the mean ratings for slides in each portion of the protocol. This sub-selection consisted of 10 high valence, sexual content slides (e.g., couples or single persons, nude or semi-clothed); nine high valence non-sexual slides (e.g., cartoon characters, ski jumps, spectacular scenery); five neutral slides showing common objects such as a plate, towel or hairdryer; and six unpleasant low valence slides (e.g., mutilation, war, or famine victims). All slides were presented in a predetermined pseudo-random order in each part of the protocol. The IAPS slides were screened in 800 by 600 pixel format to allow for the greatest possible definition of the stimuli. Apparatus Participants were tested on IBM-compatible computers equipped with 14-inch color monitors, and standard two button mouse devices. Each of the mouse buttons was labeled “yes” or “no” to assist participants during the rapid response sections of the protocol. The protocol was run on a program that enabled millisecond timing of the tasks involved, and direct writing of timing and rating data into files for later analysis. Each of the two computers used was set up in a small windowless laboratory room, which was occupied only by the participant once the procedure had been explained. Picture ratings were recorded on the Self-Assessment Manikin (SAM) scale as used by Lang et al. (1995). This figure was incorporated into the computer program written for the project. Ratings were made by the participants sliding the cursor along the horizontal rating bar under each scale until it corresponded with the intensity of feeling they wished to record. The affective valence dimension showed a figure that ranged from happy to unhappy representing the pleasant (9) to unpleasant (1) range for that rating. The arousal dimension sought a rating between excited (9) and calm (1). The dominance or control dimension ranged from controlled (1) to in control (9) with the manikin figure growing from small in size, depicting the feeling of being dominated, to a large version for the in control end of the scale. These descriptors
were in accordance with the manual instructions for ratings using the SAM with respect to the IAPS (Lang et al., 1995). Each dimension was scored on a 9-point scale. Scores from this process were directly entered into individual data files for each participant. Participants were told to rate each picture according to how they felt while viewing the image. Procedure The experimental procedure was explained to each participant individually. This process was assisted by the use of a series of instruction screens to ensure no important aspects of the protocol were omitted from the explanation. These screens were based on the instructions supplied in the Technical Manual of the IAPS. Each participant then worked alone completing the tasks in the sequence that follows. Picture Rating of 30 IAPS Slides Participants were shown a series of pictures and asked to rate each one according to how it made them feel using the SAM rating screen to record affective valence, arousal, and dominance ratings. Questionnaires Participants then completed self-report questionnaires assessing sexual desire. These were the Hurlbert Index of Sexual Desire (HISD; Apt & Hurlbert, 1992) and the Sexual Desire Inventory-2 (SDI-2; Spector, Carey, & Steinberg, 1996). The HISD was developed for use with women, but has also been utilized in studies of dyads (Apt, Hurlbert, Sarmiento, & Hurlbert, 1996), and has been shown to reflect desire changes in men undergoing medical treatment for hypogonadism, which affects sexual desire levels in most patients (Conaglen, Hunt, & Conaglen, 2003). Participants respond on a scale from 0 (all of the time) to 4 (never) to a series of 25 statements such as “I have a strong sex desire,” and “I try to avoid situations that will encourage my partner to want sex.” Some questions were reverse scored to eliminate response bias. Total score, from 0 (low) to 100 (high), gives an estimate of sexual desire. The authors of the test stated that “it has evidenced good internal consistency (alpha = .89), test-retest reliability (r = .86 across two weeks), and construct validity” (Apt & Hurlbert, 1994). Hurlbert (1993) has also reported that the test was not influenced by social desirability response sets. However,
Conaglen and Evans Sexual Content Induced Delay
there are, at this point in time, no published norms or cut-off points for use of the instrument in clinical settings. The SDI-2 is a 14-item self-report measure that elicits information about the respondent’s dyadic sexual desire and solitary desire characteristics that, when combined, give an overall measure of sexual desire (Spector et al., 1996). It focuses on cognitive rather than behavioral aspects of sexual desire. Items include “When you have sexual thoughts, how strong is your desire to engage in sexual behavior with a partner?” and “How important is it for you to fulfill your sexual desires to behave sexually by yourself?” Each item was endorsed according to intensity of feeling or frequency of occurrence of a thought, activity or feeling (0 to 7 or 8), yielding a possible score range from 0 to 109. The measure was reported to have an internal consistency index of.86 for the dyadic desire scale and .96 for the solitary scale.
In order to understand whether SCID to pictorial stimuli varied according to levels of sexual desire, individuals’ recognition time responses to neutral images were subtracted from those to sexual images and group differences examined by analysis of variance. Viewing Tasks Following the recognition task, participants were able to view the picture for as long as they wished, this viewing time being recorded as a secondary dependent variable. Participants then rated the picture as in the first portion of the experiment, and also gave the image an interest rating. The repeat ratings allowed for a determination of the consistency of ratings given by participants and the extent of habituation to the various stimuli during the experimental procedure. The interest rating involved a choice ranging from 1 (not at all interesting) to 29 (very interesting).
Picture Recognition Participants then carried out further pictorial tasks, involving a memory decision, a self-determined viewing time, and then a rating process using the same screen arrangement as in Part 1 of the experiment. The recognition memory task was generated by adding 31 slides to those viewed in the first portion of the experiment and arranging the two groups in a pseudorandom order that did not allow more than three of any type of stimulus to be viewed in sequence. Thus, the subjects had seen half of the stimuli, but not the rest, and were asked to decide as quickly as possible whether the picture on their screen was one they had seen before or not. Responses were by mouse button press. This recognition variable was a key dependent measure in the study.
Desire Group Categorization Self-report data from the two desire questionnaires were used as a basis for grouping participants according to their relative levels of sexual desire. Scores on the HISD varied from 36 to 100, with the overall mean of 68.8 (SD = 12.2). There were significant sex differences for this measure: t(134) = 4.16, p < .0001, with men reporting a mean of 73.9 (SD = 11.96) and women 65.5 (SD = 11.26). Internal consistency (Cronbach’s alpha) for this measure for the men was .92 and .89 for the women. Scores for the SDI-2 ranged from 36 to 99 out of a possible 109. The overall mean was 64.7 (SD = 14.0).
Table I. Characteristics of Sexual Desire Groups by Sex Hurlbert Index of Sexual Desire Sex Men
Group (n)
Low (n = 15) Average (n = 24) High (n = 15) Women Low (n = 21) Average (n = 38) High (n = 23) a Group b Group
Sexual Desire Inventory
Age
M
SD
M
SD
M
60.53a 75.00a 85.47a 55.00b 64.66b 76.39b
6.90 8.32 6.61 8.07 8.22 7.96
56.87a 72.46a 82.93a 48.05b 59.13b 74.00b
11.12 6.60 7.76 6.74 7.76 12.25
32.85 27.35 32.54 29.44 27.69 35.00
means significantly different, p = .0001. means significantly different, p = .0001.
SD 14.32 10.78 11.70 11.91 9.56 12.09
Pictorial Cues and Sexual Desire Sex differences were significant, t(134) = 4.63, p < .0001, with men recording a mean score of 71.0 (SD = 12.9) and women 60.5 (SD = 13.1). Internal consistency (Cronbach’s alpha) for this measure was .79 for both men and women. Because of these sex differences, the categorization into relatively high, average, and relatively low sexual desire groups was carried out separately by sex. The two desire questionnaire scores were summed and participants categorized on the basis of the quartile splits of that sum. Characteristics of those in each group by sex are given in Table I. The significance of the group differences was tested by a 3 (Desire Group) × 2 (Desire Measure) ANOVA for each sex. For the men, there was a significant main effect of Desire Group, F(2, 51) = 101.31, p < .0001. For the women, there was also a significant main effect of Desire Group, F(2, 79) = 114.63, p < .0001. There was no significant difference in age among the men’s desire groups, F(2, 51) = 1.31. There was, however, significant variation with age in the women’s desire groups, F(2, 79) = 3.27, p < .05. Post-hoc analysis showed that the women in the relatively high desire group (M = 35.00, SD = 11.23) were significantly older (p < .05) than those in the average desire group
(M = 27.69, SD = 9.56). For this reason, age was covaried whenever an analysis of variance was carried out for the women in this study. RESULTS Picture Rating of 30 IAPS Slides All participants rated 30 slides from the IAPS in the initial rating task, and then rated those slides again later in the protocol. The first task ratings were examined for differences associated with desire group and the four picture types. The repeat ratings were also examined for changes in response across the two rating tasks. Ratings by Men Table II shows the mean ratings for each picture type from the first task as a function of sexual desire group. A 3 (Desire Group) × 4 (Picture Type) analysis of variance (ANOVA) of the valence ratings revealed only a significant main effect of Picture Type, F(3,153) = 360.66, p < .0001. Newman-Keuls post-hoc analysis showed that valence ratings for the sexual and non-sexual
Table II. Mean Ratings by Men as Function of Desire Group and Picture Type for First Picture Rating Task Picture types High valence sexual Dimension Affective valence
Arousal
Dominance or control
High valence non-sexual
Neutral
Unpleasant low valence
Sexual desire group
M
SD
M
SD
M
SD
M
Low (n = 15) Average (n = 24) High (n = 15) Combined Low Average High Combined Low Average High Combined
6.9 6.6 6.9 6.8a 5.5 5.5 5.7 5.6c,d 7.0 6.3 6.5 6.6f ,g
1.1 0.7 0.5 0.8 1.5 1.2 1.2 1.3 1.4 1.4 1.8 1.5
6.9 6.7 6.4 6.6a 5.4 5.4 4.8 5.2c,e 6.3 5.8 6.2 6.1g
0.9 1.0 1.3 1.1 1.4 1.4 1.4 1.4 1.2 1.3 1.9 1.5
5.4 5.0 5.3 5.2b 2.8 2.3 2.3 2.5 7.5 7.7 7.6 7.6
1.2 0.5 0.6 0.8 1.5 1.4 1.2 1.4 1.4 1.4 1.9 1.5
2.0 2.0 1.8 2.0 6.1 6.4 6.0 6.2 4.4 3.7 3.8 3.9
Note. Absolute range, 1–9. a Valence rating significantly different to neutral and unpleasant picture valence ratings, ps < .0001. b Valence rating significantly different to unpleasant picture valence rating, p < .0001. c Arousal rating significantly different to neutral picture arousal ratings, ps < .0001. d Arousal rating significantly different to unpleasant picture arousal rating, p < .05. e Arousal rating significantly different to unpleasant picture arousal rating, p < .01. f Dominance rating significantly different to high valence non-sexual picture rating, p < .05. g Dominance rating significantly different to neutral and unpleasant picture ratings, ps < .0001.
SD 0.9 0.8 1.0 0.9 2.1 1.8 1.8 1.9 2.5 1.7 2.1 2.0
Conaglen and Evans Table III. Mean Ratings by Men as Function of Picture Type and Rating Task Picture types High valence sexual Dimension Valence Arousal Dominance or control
High valence non-sexual
Neutral
Unpleasant low valence
Rating task
M
SD
M
SD
M
SD
M
SD
First task Repeat First task Repeat First task Repeat
6.8 6.7 5.6 5.2∗∗∗ 6.6 6.8∗∗
0.8 0.9 1.3 1.4 1.5 1.6
6.6 6.7 5.2 5.1 6.1 6.2
1.1 0.8 1.4 1.4 1.5 1.6
5.2 5.0∗ 2.5 2.2 7.6 7.6
0.8 0.5 1.4 1.1 1.5 1.6
2.0 2.2∗∗ 6.2 5.7∗∗∗ 3.9 4.3∗∗∗
0.9 0.8 1.9 1.9 2.0 2.1
Note. Absolute range, 1–9. Repeat rating significantly different to first rating: ∗ p < .05; ∗∗ p < .01; ∗∗∗ p < .001.
pictures were not significantly different, but those for the unpleasant pictures were lower than all other picture types (all ps < .0001), while ratings for the neutral pictures varied (see Table II for details).
A 3 (Desire Group) × 4 (Picture Type) ANOVA of the arousal ratings showed only a significant main effect of Picture Type F(3,153) = 87.38, p < .0001. Post-hoc analysis showed that the arousal ratings for high valence
Table IV. Mean Ratings by Women as Function of Desire Group and Picture Type for First Picture Rating Task Picture types High valence sexual Dimension Valence
Sexual desire group
Low (n = 21) Average (n = 38) High (n = 23) Combined Arousal Low Average High Combined Dominance or control Low Average High Combined
High valence non-sexual
Neutral
Unpleasant low valence
M
SD
M
SD
M
SD
M
SD
6.4 6.9 7.2a 6.9 5.7 6.0 6.4 6.0b 5.8 6.1 6.5 6.1d,e
0.8 0.9 1.0 1.0 1.2 1.4 1.2 1.3 1.5 1.2 1.6 1.4
6.9 6.7 6.4 6.9 5.7 5.9 5.8 5.8b 5.6 5.6 5.8 5.6e
0.9 1.0 1.3 0.9 1.1 1.4 1.4 1.3 1.3 1.3 1.6 1.4
5.4 5.0 5.3 5.2 3.0 2.7 2.7 2.7c 7.5 7.6 7.3 7.5f
1.2 0.5 0.6 1.1 1.4 1.3 1.5 1.4 1.4 1.3 1.8 1.4
2.0 2.0 1.8 1.6 6.5 7.4 7.3 7.1 2.7 2.6 3.3 2.8
0.9 0.8 1.0 0.9 2.3 1.7 1.4 1.8 1.7 1.4 2.1 1.7
Note. Absolute range, 1–9. a Interaction: Low desire women rated the valence of the sexual pictures significantly lower than high desire women, p = .02. b Main effect of picture type: Arousal rating significantly different from neutral and unpleasant picture arousal ratings, ps < .0001. c Main effect of picture type: Arousal rating significantly different from unpleasant picture arousal rating, p < .0001. d Main effect of picture type: Dominance rating significantly different from high valence non-sexual picture rating, p < .01. e Main effect of picture type: Dominance rating significantly different from neutral and unpleasant picture ratings, ps < .0001. f Main effect of picture type: Dominance rating significantly different from unpleasant picture dominance rating, p < .0001.
Pictorial Cues and Sexual Desire Table V. Mean Ratings by Women for First and Repeat Tasks as Function of Picture Type Picture types High valence sexual Dimension
Rating task
Valence
First Repeat First Repeat First Repeat
Arousal Dominance or control
High valence non-sexual
Neutral
Unpleasant low valence
M
SD
M
SD
M
SD
M
6.9 6.7 6.0 5.8∗∗ 6.1 6.5∗∗∗
.96 1.0 1.3 1.4 1.4 1.3
6.9 6.5∗∗∗ 5.8 5.1∗∗∗ 5.6 6.1∗∗
.93 .84 1.3 1.3 1.4 1.5
5.2 4.9∗ 2.7 2.7 7.5 7.3
1.1 1.0 1.4 1.3 1.4 1.3
1.6 1.7∗ 7.1 7.0 2.8 3.1∗
SD .90 .82 1.8 1.7 1.7 1.7
Note. Absolute range, 1–9. Repeat rating significantly different from first rating: ∗ p < .05; ∗∗ p < .01; ∗∗∗ p < .001.
sexual and non-sexual pictures did not differ significantly, but the ratings for the unpleasant pictures were higher than those for other picture types, while those for the neutral pictures were significantly lower than ratings for the other picture types, all ps < .0001 (see Table II for details). A 3 (Desire Group) × 4 (Picture Type) ANOVA of the dominance ratings revealed only a main effect of Picture Type, F(3,153) = 80.74, p < .0001. Post-hoc analysis showed that dominance ratings for the sexual pictures were higher than those for the non-sexual (p < .05), and unpleasant pictures (p < .05), and lower than those for the neutral pictures (p < .0001). Details of comparisons are shown in Table II. Later in the protocol, participants rated the pictures again. Mean ratings for this repeat rating task were compared with those for the first viewing by t-test. On the second occasion, there were significant changes in the valence ratings for all except the sexual pictures. Ratings of arousal and dominance for the sexual pictures and the unpleasant low valence pictures also changed significantly. Ratings and levels of significance of these changes can be seen in Table III. The second set of ratings was examined using a 3 (Desire Group) × 4 (Picture Type) ANOVA of valence, arousal, and dominance, but desire group had no significant effect on the ratings during the repeat task.
Ratings by Women Table IV shows the mean ratings for each picture type from the first task as a function of sexual desire group. A 3 (Desire Group) × 4 (Picture Type) analysis of covariance (ANCOVA) of the valence ratings revealed a significant main effect of Picture Type, F(3,237) = 519.95, p < .0001,
and a significant Desire Group × Picture Type interaction, F(6,237) = 2.70, p < .05. Newman-Keuls post-hoc analysis of the interaction showed that the Low Desire women’s ratings of the sexual pictures were significantly lower than those of the High Desire women, p < .05 (see Table IV for details). A 3 (Desire Group) × 4 (Picture Type) ANCOVA of the arousal ratings showed only a significant main effect of Picture Type, F(3,237) = 181.18, p < .0001. Post-hoc analysis showed that the ratings of the unpleasant pictures were higher than those for all other picture types (all ps < .0001). Ratings for the sexual pictures were not different to the non-sexual pictures but higher than those for the neutral pictures, p < .0001 (see Table IV for details). A 3 (Desire Group) × 4 (Picture Type) ANCOVA of the dominance ratings revealed only a main effect of Picture Type, F(3,237) = 199.88, p < .0001. Post-hoc analysis showed dominance ratings for the unpleasant picture types were significantly lower than those for all other types (all ps < .0001). In addition, dominance ratings for the sexual pictures were higher than for the nonsexual pictures (p < .01), and lower than for the neutral pictures, p < .0001 (see Table IV for details). We also examined the repeat ratings by the women for significant changes from the first task ratings, using t-tests. The ratings are shown in Table V. There were significant changes in the valence ratings for all but the sexual pictures (see Table V for p values). Arousal ratings were significantly different for the sexual and non-sexual picture types, but not the neutral or unpleasant picture types. Dominance ratings changed significantly for all but the neutral picture types. The women’s second set of ratings was examined with a series of 3 (Desire Group) × 4 (Picture Type) ANCOVAs. For the valence ratings, there was a significant
Conaglen and Evans Table VI. Women’s Valence Ratings for Sexual Pictures by Task and Sexual Desire Group Dimension
Sexual desire group Rating task
Valence
Low (n = 21) Average (n = 38) High (n = 23)
First task Repeat First task Repeat First task Repeat
M
SD
6.4a 6.1b 6.9 6.8c 7.2 7.3
0.8 0.8 0.9 0.9 1.0 1.1
Note. Absolute range, 1–9. a Low desire women rated the valence of the sexual pictures significantly lower than high desire women, p = .02. b Low desire women rated the valence of the sexual pictures significantly lower than high desire women, p = .0001. c Average desire women rated the valence of the sexual pictures significantly lower than high desire women, p = .03.
main effect of Picture Type F(3,237) = 491.66, p < .0001, and a significant Desire Group × Picture Type interaction F(6,237) = 3.94, p < .001. Post-hoc analysis showed that the interaction effect was due to differences in the three group’s ratings of the sexual pictures only. Valence ratings for the Low and Average Desire women were lower on the repeat task while those of the High Desire women were higher (see Table VI). Analysis of the arousal and dominance ratings showed no significant effects of Desire Group.
desire groups with respect to numbers of slides lost from the analysis because of incorrect responses for either men or women in the study. All further analyses were based only on the correct non-outlier responses (n = 7,451). The picture recognition data for both men and women are shown in Fig. 1. For the men, a 3 (Desire Group) × 4 (Picture Type) ANOVA revealed only a significant main effect of Picture Type, F(3,153) = 13.57, p < .0001. Post-hoc analysis showed that while the responses to the high valence sexual pictures (M = 3,165 ms) and the unpleasant pictures (M = 2,996 ms) were not significantly different from each other, men responded to the sexual pictures more slowly than the high valence non-sexual (M = 2,459 ms) and neutral pictures (M = 1977 ms; both ps < .0001). They also responded to the unpleasant pictures more slowly than to the non-sexual (p < .01) and neutral pictures (p < .0001). For the women, a 3 (Desire Group) × 4 (Picture Type) ANCOVA revealed significant main effects of Desire Group, F(2, 78) = 3.95, p < .05, and Picture Type, F(3,237) = 19.81, p < .0001, and a significant Desire Group × Picture Type interaction, F(6,237) = 2.57, p < .05. Post-hoc analysis of the interaction revealed that the women in the relatively Low Desire group were faster at responding to the sexual (ps < .01), non-sexual (ps < .01), and unpleasant (ps < .001), picture types than the other women.
Picture Recognition The dependent variable in this task was the time taken to respond yes or no to the question, “Have you seen this picture before?” when each slide in the selection appeared on the computer monitor. Responses were correct in 93.3% of the cases (n = 7,740). The mean time recorded for all correct responses was 2,940 ms (SD = 3,950; range, 350–12,7,488). Some of the values were statistical outliers. Examination of the outliers, using the rule of thumb that any response >3 SD outside the individual’s overall mean should be discarded (cf. Bradley, Mogg, Millar, & White, 1995; Janssen, Everaerd, Spiering, & Janssen, 2000; Ratcliff, 1993), resulted in 292 responses being eliminated from the analysis. The majority of individuals lost no more than two data points from their responses. Responses lost were spread among picture types as follows: high valence sexual (n = 122); high valence nonsexual (n = 55); neutral (n = 20); low valence unpleasant (n = 95). Because the variation in the recognition task was to be analyzed by desire group, we examined the number of slides lost by participants to see whether there were differences in the incorrect answers according to desire group. There was no significant difference among the
Sexual Content Induced Delay We were interested in whether SCID would be demonstrated in this sample and what effect desire levels might have on SCID. We found that recognition time responses to the sexual stimuli were significantly slower than those to the neutral stimuli in both men, t(53) = 6.55, p < .00001, and women, t(81) = 6.59, p < .00001 (see Fig. 2). Then, individual’s mean response times for neutral pictures were subtracted from those for the sexual pictures and an ANOVA (for the men) and an ANCOVA (covarying age for the women) were conducted. We found that there was no significant effect of sexual desire group on SCID in either men, F(2, 51) < 1, or women, F(2, 78) = 1.26. Viewing Times Recorded After they had carried out the picture recognition task, the participants viewed the slides for an unrestricted time. For the men, a 3 (Desire Group) × 4 (Picture Type) ANOVA of the viewing time revealed no significant effects of Desire Group or Picture Type.
Pictorial Cues and Sexual Desire
Fig. 1. Mean picture recognition response times (±SE) by picture type, showing variation in responding by sex (men upper panel, women lower panel) and sexual desire group.
For the women, a 3 (Desire Group) × 4 (Picture Type) ANCOVA revealed only a significant main effect of Picture Type, F(3,237) = 4.86, p < .01. Post-hoc analysis showed that the women spent significantly longer viewing the low valence unpleasant slides (M = 24,689 ms) than the sexual (M = 22,584 ms, p < .001), non-sexual (M = 23,499, p < .05), and neutral (M = 23,131 ms, p < .01) pictures.
Rating Times Recorded The rating times for men varied from a minimum of 1921 ms for a neutral slide to a maximum of 31,929 ms for an unpleasant slide. The ratings were examined using a 3 (Desire Group) × 4 (Picture Type) ANOVA, which revealed only a significant main effect of Picture Type, F(3,153) = 4.91, p < .01. Post-hoc analysis showed that
Conaglen and Evans
Fig. 2. Sexual content induced delay (mean ± SD) examined with respect to gender and sexual desire levels.
it took the men significantly longer to rate the unpleasant pictures (M = 11,775 ms) than the sexual (M = 10,526 ms, p < .05), non-sexual (M = 10,203 ms, p < .01), or neutral (M = 9,772 ms, p < .001) pictures. Rating times for the women varied from a minimum of 5,215 ms for a sexual slide to a maximum of 27158 ms for a neutral slide. A 3 (Desire Group) × 4 (Picture Type) ANCOVA of the rating times revealed only a significant main effect of Picture Type, F(3,237) = 14.89, p < .0001. Post-hoc analysis showed that, similar to the men, it took the women significantly longer to rate the unpleasant pictures (M = 12,780 ms) than the sexual (M = 11,234 ms, p < .0001), non-sexual (M = 11,094 ms, p < .0001), or neutral (M = 10,515 ms p < .0001) picture types.
than the neutral pictures (p < .05) but less so than the non-sexual pictures (p < .0001). Interest ratings reported by women varied from 4.4 to 25.2 for the sexual pictures (M = 17.7, SD = 3.9), 4.6 to 24.7 for the non-sexual pictures (M = 16.0, SD = 4.0), 1.9 to 17.4 for the neutral pictures (M = 8.2, SD = 3.5), and 1.0 to 28.7 for the unpleasant pictures (M = 18.5, SD = 5.6). A 3 (Desire Group) × 4 (Picture Type) ANCOVA revealed only a significant main effect of Picture Type, F(3,237) = 102.2, p < .0001. Post-hoc analysis showed that the women’s interest in sexual pictures was significantly higher than in non-sexual (p < .01) or neutral pictures (p < .0001), and that interest in the neutral pictures was less than in all other picture types (all ps < .0001). Neutral (p < .0001) and non-sexual pictures (p < .001) were also rated more interesting than the unpleasant pictures.
Interest Ratings Interest ratings reported by men varied from 11.9 to 24.9 for the sexual pictures (M = 18.5, SD = 3.0), 12.8 to 25.9 for the non-sexual pictures (M = 18.0, SD = 2.8), 1.4 to 15.1 for the neutral pictures (M = 8.6, SD = 3.7), and 1.0 to 27.4 for the unpleasant pictures (M = 16.6, SD = 5.8). A 3 (Desire Group) × 4 (Picture Type) ANOVA revealed only a significant main effect of Picture Type, F(3,153) = 68.57, p < .0001. Post-hoc analysis showed the men’s interest in sexual pictures was significantly higher than in the neutral (p < .0001) or unpleasant pictures (p < .05). The latter were rated more interesting
DISCUSSION This study examined whether participants responded to pictorial cues in ways that differed according to their level of self-reported sexual desire, but largely failed to confirm expectations based on a previous study involving a semantic protocol (Conaglen, 2004). In that study, both men and women with lower levels of sexual desire responded more slowly to sexual stimuli than other participants. There was also a tendency for SCID to be greater in those with relatively low sexual desire.
Pictorial Cues and Sexual Desire In the present study, there was a relationship between recognition of the pictures and levels of desire for women. However, responses to pictorial sexual stimuli did not differ significantly by desire grouping in the men. We found that SCID was not significantly related to sexual desire levels in either men or women. One possible reason for the lack of significant differences in the men was their level of sexual desire. The self-report sexual desire scores for the men categorized as relatively low desire were in the range one might regard clinically as normal or better with that group scoring means of 60.5 (SD = 6.9) on the HISD and 56.9 (SD = 11.1) on the SDI-2. By comparison, in a recent study of men who had clinically significant low desire due to hypogonadism (low testosterone or high prolactin), mean scores of 39.4 (SD = 21.4) on the HISD and 34.5 (SD = 23.9) on the SDI-2 were obtained (Conaglen & Conaglen, 2005). In their clinically recovered state, these men scored 61.4 (SD = 12.6) and 55.3 (SD = 16.7) on the two measures, which can be seen to compare with the men in the relatively low desire group in the present study. We would not, therefore, expect the men in the present study to demonstrate responses or ratings typical of men with low levels of sexual desire. An additional consideration parallels the finding by Spiering et al. (2004) that when cognitive tasks are required, no significant gender differences were found, but when men and women were asked to perform an emotional or affective task associated with sexual images, significant gender differences emerged. Our findings may reflect the same mechanism that they postulated, namely that involvement of explicit or conscious memory generated the gender differences. The significant findings specific to each hypothesis for the women studied are examined in turn. We had expected that ratings for the sexual pictures would reflect the participants’ interest or lack of interest in sex, and thus be related to their sexual desire levels and categorization. However, we found that desire grouping only affected the women when rating and picture type interacted, specifically when the lower desire women’s ratings of valence (whether a picture is pleasant or unpleasant) for sexual pictures was contrasted and shown to be lower than the high desire women’s ratings. When the women rated pictures a second time, later in the protocol, only the valence ratings given to the sexual pictures did not change significantly. It is of interest that the sexual images retained their ratings while other types changed significantly, since this suggests something different about the sexual aspect of these images. It is important to note that we found that sexual images were not only initially rated less pleasant by women with
lower sexual desire, an expected result, but were also rated lower by that group when viewed a second time. In addition, the higher desire women rated sexual pictures as more pleasant on the second viewing, although the difference between this and the first rating did not reach significance. Perhaps sexual cues are particularly salient for both these groups for opposing reasons with the lower desire women showing that their underlying tendency is to inhibit subsequent sexual related behavior, while the higher desire women demonstrate a priming to engage in such behavior, as has been suggested by Bass and Walen (1986). The second ratings for arousal showed that the low desire women’s ratings of the sexual pictures were significantly lower on the second occasion (p < .001), while those of the higher desire women did not differ significantly. When one considers the low group, a parallel can be seen with findings by Hamm et al. (1996) among fearful women showing that they exhibited less fear when stimuli were presented a second time, and suggests that for women with low desire habituation to the images occurs, resulting in less arousal to the sexual images. However, those with higher sexual desire levels appear to retain their arousal as well as their rating of the pleasantness of these images, encouraging the view that they are disposed to appetitive sexual behavior given the opportunity. The women in the present study did not demonstrate the expected slower responding to sexual pictures that we thought would be associated with lower levels of sexual desire. In contrast to our expectation, women with relatively low desire were significantly quicker at the task than either the average or higher desire women. While this finding was contrary to expectations derived from the semantic cues study (Conaglen, 2004), it accords with the avoidance behaviors of people fearful of snakes and spiders, or mutilation fear as observed by Hamm et al. (1997). Lang et al. (1990) have suggested emotional responses to stimuli will be either appetitive or aversive, and these findings suggest the lower desire women found these images aversive, while the high desire group found them appetitive. However, the lower desire women did respond more quickly across all picture types, which suggests there may be something else generating this pattern of responding. One factor that could be involved, and is also frequently associated with lower levels of desire is anxiety. We did measure sexual anxiety in these participants (Conaglen, 1999) and those women with lower desire were found to have significantly higher levels of anxiety than the other participants. This could account for faster responding across all categories of pictures, and adds to the case for a study comparing a clinical with a control group of women.
Conaglen and Evans Both men and women in the present study demonstrated significant SCID, but there was no consistent relationship between desire groups and SCID. This might have been due to the variability in responding in a pictorial protocol, since the range of response times and the individual variation was much greater than one finds in a semantic protocol. The last and perhaps simplest of our hypotheses, but that which underpins the study, related to whether people with lower desire would view sexual pictures less interesting than others did. While the global analysis showed that interest ratings for each picture type followed patterns previously observed in similar studies, that is greater interest in sexual and unpleasant pictures than in neutral images (see e.g., Lang, Bradley & Cuthbert, 1990), the desire group analysis did not reach significance. However a one-way ANOVA did show that low desire women rated sexual pictures less interesting than either the average (p < .05) or the higher desire groups (p < .01), lending credibility to the construct that interest in sexual images and sexual desire can be related. By showing significant differences in responses to sexual pictures in women with relatively high desire compared to those with lower desire levels, we have contributed further evidence that sexual responding can be just as likely to vary with respect to the emotional content of the stimuli as other forms of human responding. Viewed in tandem with the results of a previous study using semantic stimuli (Conaglen, 2004), these findings suggest a possible surrogate means of testing for changes in sexual desire in women. Moreover, we have also reconfirmed the validity of the concept of SCID or delay due to sexual content within pictorial contexts. Earlier word studies, within which this construct was first demonstrated, showed that responses to sexual cues differed from neutral or romantic cues. This study examining the construct in a pictorial protocol showed the relative strength of the phenomenon in such contexts. It accords with the Spiering et al. (2002) finding that SCID can be demonstrated with pictorial stimuli as well as words. While re-asserting that sexual content and the emotional impact of that content influences how people respond to pictures, this study also found that pictures of equivalent emotional content, or valence, were responded to differently depending on whether the content was sexual or not. This belies the suggestion made previously, namely that variation in responding might be based simply on the emotional content of the stimuli involved. In trying to reconcile the apparently contradictory findings that low desire persons responded most slowly
to sexual word stimuli, but most rapidly when the stimuli were pictures, we can return to the involvement of implicit and explicit memory in cognitive processing. It would appear that the semantic stimuli task used in our first study required unitizing (Mori & Graf, 1996), whereas the pictorial cues recognition task used here required elaboration of the information. In addition, the elaboration of sexual images when compared to elaboration of nonsexual images, takes longer when they are more salient for the individual. So the two tasks that yield apparently disparate response patterns in persons with low sexual desire—slow responses to word tasks and greater SCID, compared to faster responses to picture recognition and less SCID, involve different aspects of memory. Further, examining our findings in conjunction with the ideas suggested by Spiering, Everaerd, and Janssen (2003)—that subjective responding is based on explicit or declarative memory and involves a conscious elaboration of emotional information—this elaborative process may account for the longer response times recorded by persons with higher levels of sexual desire. In addition, the contrast with the semantic protocol finding suggests that the word task involved implicit memory rather than the explicit memory involved in our picture recognition tasks. A further possibility is that for a person with low desire, fear or worry, generated by a sexual image, prevented them from attending to the sexual stimuli in a sexual fashion, and inhibited further processing, a reason for the speedier responses recorded to sexual pictures by those with low sexual desire. However one explains the findings, the clinical concern, for those troubled by low sexual desire, is to increase for the client the salience of sexual images, or in the wider context, sexual cues of any kind. In demonstrating the tendency for subjects to take longer to process sexual images, this study raises several issues. Some of these are methodological, others more theoretical, or clinical in nature. On the methodological side, the experimental procedure did not control for any effect of order on the components of the protocol. While order had not been found to influence the picture rating process in pilot work, the present study may have inadvertently primed responding in some way by showing pictures to all participants in a specific (though pseudorandomized) order. Using information processing techniques to investigate the emotionally charged area of sexuality and its expression within a New Zealand context raises the issue of the cultural validity of the rating technique. By piloting the rating method with New Zealand students and a selection of images from the IAPS, we established that Lang’s claims for the cross-cultural validity of this
Pictorial Cues and Sexual Desire methodology were also true in the New Zealand context (Conaglen, 1999). The ratings given the pictures by New Zealanders correlated with those of the Americans at a .83 to .97 level. Thus, it was established that using the American ratings of images as a way of choosing comparable selections for the laboratory protocol would be sufficiently applicable for use in the New Zealand environment. Conversely, and more importantly for the field at large, any results within the New Zealand context can reasonably be expected to have validity for other Western contexts. One further influence on these results involves a combination of methodology, theory, and clinical concern. That is the manner in which the participants were separated into high, average, and lower sexual desire groups as a result of their self-report measure scores. The summation of raw scores and subsequent separation into quartile groups was a replicable way in which to determine membership of the three desire groups, but does not necessarily conform precisely to clinical criteria of low sexual desire. Comparison of the questionnaire data for the females in this study, with women seen in clinical practice, shows that the low sexual desire group participants in this study had slightly higher scores than clinic clients with low desire problems. A replication in a clinical sample would be important to better understand the potential of using these techniques in practical contexts. In conclusion, this study has demonstrated the potential of an information processing approach in the emotional area of sexual behavior, particularly for understanding sexual desire. The methodology should prove adaptable to further investigate the clinically relevant questions that arise from this demonstration of the relevance of responses to cues of this nature. What we have demonstrated supports the hypotheses that persons with low sexual desire react differently to sexual cues in their environment, although we found that pictorial sexual cues were processed more speedily by low libido persons, perhaps indicating an inhibition of translation of recognized cues (or a lack of elaboration) into interest in things sexual. In addition, the pleasure ratings for sexual images demonstrated a more negative view of sexual images being held by that group, particularly in contrast to the people recording high levels of sexual desire. Overall, we obtained a picture of less engagement, in those with low libido, with images of a sexual nature, compared to other images. The therapeutic challenge is how to assist such people to engage more fully with such material as a step towards a greater appreciation of, and interest in, matters sexual within their personal contexts.
APPENDIX International Affective Picture System slides used in protocol were as follows: High valence sexual slides. First selection: 4510, Male nude; 4532, Man; 4535, Biceps; 4603, Couple; 4607, Couple; 4640, Couple; 4652, Couple; 4659, Couple; 4660, Couple; 4680, Nudes. Second selection: 4533, Man; 4599, Couple; 4608, Couple; 4610, Couple; 4641, Couple; 4651, Couple; 4664, Couple; 4690, Nudes; 4700, Couple. High valence non-sexual slides. First selection: 1999, Disney; 4150, Skater; 5260, Water falls; 5480, Fireworks; 8034, 8180, Cliff divers; 8190, Skiers; 8370, Rafting; 8500, Gold. Second selection: 1463, Kittens; 1710, Puppies; 5450, Space shuttle; 5470, Astronaut; 5700, Mountains; 7410, M & Ms; 8080, Sailing; 8120, Tennis player; 8170, Sail boat; 8501. Neutral slides. First selection: 2840, Chess; 7002, Towel; 7050, Hair dryer; 7233, Plate; 7235, Chair. Second selection: 5510, Mushroom; 6150, Outlet; 7009, Mug; 7034, Hammer; 7035, Mug; 9070, Boy. Low valence slides. First selection: 3000, Mutilation; 3053, Burn victim; 3060, Mangled face; 3080, Mangled face; 3120, Body; 9040, Child. Second selection: 3102, Burn victim; 3130, Body; 3170, Baby with tumor; 3350, Infant; 9410, Soldier; 9570, Dog. ACKNOWLEDGMENTS This article describes one aspect of the first author’s doctoral dissertation, which was supervised by the second author. The authors thank Rob Bakker for writing the program used for the study and express their gratitude to the reviewers whose constructive comments assisted in the revision of the article. REFERENCES Apt, C. V., & Hurlbert, D. F. (1992). Motherhood and female sexuality beyond one year postpartum: A study of military wives. Journal of Sex Education and Therapy, 18, 104–114.
Conaglen and Evans Apt, C., & Hurlbert, D. F. (1994). The sexual attitudes, behavior, and relationships of women with histrionic personality disorder. Journal of Sex and Marital Therapy, 20, 125– 133. Apt, C., Hurlbert, D. F., Sarmiento, G. R., & Hurlbert, M. K. (1996). The role of fellatio in marital sexuality: an examination of sexual compatibility and sexual desire. Sexual and Marital Therapy, 11, 383–392. Bass, B. A., & Walen, S. R. (1986). Rational-emotive treatment for the sexual problems of couples. Journal of Rational-Emotive Therapy, 4, 82–94. Basson, R. (2001). Using a different model for female sexual response to address women’s problematic low sexual desire. Journal of Sex and Marital Therapy, 27, 395–403. Basson, R. (2002). Women’s sexual desire. Disordered or misunderstood? Journal of Sex & Marital Therapy,28, 17–28. Beck, J. G. (1995). Hypoactive sexual desire disorder: An overview. Journal of Consulting and Clinical Psychology, 63, 919– 927. Bradley, B. P., Mogg, K., Millar, N., & White, J. (1995). Selective processing of negative information: Effects of clinical anxiety, concurrent depression, and awareness. Journal of Abnormal Psychology, 104, 532–536. Catania, J. A., Gibson, D. R., Chitwood, D. D., & Coates, T. J. (1990). Methodological problems in AIDS behavioral research: Influences on measurement error and participation bias in studies of sexual behavior. Psychological Bulletin, 108, 339– 362. Conaglen, H. M. (1999). Factors involved in sexual desire: Environmental cues and their impact. Unpublished doctoral dissertation, University of Waikato, Hamilton, New Zealand. Conaglen, H. M. (2004). Sexual Content Induced Delay: A reexamination investigating relation to sexual desire. Archives of Sexual Behavior, 33, 359–367. Conaglen, H. M., & Conaglen, J. V. (2005). Measurement of sexual desire during treatment for hypogonadism. Manuscript in preparation. Conaglen, H. M., Hunt, P., & Conaglen, J. V. (2003). Measuring sexual desire in couples undergoing treatment for hypogonadism. Poster presented at the meeting of the International Academy of Sex Research, Bloomington, Indiana. Donahey, K. M., & Carroll, R. A. (1993). Gender differences in factors associated with hypoactive sexual desire. Journal of Sex & Marital Therapy, 19, 25–40. Geer, J. H. (1996). Gender differences in the organization of sexual information. Archives of Sexual Behavior, 25, 91–107. Geer, J. H., & Bellard, H. S. (1996). Sexual content induced delays in unprimed lexical decisions: Gender and context effects. Archives of Sexual Behavior, 25, 379–395. Geer, J. H., Judice, S. L., & Jackson, S. R. (1994). Reading times of erotic material: The pause to reflect. Journal of General Psychology, 121, 345–352. Geer, J. H., & Manguno-Mire, G. M. (1996). Gender differences in cognitive processes in sexuality. Annual Review of Sex Research, 7, 90–124. Geer, J. H., & McGlone, M. S. (1990). Sex differences in memory for erotica. Cognition and Emotion, 4, 71–78. Geer, J. H., & Melton, J. S. (1997). Sexual content-induced delay with double-entendre words. Archives of Sexual Behavior, 26, 295– 316. Greenwald, M. K., Cook, E. W., & Lang, P. J. (1989). Affective judgment and psychophysiological response: Dimension covariation in the evaluation of pictorial stimuli. Journal of Psychophysiology, 3, 51–64. Haerich, P., & Khoury, L. (1994). Erotophobia, erotophilia, and startle modification differences in affective responding to the same stimulus. Psychophysiology, 31,S54.
Hamm, A. O., Cuthbert, B. N., Globisch, J., & Vaitl, D. (1997). Fear and the startle reflex: Blink modulation and autonomic response patterns in animal and mutilation fearful subjects. Psychophysiology, 34, 97–107. Heiman, J. R. (2001). Sexual desire in human relationships. In W. Everaerd, E. Laan, & S. Both (Eds.), Sexual appetite, desire and motivation: Energetics of the sexual system (pp. 117–134). Amsterdam: Royal Netherlands Academy of Arts and Sciences. Hurlbert, D. F. (1993). A comparative study using orgasm consistency training in the treatment of women reporting Hypoactive Sexual Desire. Journal of Sex and Marital Therapy, 19, 41–55. Janssen, E., Everaerd, W., Spiering, M., & Janssen, J. (2000). Automatic processes and the appraisal of sexual stimuli: Toward an information processing model of sexual arousal. Journal of Sex Research, 37, 8–23. Kaplan, H. S. (1977). Hypoactive sexual desire. Journal of Sex & Marital Therapy, 3, 3–9. Lang, P. J. (1995). The emotion probe: Studies of motivation and attention. American Psychologist, 50, 372–385. Lang, P. (1979). A bio-informational theory of emotional imagery. Psychophysiology, 16, 495–512. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1995). International Affective Picture System (IAPS): Technical manual and affective ratings. Gainesville, FL: NIMH Center for the Study of Emotion and Attention. Lang, P., Bradley, M., & Cuthbert, B. (1996). The International Affective Picture System [CD-ROM]. Gainesville, FL: NIMH Center for the Study of Emotion and Attention. Leiblum, S. R., & Rosen, R. C. (1988). Introduction: Changing perspectives on sexual desire. In S. R. Leiblum & R. C. Rosen (Eds.), Sexual desire disorders (pp. 1–17). New York: Guilford. Levine, S. B. (2003). The nature of sexual desire: A clinician’s perspective. Archives of Sexual Behavior, 32, 279–285. McCabe, M. P. (1991). The development and maintenance of sexual dysfunction: An explanation based on cognitive theory. Sexual and Marital Therapy, 6, 245–260. Mori, M., & Graf, P. (1996). Nonverbal local context cues explicit but not implicit memory. Consciousness and Cognition, 5, 91–116. Ratcliff, R. (1993). Methods for dealing with reaction time outliers. Psychological Bulletin, 114, 510–532. Schnarch, D. M. (1991). Constructing the sexual crucible: An integration of sexual and mental therapy. New York: W. W. Norton. Sipski, M. L., Alexander, C. J., & Rosen, R. C. (1977). Physiologic parameters associated with sexual arousal in women with incomplete spinal cord injuries. Archives of Physical Medicine and Rehabilitation, 78, 305–313. Spector, I. P., Carey, M. P., & Steinberg, L. (1996). The Sexual Desire Inventory: Development, factor structure, and evidence of reliability. Journal of Sex and Marital Therapy, 22, 175–190. Spiering, M., Everaerd, W., & Elzinga, B. (2002). Conscious processing of sexual information: Interference caused by sexual primes. Archives of Sexual Behavior, 31, 159–164. Spiering, M., Everaerd, W., & Janssen, E. (2003). Priming the sexual system: Implicit versus explicit activation. Journal of Sex Research, 40, 134–145. Spiering, M., Everaerd, W., & Laan, E. (2004). Conscious processing of sexual information: Mechanisms of appraisal. Archives of Sexual Behavior, 33, 369–380. Tolin, D. F., Lohr, J. M., Lee, T. C., & Sawchuk, C. N. (1999). Visual avoidance in specific phobia. Behaviour Research and Therapy, 37, 63–70. Williams, J. M. G., Mathews, A., & MacLeod, C. (1996). The emotional Stroop task and psychopathology. Psychological Bulletin, 120, 3–24. Zajonc, R. B. (1984). On the primacy of affect. American Psychologist, 39, 117–123.
