The influence of Emotion Down-Regulation on the Expectation of Sexual Reward

Available online at www.sciencedirect.com

ScienceDirectBehavior Therapy xx (2015) xxx–xxx

www.elsevier.com/locate/bt

BETH-00544; No of Pages 16; 4C:

The Influence of Emotion Down-Regulation on theExpectation of Sexual Reward

Mirte BromLeiden University and Leiden University Medical Centre

Ellen LaanWalter Everaerd

University of Amsterdam

Philip SpinhovenLeiden University and Leiden University Medical Centre

Janna CousijnUtrecht University

Stephanie BothLeiden University Medical Centre

Emotion regulation research has shown successful alteringof unwanted aversive emotional reactions. Cognitive strat-egies can also regulate expectations of reward arising fromconditioned stimuli. However, less is known about the effi-cacy of such strategies with expectations elicited by condi-tioned appetitive sexual stimuli, and possible sex differencestherein. In the present study it was examined whether acognitive strategy (attentional deployment) could successfullydown-regulate sexual arousal elicited by sexual reward-conditioned cues in men and women. A differential condi-tioning paradigm was applied, with genital vibrostimulationas unconditioned stimulus (US) and sexually relevant pictures

This research was supported by a grant of the NetherlandsOrganization for Scientific Research (NWO) awarded to Dr. S.Both. The funders had no role in study design, data collection andanalysis, decision to publish, or preparation of the manuscript. Theauthors have declared that no competing interests exist. The authorswish to thank Ashray Dihal, Lizzy Ooms and Marieke Bremer fortheir assistance in collecting the data.

Address correspondence to Mirte Brom, MSc, Department ofPsychosomatic Gynaecology and Sexology, Leiden UniversityMedical Centre, Leiden, The Netherlands; e-mail: [email protected]/© 2015 Association for Behavioral and Cognitive Therapies.Published by Elsevier Ltd. All rights reserved.

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

as conditional stimuli (CSs). Evidence was found for emotiondown-regulation to effect extinction of conditioned sexualresponding in men. In women, the emotion down-regulatorystrategy resulted in attenuated conditioned approach tenden-cies towards the CSs. The findings support that top-downmodulation may indeed influence conditioned sexual re-sponses. This knowledge may have implications for treatingdisturbances in sexual appetitive responses.

Keywords: emotion regulation; sexual conditioning; sexual response;sexual reward learning; sexual arousal

RESEARCH IN ANIMALS AND HUMANS supports thenotion that reward learning in the form of classicalconditioning can contribute to the etiology of bothnormal and maladaptive sexual behaviors, likeparaphilias, or deviant sexual preferences (Brom,Both, Laan, Everaerd, & Spinhoven, 2014; Pfaus,Kippin, & Centeno, 2001). In classical condition-ing, through the repeated association with the un-conditional stimulus (US), a neutral stimulus (NS)can eventually elicit the same reaction as the US(Bindra, 1974; Pavlov, 1927). The NS is now called

own-Regulation on the Expectation of Sexual Reward, Behavior

2 brom et al .

the conditioned stimulus (CS) and the reaction tothe CS is called the conditioned response (CR).Several notable studies have demonstrated condi-tioned sexual arousal responses in humans (for areview see Brom, Both, et al., 2014). Both from alearning theory and neuroscience perspective,disorders in sexual motivation, like hypersexuality,can potentially be characterized as disorders in-volving disturbed emotional learning and memoryprocesses resulting in enhanced sexual responseacquisition and maintenance.The expectation of a potential sexual reward can

elicit positive feelings and sexual arousal and there-fore can aid in the learning about environmentalcues that predict future sexual rewards. However,this reward expectation signal can also be maladap-tive, potentially eliciting sexual urges that may bedifficult to control, like in the case of hypersexuality.Therefore, it is important to understand how toregulate or control the positive feelings associatedwith reward expectation. One promising methodfor examining this is the utilization of cognitivestrategies. The term emotion regulation signifies anyprocess that serves to initiate, inhibit, or modulate(e.g., cognitively reevaluate) emotional feelings orbehavior (Aldao, 2013; Gross, 2002; Gross &Thompson, 2007). Successful emotion regulationmay be dependent on top-down control from theprefrontal cortex over subcortical regions involvedin reward and emotion. Failures in this deploymentof top-down cognitive control mechanisms or over-active bottom-up processesmay contribute to severalforms of psychopathology (Heatherton & Wagner,2011; Ray&Zald, 2012), including sexual disorderswith a learned component (Bancroft & Janssen,2000; Both, Laan & Everaerd, 2011; Klucken et al.,2014; van Lankveld, van den Hout, & Schouten,2004; Salemink & van Lankveld, 2006). Cognitivestrategies can successfully alter unwanted aversiveemotional reactions. Emotional down-regulationstrategies can influence emotions at the input phases(i.e., antecedent focused like cognitive reappraisalor attentional deployment) and at the output phase(i.e., response focused like suppression; Gross,1998; Webb, Miles, & Sheeran, 2012). Gross andThompson (2007) suggest that antecedent-focusedstrategies (e.g., attentional deployment) are moreeffective than response-focused strategies. As rela-tively few studies on negative emotions, and evenfewer studies on positive emotions, have investigatedthe effects of the promising active distraction strat-egies (where the emphasis is on participants to bringto mind something unrelated to the emotion or emo-tional stimulus to serve as a distraction), especially onbehavioral and physiological measures of emotion,this is an important avenue for future research (Webb

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

et al., 2012). At present, there is growing evidencethat cognitive strategies like attentional deploymentcan also regulate expectations of reward arisingfrom conditioned stimuli (Delgado, Gillis, & Phelps,2008). However, less is known about the efficacy ofsuch strategies with expectations elicited by condi-tioned appetitive sexual stimuli. To our knowledge,the present study is the first to investigate whethera cognitive down-regulatory strategy can efficientlyregulate sexual arousal elicited by sexual reward-conditioned cues.At present, it is unclear if men and women are

equally prone to conditioning of sexual responseand if sex differences do exist in the emotionregulation of positive emotions, like sexual arousal.Given the fact that paraphilia and hypersexualityare predominantly observed in men (Kafka 1994;Kuzma & Black, 2008; Rosen, 2000), it is spec-ulated that men are more receptive to increased CRacquisition (Domjan, 2005; Gutiérrez & Domjan,1997; Klucken et al., 2009; Pfaus et al., 2001).Nevertheless, few studies have addressed sexualconditioning in both men and women (Brom, Both,et al., 2014), and some results are contradictoryto this general assertion (Brom, Laan, Everaerd,Spinhoven, & Both, 2014; Hoffmann, Janssen &Turner, 2004). Second, with respect to emotionregulation, the general assertion is that women usemore emotion-focused strategies, while men arethought to use more efficient cognitive (rational)strategies (Whittle et al., 2011). However, most—ifnot all—of these results relate to the regulation ofparticularly negative emotions (Gross, 2007; Maket al., 2009; McRae et al., 2008). Hence, thecontradictory results of previous sexual condition-ing studies and the lack of studies on sex differencesin positive emotion regulation point to the impor-tance for further investigation of possible genderdifferences in sexual learning and cognitive regula-tion thereof.In the present study, a differential conditioning

paradigm was applied, with instructions adaptedfrom Delgado et al. (2008). It was predicted thatparticipants in two conditions (the control condi-tion Attend and the experimental Down-Regulatecondition) would show conditioned genital andsubjective sexual responding to the CS that waspaired with the US (the CS+), which was expectedto gradually decrease during extinction trials. Whenthe Attend instruction preceded the CSs, the par-ticipant was instructed just to pay attention to thestimulus. In contrast, when the instruction Regulateappeared onscreen, participants were instructed toconjure a soothing image from nature prompted bythe color of the stimulus. Instructions were pre-sented in acquisition and extinction phases. It was

own-Regulation on the Expectation of Sexual Reward, Behavior

3i n f luence of emot ion down-regulat ion on sexual reward

predicted that an emotion down-regulation strategywould successfully decrease arousal elicited by thesexual reward-conditioned cue, in men and women,in both the acquisition and extinction phases. Sincesubjective ratings are susceptible to demand char-acteristics, in addition a task was included to assessimplicit approach and avoidance tendencies to-wards the CS (Cousijn, Goudriaan, & Wiers,2011). We assumed participants should be fasterwhen instructed to approach the CS+ and avoidfrom the CS- than when instructed to avoid the CS+and approach the CS-, and an emotion down-reg-ulation strategy should decrease these responseselicited by reward-conditioned cues.

Methodparticipants

Research participants were 40 men and 53 women.Participants were paid (€30,-) for their participationand were recruited using posted advertisements.The advertisement stated that the focus of the studywould be on the relationship between erotic (genital)stimulation and sexual arousal. Inclusion criteriawere: age between 18 and 45 years and a hetero-sexual orientation. Exclusion criteria were: sexualproblems, a Diagnostic and Statistical Manual ofMental Disorders (DSM-IV) diagnosis of an affectiveor psychotic disorder or abusive drug use, pregnancyor breastfeeding, and a medical illness or use ofmedication that could interfere with sexual response.Written informed consent was obtained from allparticipants. The study was approved by the EthicalCommittee of the Medical Center.

design and conditioning procedure

One stimulus (the CS+) was followed by the genitalvibrostimulation (US) during the acquisition phase,

FIGURE 1 Schematic representation of the experimentalextinction phase, before every CS presentation a written ccondition received the instruction “Regulate” whereas partici“Attend” prior to each CS+. Assignment of the color ofcounterbalanced across participants and conditions.

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

whereas the other stimulus (CS-) was never followedby genital vibrostimulation. Participants were ran-domly assigned to one of the two conditions:Down-Regulate or Attend, with restriction thatconditions matched on sex as close as possible. Fora schematic overview of the procedure see Figure 1.In the preconditioning phase, participants sawfour nonreinforced presentations of the CS+ andfour presentations of the CS-, for 9 seconds each.Subsequently, in the acquisition phase the CS+ andCS- were presented 10 times each and the CS+ wasalways followed by the US. In the extinction phasethe CS+ was no longer followed by the US. Priorto CS presentation, in the acquisition and extinctionphases participants were presented with a writtencue (attend or regulate) on screen for 2 s thatreminded participants to either Attend or Down-Regulate. All phases were presented without inter-ruption. Genital response was measured continu-ously during resting baseline, preconditioning,acquisition, and extinction phases. There were tworandomCS orders for each phase (that was counter-balanced across participants); with the restrictionof only two successive presentations of each CS.During the whole procedure intertrial intervals(ITIs) were 20, 25, or 30 seconds. The order of thelength of the ITI was random, with the restriction ofonly two successive lengths. Stimuli and cues werepresented byusing E-prime 2.0 Software (PsychologySoftware Tools, Inc).

materials, apparatus, and recordingStimulus MaterialsTwo identical pictures that served as CSs portrayedan abdomen of an individual of the opposite sex(wearing underwear), with the color of the under-wear in the picture (blue or yellow) being the only

procedure in both conditions. In the acquisition andue was presented: participants in the Down-Regulatepants in the control condition received the written cuethe pictures (blue or yellow) as CS+ and CS- was

own-Regulation on the Expectation of Sexual Reward, Behavior

4 brom et al .

difference. The CSs were shown in the middle of acomputer monitor, approximately 1.5 m in front ofthe participant. The size of the presented pictureswas 14 × 21 cm. Assignment of the pictures as CS+and CS- was counterbalanced across participantsand conditions.

Written InstructionsIn the Attend condition participants received thewritten cue Attend prior to each trial in the acqui-sition and extinction phases. They were instructedto "just pay attention" to the CSs when they werepresented this cue. In contrast, in the Down-Regulate condition participants were only present-ed with the Regulate cue in the acquisition andextinction phases, and were instructed that whenthe cue Regulate appeared on the monitor, theyshould conjure a soothing image from natureprompted by the color of the CS. For example,upon seeing the blue CS, participants could imaginethe ocean or blue sky, while imagining a sunnybeach or a field of flowers for the yellow CS.Participants were asked to generate the same imageevery time each color CS was presented.

Genital Vibrostimulation (US)Genital vibrostimulation was provided 8 s follow-ing the start of the CS+ for 2 s. For men, the US wasadministered by means of a ring-shaped vibrator.They were instructed to place the vibrator justbelow the coronal ridge (Janssen et al., 1994). Forwomen, a small hands-off vibrator (2 cm diameter)was used (Laan & van Lunsen, 2002). The vibratorwas placed on the clitoris using a lycra panties thathad an opening for the vaginal plethysmograph. Allparticipants were instructed to position the vibratoras most sexually stimulating.

Male Genital Sexual ArousalAn indium/gallium-in-rubber penile gauge assessedchanges in penile circumference (Bancroft, Jones, &Pullan, 1966). The gauges were calibrated beforeeach laboratory session using a set of calibrated rings(Janssen, Prause, & Geer, 2007). The penile gaugewas positioned two-thirds of the way down the shaftof the penis toward the base. Changes in electricaloutput caused by expansion of the gauge wererecorded by a continuous DC signal. The Indium-Gallium penile gauges were disinfected after eachuse, according to Sekusept plus disinfection proce-dure (MedCaT B.V.). Sekusept plus contains Gluco-protamine, which action spectrum covers bacteriaincluding mycobacteria, fungi, and viruses (e.g.,Human Papillomavirus [HPV]) (MedCaT B.V.).

Women’s Genital ArousalVaginal photoplethysmography assessed vaginalpulse amplitude (VPA; Laan, Everaerd, & Evers,

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

1995). Depth of the probe and orientation of thelight emitting diode were controlled by a device(a 6 × 2 cm plate) attached to the cable. The vaginalphotoplethysmograph was disinfected by means of aplasma sterilization procedure between uses. Plasmasterilization is a highly effective method for the com-plete removal of all organic (and certain inorganic)materials.

Subjective RatingsRatings of affective value, sexual arousal, and USexpectancy were collected during the precondition-ing and extinction phases. Participants were firstasked to rate, after each CS presentation, the affec-tive value of the CSs by answering the question,“What kind of feeling does this picture evoke inyou?” The question could be answered on a 7-pointLikert scale on a keyboard that varied from verynegative to very positive. Then, subjective sexualarousal was rated by answering the question, “Howsexually arousing is this picture to you?” Thequestion could be answered on a 7-point scale thatvaried from not sexually arousing at all to verysexually arousing. Then, participants were requiredto rate the expectancy of a vibration followingthe presentation of each CS on a 7-point scale byanswering the question, “To what extent did youexpect a vibration after this picture?” The scaleconsisted of 7 points labeled from certainly novibration to certainly a vibration. The questionswere presented at the monitor 1 second followingthe end of picture presentation.

other measuresApproach Avoidance Task (AAT; see Cousijn et al.,2011; E-prime 2.0 Software, Psychology SoftwareTools, Inc.)Participants were presented with the CS+, CS-,and neutral pictures from the International AffectivePicture System (IAPS; Lang, Bradley, & Cuthbert,2005). All images were rotated 3° left or right. Imagecontent was irrelevant to the task: participants wereinstructed to pull or push the joystick in response torotation direction. Pulling and pushing the joystickgradually increased and decreased image size. TheCS+, CS-, and the neutral pictures were presented80 times each, 40 times in push- and 40 times inpull-format, resulting in 240 test trials. The latencywas recorded between picture onset and completionof a full push or pull response. Literature supportsthe AAT’s validity in measuring approach/avoidancemotivational processes (Wiers et al., 2011).

The International Index of Erectile Function (IIEF)This is a validated 15-question questionnaire thatexamines 4 main domains of male sexual function:erectile function (6 questions, range 0-5), orgasmic

own-Regulation on the Expectation of Sexual Reward, Behavior

5i n f luence of emot ion down-regulat ion on sexual reward

function (2 questions, range 0–5), sexual desire(2 questions, range 0–5), and intercourse satisfac-tion (3 questions, range 0–5). Higher scores indicatebetter sexual function. Psychometric properties ofthe IIEF are good (Rosen et al., 1997).

The Female Sexual Function Index (FSFI)Women’s sexual functioning was assessed by theFSFI (Rosen et al., 2000; Ter Kuile, Brauer & Laan,2006), consisting of six subscales: desire (two items;range 1–5), arousal (four items; range 0–5), lubri-cation (four items; range 0–5), orgasm (three items;range 0–5), satisfaction (three items; range 0–5),and pain (three items; range 0–5). A higher scoreindicates better sexual functioning. The FSFI hasgood internal reliability and is able to differentiatebetween clinical samples and nondysfunctionalcontrols.

Exit InterviewParticipants were asked, among others things,about their reactions to the experimental proce-dure, the use of the genital device, and their eval-uation of the genital vibrostimulation. For instance,participants were asked to what extent they likedthe vibrostimulation. This could be rated on a5-point scale ranging from (1) not pleasant at all, to(5) very pleasant. Likewise, participants were askedhow sexually aroused they became by the vibration.In addition, they were asked about any prior ex-perience with vibrostimulation. Participants werealso asked about the used cognitive strategies, andthey were asked to rate how successful they werein concentrating and in the deployment ofthe cognitive strategy on a scale from 1 to 5 (i.e.,1 = trouble keeping concentrated to 5 = wellcapable keeping concentrated; and 1 = notsuccessful at all to 5 = very successful).

procedure

After completing the first session of the AAT,participants were instructed that the purpose of theexperiment was to measure physiological responsesto different pictures and to genital vibrostimula-tion. Before entering the experimental conditioningsession, participants were instructed about the vibro-stimulation, the colors of the CSs, and the writtencues that would appear on screen. Participants weremade aware of the contingencies (e.g., only the colorblue or yellow predicted a potential genital vibros-timulation). Then Attend or Regulate instructionswere explained. Participants were asked to verbalizewhat they were planning to think about when beingpresented with the written cues Attend and Regulateto assure that they were following the instructionsthey were given. In addition, participants werenotified that regardless of the instruction, the CS+

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

always indicated the possibility of receiving genitalvibrostimulation. Subsequently, the experimentalconditioning experiment followed (see Both et al.,2008; Both, Laan, & Everaerd, 2011; Both, Brauer,& Laan, 2011; Brom, Laan, et al., 2014, forconditioning procedure), starting with the precondi-tioning phase, followed by the acquisition and ex-tinction phases. In the acquisition and extinctionphase participants were presented with the writtencue Attend or Regulate prior to each CS. Directlyafter this experimental procedure, the second sessionof the AAT was completed. Then participantsprivately filled in questionnaires (e.g., FSFI, Rosenet al., 2000; IIEF, Rosen et al., 1997) and an exitinterview questionnaire was administered.

data reduction, scoring and analysis

After artifact removal, mean penile circumferenceor mean VPA level during the 2-minute restingbaseline period was calculated. Genital responses tothe CSs were scored in three latency windows:during 4–8, 9–12, and 13–16 seconds following CSonset, respectively FIR (first interval response), SIR(second interval response) and TIR (third intervalresponse) (see also Both et al., 2008; Both, Laan, &Everaerd, 2011; Both, Brauer, & Laan, 2011). ForFIR, SIR and TIR, change scores were calculated foreach CS presentation by subtracting mean genitalresting baseline from genital measures following CSpresentation. Since direct gender comparison ofgenital responses cannot be made because of thedifferent measures used, genital data for men andwomen were analyzed separately. For genitalresponses, effects were tested with mixed factorunivariate analysis of variance procedures (GeneralLinear Model in SPSS), with Stimulus and Trialas within-subject factors and Condition as betweensubjects factor. Analyses of subjective measure-ments and AAT scores were conducted for men andwomen combined, with Condition and Gender asbetween-subjects factor (General Linear Model inSPSS). The Greenhouse–Geisser correction wasapplied to adjust for violation of the sphericityassumption in testing repeated measures effects.All phases were analyzed separately. Precondition-ing and acquisition phases were both analyzedas a whole, whereas individual extinction trialswere analyzed separately, since sexual conditioningeffects have generally been found to be small (Brom,Laan, et al., 2014; Hoffmann et al., 2004), andthe deployment of the emotion-regulation strat-egy is expected to affect not only the magnitudeof conditioned responding (trial 1 and 2 of theextinction phase) but also the extinction of condi-tioned responding (trial 3 and 4 of the extinctionphase). Effect sizes are reported as proportion of

own-Regulation on the Expectation of Sexual Reward, Behavior

Table 1Subject Characteristics: Descriptive Subject Variables for Men and Women, and for Each Condition

Variable: Men Women Men & Women

Attend(n = 20)

Down-Regulate(n = 20)

Effect size(Cohen’s d)

Attend(n = 29)

Down-Regulate(n = 33)

Effect size(Cohen’s d)

Men(N = 40)

Women(N = 62)

Effect size(Cohen’s d)

MEAN SD MEAN SD p MEAN SD MEAN SD p MEAN SD MEAN SD p

Age (years) 25 6.1 22 2.6 .08 .66 28.8 8.2 27.7 8.1 .64 .14 23.6 4.8 28.2 8.0 b .01* .70Sexual Functioning (IIEF/ FSFI- score) 36.2 5.6 34.9 5.8 .69 .41 27.3 2.8 28.1 2.8 .32 .29Prior Experience Vibrostimulation 1.7 1.1 1.7 0.9 .96 .00 3.8 1.3 3.7 1.1 .87 .08 1.7 1.0 3.8 1.1 b .01* 2.0Pleasantness US 3.2 1.4 3.3 0.9 .81 .09 3.5 0.7 3.2 0.6 .21 .47 3.2 1.2 3.4 0.8 .48 .03US Perceived as Sexually Arousing 3.1 1.2 3.0 1.1 .89 .09 3.2 0.9 3.1 0.8 .90 .11 3.0 1.2 3.1 0.9 .59 .10Declared Sexual Arousal 2.5 1.4 2.5 1.0 .95 .00 2.5 0.8 2.5 0.8 .86 .00 2.5 1.2 2.6 0.9 .66 .10Instructions: Able to concentrate 4.3 0.5 3.93 0.6 .09 .69 4.0 0.4 3.5 0.6 b .01* .98 4.1 0.6 3.8 0.6 b .01* .50Instructions: successful deploymentof cognitive strategies

4.2 0.7 4.0 0.5 .36 .34 3.5 0.6 3.7 0.5 .08 .37 4.1 0.6 3.8 0.5 .02* .56

Note. IIEF = International Index of Erectile Function (Rosen et al. 1997); FSFI = Female Sexual Function Index (Rosen et al., 2000; Ter Kuile, Brauer & Laan. 2006). Questions from exit interview.Scales: Prior experience vibrostimulation: 1 (never) – 5 (very often); Pleasantness US: 1 (not pleasant at all) - 5 (very pleasant); US perceived as sexually arousing: 1 (not sexually arousing at all) –5 (very sexually arousing); Declared sexual arousal: 1 (not sexually aroused) – 5 (very sexually aroused); Instructions: Able to concentrate: 1 (trouble keeping concentrated) – 5 (well capablekeeping concentrated); Instructions: successful deployment of cognitive strategies: 1 (not successful at all) – 5 (very successful); Examples of what participants thought of in the Regulate conditionwhen presented with their CS+ are: seeing a blue sky with contrails, the sea, a yellow beach, or a yellow dessert. * p b .05.

6brom

etal.

Pleasecite

thisarticle

as:Mirte

Brom

,etal.,T

heInfluence

ofEmotion

Dow

n-Regulation

onthe

Expectation

ofSexualR

eward,

Behavior

Therapy

(2015),http://dx.doi.org/10.1016/j.beth.2015.01.005

7i n f luence of emot ion down-regulat ion on sexual reward

partial variance (ηp2) or as Cohen's d for paired

comparisons (Cohen, 1988). Data from the AATwere corrected for outliers. Median RTs were usedbecause they are less sensitive to outliers than means(see Cousijn et al., 2011). Bias scores (median push –pull) were computed for CS+, CS-, and the neutralpictures. A positive bias score will be referred tofurther as an approach-bias and a negative bias scoreas an avoid-bias. AAT bias scores were analyzedusing standard analysis of variance (ANOVA).

ResultsParticipants were randomly assigned to one of thetwo conditions with the restriction that conditionswere matched on sex as close as possible: Down-Regulate (N = 46;Men,n = 20) andAttend (N = 47;Men, n = 20); see Table 1 Subject characteristics.

genital sexual arousalPreconditioning PhaseFor all latency windows (FIR, SIR and TIR), nodifference in circumference following presentationof the CS+ and CS- was found, all ps N .42. Inaddition, for women, on all time latencies, no dif-ference in VPA following presentation of the CS+and CS- was found, all ps N .20.

Acquisition PhaseMen. Figure 2 summarizes penile circumference

(SIR) to CS+ and CS- across trials for the conditionsAttend and Down-Regulate. A main effect of Stim-ulus was found on FIR, F(1, 38) = 8.29, p b .01,ηp2 = .18; and SIR, F(1, 38) = 90.88, p b .01, ηp

2 =.71, indicating the vibrostimulation resulted in agenital response. In line with Brom, Laan, et al.(2014), penile circumference was smaller in re-sponse to the CS+ and vibrostimulation than inresponse to the CS-. On TIR no main effect ofStimulus was found, p = .23. No differences indifferential responding were observed between the

FIGURE 2 Mean penile circumference change scores (with stand(SIR) following the CS+ and CS- during the preconditioning phase,Attend and Down-Regulate. Note that during the acquisition phase

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

conditions, FIR p = .47; SIR p = .40; TIR p = .38,and no main effect of Condition was found, FIRp = .68; SIR p = .71; TIR p = .71.

Women. Figure 3 summarizes VPA (SIR) toCS+ and CS- across trials for both conditionsseparately. In line with previous studies (Both et al.,2008; Both, Laan, & Everaerd, 2011; Both, Brauer,& Laan, 2011), the 2 (Stimulus) × 10 (Trial) × 2(Condition) mixed ANOVA of VPA revealed nosignificant main effect of Stimulus on FIR, p = .07,but did on SIR, F(1, 51) = 18.77, p b .01, ηp

2 = .27,and TIR, F(1, 50) = 50.51, p b .01, ηp

2 = .50. AStimulus × Condition interaction was not found,FIR p = .15; SIR p = .15; TIR p = .34, nor ofStimulus × Trial × Condition, FIR p = .25; SIRp = .59; TIR p = .38.

Extinction PhaseMen. Analysis of the first extinction trial

revealed a significant main effect of Stimulus onFIR F(1, 38) = 4.19, p b .05, ηp

2 = .10; and SIR,F(1, 38) = 4.16, p b .05, ηp

2 = .10, indicating con-ditioned responding. A Stimulus × Condition inter-action was not found, FIR p = .27; SIR p = .25,TIR p = .30. Analysis of the entire extinction phaserevealed overall smaller penile responses to CS+than to CS-, as reflected by the significant maineffect of Stimulus on SIR, F(1, 38) = 4.29, p b .05,ηp2 = .10, indicating conditioned responding. No

interaction effects of Stimulus × Trial × Condition,and Stimulus × Condition were seen, all ps N .17.On FIR and TIR a significant interaction effect ofStimulus × Trial was found, FIR F(2, 79) = 3.46,p b .04, ηp

2 = .08; TIR F(2, 80) = 3.07, p b .05,ηp2 = .08, indicating extinction. On the last extinc-

tion trial no significant main effect of Stimulus wasfound, FIR p = .13, SIR p = .36, TIR p = .21.Analysis of only responses towards the CS+ duringthe preconditioning trials and the extinction trials

ard error bars) during the second interval response windowacquisition phase, and extinction phase for the two conditions, the response represents responding to the CS+ plus the US.

own-Regulation on the Expectation of Sexual Reward, Behavior

FIGURE 3 Mean vaginal pulse amplitude (VPA) change scores (with standard error bars) during the second interval responsewindow (SIR) following the CS+ and CS- during the preconditioning phase, acquisition phase, and extinction phase for the twoconditions Attend and Down-Regulate. Note that during the acquisition phase, the response represents responding to the CS+plus the US.

8 brom et al .

revealed no differences in conditioned respondingbetween the Attend and Down-Regulate condition,as reflected by nonsignificant Trial × Conditioninteractions, all ps N .10.

Women. On the first extinction trial no signif-icant main effect of Stimulus was found, FIR p =.45, SIR p = .35, TIR p = .47. No differences wereseen between the conditions, Stimulus × Condition,FIR p = .60; SIR, p = .88; TIR p = .98. Analysis ofthe entire extinction phase revealed no significanteffect of Stimulus, FIR p = .97, SIR p = .13, TIR,p = .71. Analysis of the preconditioning phase(MEAN precon 1–4) and the first extinction trialdemonstrated no significant main effect of Stimu-lus, with no differences between the two conditions,all ps N .31. On the last extinction trial, no sig-nificant main effect was found of Stimulus, FIR p =.74; SIR p = .61; TIR p = .54, and no differencestherein between conditions, all ps N .18.

subjective measuresPreconditioning PhaseThe 2 (Stimulus) × 4 (Trial) × 2 (Condition) × 2(Gender) mixed factor ANOVA to verify equallevels of responding to the CSs revealed no dif-ference in responding following presentation ofthe CS+ and CS- on affective value and subjectivesexual arousal and US expectancy between condi-tions and sexes, all ps N .05.

Extinction PhaseUS expectancy. As can be seen in Figure 4, both

conditions showed a robust increase of differentialresponding towards CS+ and CS- after the acquisi-tion phase, and both conditions showed a decrease inthis differential responding over trials. Indeed, the 2(Stimulus) × 4 (Trial) × 2 (Condition) × 2 (Gender)mixed factor ANOVA revealed a main effect ofStimulus, F(1, 86) = 227.09, p b .01, ηp

2 = .73, and

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

a significant interaction effect of Stimulus × TrialF(2, 212) = 43.97, p b .01, ηp

2 = .34. No differenceswere seen between conditions, Stimulus × Condi-tion, p = .73, and Stimulus × Trial × Condition,p = .59. An interaction of Stimulus × Gender wasobserved, F(1, 86) = 8.96, p b .01, ηp

2 = .09.Women in both conditions showed increased differ-ential responding towards the CS+ and CS- after theacquisition phase compared to men. Analysis of theextinction phase for men and women separately didnot reveal significant differences between the twoconditions, all ps N .18. Analysis of the first extinc-tion trial did not reveal differences in conditionedresponding between the two conditions, as reflectedby nonsignificant Stimulus × Condition interactions,men p = .25 and women p = .32, and neither didanalysis of the last extinction trial, men p = .78 andwomen, p = .15.

Affective value. As can be seen in Figure 5, menand women differed in conditioned respondingafter the acquisition phase. The 2 (Stimulus) × 4(Trial) × 2 (Condition) × 2 (Gender) mixed factorANOVA revealed a main effect of Stimulus, F(1,75) = 27.15, p b .01, ηp

2 = .27, and an interactioneffect of Stimulus × Trial, F(2, 166) = 4.05,p b .02, ηp

2 = .05. Also a significant interaction ofStimulus × Trial × Condition × Gender wasfound, F(2, 166) = 4,31, p b .02, ηp

2 = .05. Addi-tional analyses for men and women separatelyrevealed a significant main effect of Stimulus inmen, F(1, 32) =11.39, p b .01, ηp

2 = .26. Nointeraction of Stimulus× Trial was found, p = .36,indicating no extinction of conditioned respondingin men. A significant interaction was found ofStimulus × Trial × Condition, F(2, 75) = 3.31,p b .04, ηp

2 = .09, and as can be seen in Figure 5,the Down-Regulate condition demonstrated en-hanced extinction of conditioned respondingcompared to the Attend condition. Analysis of

own-Regulation on the Expectation of Sexual Reward, Behavior

FIGURE 4 US expectancy ratings (with standard error bars) following the CS+ and CS- during the preconditioning phase andextinction phase for men (top) and women (bottom) in the two conditions Attend (left) and Down-Regulate (right).

9i n f luence of emot ion down-regulat ion on sexual reward

the last extinction trial revealed no significantinteraction of Stimulus × Condition, p = .34, buta main effect was found of Stimulus, F(1, 37) =5.66, p b .03, ηp

2 = .13, indicating incompleteextinction of conditioned responding with nodifferences therein between conditions.

FIGURE 5 Subjective affect ratings (with standard error bars) folloextinction phase for men (top) and women (bottom) in the two c

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

For women a main effect of Stimulus was found,F(1, 43) = 20.01, p b .01, ηp

2 = .32, and an inter-action effect of Stimulus × Trial, F(2, 88) = 5.06,p b .01, ηp

2 = .11. Also a main effect of Conditionwas found, F(1, 43) = 4.41, p = .04, ηp

2 = .09. Ascan be seen in Figure 5, compared to the Attend

wing the CS+ and CS- during the preconditioning phase andonditions Attend (left) and Down-Regulate (right).

own-Regulation on the Expectation of Sexual Reward, Behavior

10 brom et al .

condition, women in the Down-Regulate conditiondemonstrated overall higher responses towards theCS+ and CS- in the extinction phase. No interactioneffects of Stimulus × Condition and Stimulus ×Trial × Condition were seen, ps N .33. Analysis ofthe first extinction trial for women revealed nosignificant interaction of Stimulus × Condition,p = .33. Analysis of the last extinction trial alsodid not reveal differences in conditioned differentialresponding towards the CS+ and CS- between thetwo conditions, p = .60. On this last trial there wasstill a main effect of Stimulus, F(1, 50) = 13.32,p b .01, ηp

2 = .21, indicating no extinction of con-ditioned differential responding.

Subjective sexual arousal. Figure 6 shows in-creased ratings of subjective sexual arousal towardsthe CS+ on the first trials of the extinction phase inmen and women. The 2 (Stimulus) × 4 (Trial) × 2(Condition) × 2 (Gender) mixed factor ANOVArevealed a significant main effect of Stimulus, F(1,81) = 23.22, p b .01, ηp

2 = .23, and an interactioneffect of Stimulus × Trial, F(3, 213) = 8.05,p b .01, ηp

2 = .09, and Stimulus × Trial × Condi-tion × Gender, F(3, 213) = 2.85, p b .05, ηp

2 = .03.Therefore, additional analyses were conducted formen and women separately. In men, no significantinteraction of Stimulus × Condition was found,p = .12, nor of Stimulus × Trial × Condition, p =.15. However, analysis of the last two extinction

FIGURE 6 Ratings (with standard error bars) of subjective sexualphase and extinction phase for men (top) and women (bottom) in

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

trials revealed a significant interaction of Stimulus ×Condition, F(1, 37) = 4.34, p b .05, ηp

2 = .11. Anal-ysis of the last extinction trial also revealed a sig-nificant Stimulus × Condition interaction, F(1, 38) =5.12, p b .03, ηp

2 = .12.For women, analysis of the extinction phase

revealed a significant Stimulus × Trial interactioneffect, F(2, 106) = 2.91, p b .01, ηp

2 = .15. No sig-nificant interaction effects of Stimulus × Condition,p = .38, or Stimulus × Trial × Condition, p = .19,were observed. A main effect of Condition wasseen, F(1, 45) = 4.16, p b .05, ηp

2 = .09. As can beseen in Figure 6, women in the Down-Regulatecondition demonstrated overall higher ratings ofsubjective sexual arousal towards both CS+ andCS- in the extinction phase, as compared withwomen in the Attend condition. Additional analysisof only the first extinction trial for men and womenseparately did not reveal a significant Stimulus ×Condition interaction, men p = .55, women p = .13.Analysis of the last preconditioning trial and the firstextinction trial for only CS+ responses, revealed amain effect of Stimulus in men, F(1, 37) = 8.39,p b .01, ηp

2 = .19, and women, F(1, 47) = 24.97,p b .01, ηp

2 = .35, and a significant interaction effectof Stimulus × Condition in women, F(1, 47) = 5.11,p b .03, ηp

2 = .10. The Down-Regulate conditiondemonstrated attenuated responding towards theCS+ compared to women in the Attend condition.Analysis of the last extinction trial did not reveal

arousal following the CS+ and CS- during the preconditioningthe two conditions Attend (left) and Down-Regulate (right).

own-Regulation on the Expectation of Sexual Reward, Behavior

11i n f luence of emot ion down-regulat ion on sexual reward

a significant Stimulus × Condition interaction, p =.28, but a significant main effect was observed ofStimulus, F(1, 50) = 6.84, p b .02, ηp

2 = .12, indi-cating there was still differential conditionedresponding on the last extinction trial, with no dif-ferences therein between conditions.

approach and avoidance tendencies

The preconditioning AAT bias scores were ana-lyzed with a mixed factor ANOVA with Genderand Condition as between-subjects factors andImage as within-subject factor with three levels(CS+, CS-, and neutral pictures). In line with theexpectations, no interaction effect of Image andCondition was found, p = .45. Men and womenalso did not seem to behave differently inapproach and avoidance tendencies towards thestimuli before the conditioning procedure, asreflected by the nonsignificant Image × Genderinteraction, p = .60.

FIGURE 7 Approach Avoidance Task (AAT) bias scoreswomen in the Attend and Down-Regulate condition (msconditioning. A positive score indicates faster reaction times o

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

The mixed factor ANOVA with Gender andCondition as between-subjects factors and Image aswithin-subject factor with three levels (CS+, CS-,and neutral pictures) and Trial as within-subjectsfactor with two levels (preconditioning and post-conditioning), of the AAT preconditioning and AATpostconditioning bias scores, revealed an interactionof Image × Trial × Gender, F(1, 127) = 22.07,p b .01, ηp

2 = .20. No Image × Trial × Conditioneffect was observed, p = .37. Analysis for men andwomen separately revealed no significant results formen, all ps N .31, whereas for women a significantImage × Trial interaction was found, F(2, 81) =61.52, p b .01, ηp

2 = .55.Analysis of only the postconditioning AAT scores

demonstrated a significant main effect of Image, F(2,137) = 55.97, p b .01, ηp

2 = .39, and of Image ×Gender, F(2, 137) = 52.64, p b .01, ηp

2 = .37. Nosignificant interaction of Image × Condition wasfound, p = .61. Analysis of postconditioning bias

for CS+, CS-, and neutral images in men (above) andwith standard error bars), preconditioning and postn approach (pull) trials compared to avoid (push) trials.

own-Regulation on the Expectation of Sexual Reward, Behavior

12 brom et al .

scores for men and women separately demonstrateda main effect of Image in women, F(1, 68) = 91.46,p b .01, ηp

2 = .64,whereas inmen it did not, p = .41.For women, also a main effect of Condition wasseen, F(1, 51) = 4.19, p b .05, ηp

2 = .08. Comparedto women in the Attend condition, women in theDown-Regulate condition had attenuated approachbiases towards all stimuli, as can be seen in Figure 7.

correlations between conditionedresponses

To investigate relationships between conditionedresponses, additional correlational analyses were

Table 2Correlations Between Conditioned Genital Response, ConditionedConditioned US Expectancy and Conditioned Approach and AvoWomen in the Attend and Regulate Condition

AffectiveValue

SubjectiSexualArousal

Men Attend Affective Value .67⁎⁎

Subjective Sexual Arousal .67⁎⁎

US Expectancy .45 .45⁎

Conditioned GenitalResponse SIR

-.19 .14

Conditioned GenitalResponse TIR

-.18 .14

Bias Score CS+ -.34 -.32Bias Score CS- -.39 -.12

Regulate Affective Value .49⁎

Subjective Sexual Arousal .49⁎

US Expectancy .11 .12Conditioned GenitalResponse SIR

.30 .14

Conditioned GenitalResponse TIR

.18 -.21

Bias Score CS+ -.17 .14Bias Score CS- .18 .13

Women Attend Affective Value .71⁎⁎

Subjective Sexual Arousal .71⁎⁎

US Expectancy .38 .40Conditioned GenitalResponse SIR

-.40 -.14

Conditioned GenitalResponse TIR

-.31 -.20

Bias Score CS+ .10 -.16Bias Score CS- -.17 -.31

Regulate Affective Value .32Subjective Sexual Arousal .32US Expectancy .13 .39Conditioned GenitalResponse SIR

-.18 .04

Conditioned GenitalResponse TIR

.22 .07

Bias Score CS+ -.20 .01Bias Score CS- .11 -.11

Note. ** Correlation is significant at the .01 level (2-tailed), * Correlation

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

conducted. We expected that the strength of theconditioned genital response would be positivelyrelated to the amount of change in subjective affectand subjective arousal and US expectancy. In ad-dition, it was expected that the strength of theconditioned genital response would be positivelyrelated to the CS+ bias score. To investigate theserelationships, for genital responses on SIR and TIRand ratings of affect, and subjective sexual arousaland US expectancy, the difference between the re-sponse to the CS+ and the CS− during the first trialin the extinction phase was calculated by subtract-ing the response to the CS− from the response to

Affective Change, Conditioned Subjective Sexual Arousal,idance Tendencies Towards the CS+ and CS- for Men and

ve USExpectancy

ConditionedGenitalResponse SIR

ConditionedGenitalResponse TIR

BiasScoreCS+

BiasScoreCS-

.45 -.19 -.18 -.34 -.39

.45⁎ .14 .14 -.32 -.12.14 .13 .01 .12

.14 .99⁎⁎ .07 .31

.13 .99⁎⁎ .06 .30

.01 .07 .06 .49⁎

.12 .31 .30 .49⁎

.11 .30 .18 -.17 .18

.12 .14 -.21 .14 .13-.01 -.13 .42 .13

-.01 .73⁎⁎ -.13 -.05

-.13 .73⁎⁎ .01 .07

.42 -.13 .01 .60⁎⁎

.13 -.05 .07 .60⁎⁎

.38 -.40 -.31 .10 -.17

.40 -.14 -.20 -.16 -.31-.11 -.15 -.15 .05

-.11 .48⁎ -.25 -.14

-.15 .48⁎ .15 .08

-.15 -.25 .15 .20.05 -.14 .08 .20.13 -.18 .22 -.20 .11.39 .04 .07 .01 -.11

-.26 .07 -.32 .18-.26 -.35 .26 .16

.07 -.35 -.33 -.23

-.32 .26 -.33 .12.18 .16 -.23 .12

is significant at the .05 level (2-tailed).

own-Regulation on the Expectation of Sexual Reward, Behavior

13i n f luence of emot ion down-regulat ion on sexual reward

the CS+. Pearson product-moment correlationsbetween genital difference scores, affect differencescore, subjective sexual arousal difference score,and US expectancy difference scores were calculat-ed (see Table 2). Table 2 shows that there wereno significant correlations between the strength ofthe conditioned genital response and conditionedsubjective and behavioral measures in men andwomen, in both the Attend and Regulate condition.

DiscussionThe present study is the first that included men andwomen in the same experimental conditioningdesign on emotion regulation, and it is remarkablethat a gender difference in subjective and behavioralsexual response was observed. First, the deploy-ment of a cognitive emotion down-regulationstrategy effectively enhanced extinction of condi-tioned affective value and subjective sexual arousalin men as compared to men in the Attend condition.This difference in enhanced extinction of condi-tioned subjective sexual arousal between the twoconditions in men is substantial given the foundeffect sizes. Intriguingly, in women no evidence wasfound that cognitive down-regulation results in en-hanced extinction of conditioned differential affectvalue or subjective sexual arousal towards the CS+and CS-. Surprisingly, and contrary to the expec-tations, women in the Down-Regulate conditiondemonstrated overall higher ratings of affectivevalue and subjective sexual arousal towards theCS+ and CS- in the extinction phase, comparedwith women in the Attend condition. Second, com-pared with an attend stimulus strategy, cognitivedown-regulation strategies resulted in attenuatedapproach tendencies towards conditioned stimulithat predicted potential sexual reward in women,but not in men. Although men demonstrated morerobust conditioned genital response, strong ap-proach tendencies were not observed. However,such tendencies need not per se translate into overtbehavior, since although emotions involve an auto-matic tendency to act, emotional impulses can beregulated by cognitive evaluation processes operat-ing under cognitive control (Frijda, 2010).It is crucial to mention that not all hypotheses

were confirmed. First, no evidence for cognitiveemotion down-regulation strategies to affect acqui-sition of conditioned genital response in men andwomen was found. Additionally, compared with anattend stimulus strategy, cognitive down-regulationstrategies did not result in decreased conditionedgenital sexual arousal, or subjective affect andsexual arousal in both sexes. Finally, it seems USexpectancy in men and women is not affected at allby cognitive emotion down-regulation strategies.

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

Results also showed that no significant correlationsexisted between the strength of the conditionedgenital, subjective, and behavioral response, withno differences therein between men and women.It is tempting to speculate that women may

indeed use less efficient cognitive strategies com-pared to men (Whittle et al., 2011). Results fromthe exit interview also revealed that womenexperienced more difficulties with the deploymentof the cognitive down-regulatory strategy. It ispostulated that in primary emotions, which arise asa result of processing innately significant environ-mental stimuli, like sexual cues, the limbic struc-tures are primarily involved. Secondary emotions—that are evoked by environmental and experientialstimuli that have acquired significance throughlearning—are thought to involve the additionalparticipation of the prefrontal and somatosensorycortices, which also function to modulate limbicsystem activation (Damasio, 1994; LeDoux, 2000).Research revealed that men rely more on prefrontaland somatosensory cortices (especially the dorso-lateral prefrontal cortex) during emotion regula-tion, whereas women rely more on limbic regionsincluding the left hippocampus, the left amygdalaand insular cortex (Kong et al., 2014; Whittle et al.,2011). The observed greater limbic activation inwomen (Whittle et al.) might suggest that theiremotional perception may be more of the primarythan the secondary type, and this may facilitatequicker and more accurate perception. In men,emotional perception may be more affected byregulatory and associative processes, leading toa greater ability to regulate emotions, includingsexual arousal (Whittle et al.). Research on theregulation of sexual arousal in men showed thatexperienced sexual arousal is associated withactivation in “limbic” and paralimbic structures,whereas inhibition of sexual arousal is associatedwith activation of the right superior frontal gyrusand right anterior cingulate gyrus (Beauregard,Lévesque, & Bourgouin, 2001). Intriguingly, noactivation was found in limbic areas duringinhibition of sexual arousal. Unfortunately, atpresent no imaging studies have been conductedthat have investigated down- or up-regulation ofsexual arousal in women. However, an imagingstudy by Klucken et al. (2014) revealed that theVal 158Met-ploymorphism in the Catechol-O-Methyl-Transferase (COMT) is associated with thealteration of neural processes of appetitive condi-tioning. Individuals who carried the Val/Val-alleledemonstrated increased hemodynamic responsesin the amygdala compared with the Met/Met-allelegroup in a differential conditioning paradigm.Although participants were not explicitly instructed

own-Regulation on the Expectation of Sexual Reward, Behavior

14 brom et al .

to use emotion-regulation strategies in this study,in Met/Met-allele carriers an increased effectiveamygdala-ventromedial prefrontal cortex connectiv-itywas found, and this could be regarded as amarkerfor altered emotion regulation during conditioning.These findings emphasize the importance of geneticvariations on appetitive conditioning and subsequentincreased vulnerability for addiction disorders ormaladaptive sexual behaviors.Given the problems in comparing genital re-

sponses of men and women directly, and possibledifferences between sexes with regard to responsesto specific types of stimulus materials, it is far tooearly to infer that women indeed are less efficientin down-regulation of positive (sexual) emotionsthan men. In addition, the effect of the emotionaldown-regulatory strategy in the present study isrelative to the other (Attend) strategy with whichit is compared and does therefore not reflect thecomplexities of the emotion-regulation repertoire(Aldao, 2013). Future studies should thereforeinvestigate if the found gender differences are alsoseen making use of multiple cognitive down-regulatory strategies (like cognitive reappraisal, orconcentrating on the neutral and asexual aspectsof the CSs). In addition, another limitation of thepresent study is the absence of a between-subjects(unpaired) control group. Without such a controlgroup it is difficult to determine whether and whatlearning has occurred. At present it is unclear if thedifferential response towards the CS+ and CS−wasdue to conditioning or to pseudo-conditioning. Thepossibility of sensitization of sexual arousal wouldtranslate into increased genital responses acrosstrials, and not in differential responding towardsthe CS+ and CS− per se (Domjan, 2010; Hoffmannet al., 2014). Therefore, making use of such acontrol group in future research is desirable.It is suggested that antecedent-focused strategies

like attentional deployment are more effective thanresponse-focused strategies (Gross & Thompson,2007). In the meta-analysis by Webb et al. (2012),passive distraction strategies (where participantsare provided with materials or a task that isunrelated to the emotion or emotional stimulus)had small effects on emotional outcomes, whereasactive distraction strategies (where the emphasisis on participants to bring to mind somethingunrelated to the emotion or emotional stimulus toserve as a distraction) had small-to-medium-sizedeffects. It was postulated that explicitly instructingparticipants to think about something unrelated tothe emotion is more effective than simply providinga distracting task. More investigations are neededas research on the regulation of positive emotionssuch as sexual arousal is extremely scarce. More-

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

over, the majority of the empirical investigationson emotion regulation (Aldao, 2013), including thepresent study, have examined processes in healthyindividuals, and only little attention has beendevotedto how those processes might differ as a functionof variability in psychopathology status. As it issuggested that personality facets and dispositionaland state-level psychological processes influenceemotion-regulatory processes (Aldao), an impor-tant venue for future research is the tailoring ofthe emotion-regulation strategies to the individualpatient.The present study is the first that found con-

ditioned genital response in men making use of atactile US. In line with Brom, Laan, et al. (2014),men showed a smaller penile circumference inresponse to the CS+ during the acquisition phaseand when vibrostimulation no longer was applied.Former research on automatic and controlled cog-nitive processing of sexual stimuli also found malegenital responses to be opposite to the predictions:genital responses towards sexually primed targetswere lower than responses to neutrally primedtargets (Janssen et al., 2000). Those results wereexplained by physiological processes of penile erec-tion. During the initial phases of erectile response,the penis undergoes an increase in length, and this isassociated with a simultaneous decrease in circum-ference. Therefore, the physiology of penile erectionmay also account for the results found in the presentstudy, with the smaller penile circumference inresponse to the CS+ reflecting the initial stage ofpenile erection.Our results suggest that in the treatment of

problematic strong sexual arousal and appetite,cognitive strategies in the processing of conditionedsexual stimuli may be helpful. It is important tomention that in the present study, instructions toregulate were given also during the acquisitionphase, which would not reflect how regulationinstructions would be offered in a clinical setting. Inthe treatment of problematic sexual arousal, clientsare taught regulation strategies after having devel-oped problematic behaviors via maladaptive con-ditioning. Nevertheless, learning to obtain effectiveemotion-regulation strategies in circumstances inwhich sexual stimuli cannot be avoided may beuseful to diminish undesirable feelings of sexualarousal and desire and to exert control over sexualbehavior. Therefore, future studies should incorpo-rate clinical samples, like individuals with hypersex-uality or deviant sexual preferences that manifestperturbed motivation. Second, as mentioned before,future conditioning studies should also make use ofa design in which the instructions to regulate aregiven only after acquisition has occurred, herewith

own-Regulation on the Expectation of Sexual Reward, Behavior

15i n f luence of emot ion down-regulat ion on sexual reward

resembling the clinical setting more closely. Still,results from the present study suggest that cognitiveemotion regulatory strategies may be more effectivein controlling unwanted sexual feelings than extinc-tion by cue-exposure treatment alone, as researchfrom our lab has shown that diminished sexualresponses can return (Brom, Laan, et al., 2014). Onthe other hand, in case of hyposexuality, increasingsexual arousal by making use of up-regulatory cog-nitive strategies may be effective. Therefore, futureresearch should investigate if cognitive up-regulatorystrategies can indeed be helpful in increasing sexualarousal elicited by conditioned sexual cues.

Conflict of Interest StatementThe authors declare that there are no conflicts of interest.

ReferencesAldao, A. (2013). The future of emotion regulation research:

Capturing context. Perspectives on psychological Science, 8,155–172.

Bancroft, J. H. J., & Janssen, E. (2000). The dual control modelof male sexual response: A theoretical approach to centrallymediated erectile dysfunction.Neuroscience& BiobehavioralReviews, 24, 571–579.

Bancroft, J. H. J., Jones, H. G., & Pullan, B. R. (1966). A simpletransducer for measuring penile erection, with commentson its use in the treatment of sexual disorders. BehaviourResearch & Therapy, 4, 239–241.

Beauregard, M., Lévesque, J., & Bourgouin, P. (2001). Neuralcorrelates of conscious self-regulation of emotion.The Journalof Neuroscience, 21, RC165.

Bindra, D. (1974). A motivational view of learning. Perfor-mance, and behavior modification. Psychological Review,81, 199–213.

Both, S., Laan, E., & Everaerd, W. (2011). Focusing "hot" orfocusing "cool": Attentional mechanisms in sexual arousalin men andwomen. Journal of Sexual Medicine, 8, 167–179.

Both, S., Brauer, M., & Laan, E. (2011). Classical conditioningof sexual response in women: A replication study. Journal ofSexual Medicine, 8, 3116–3131.

Both, S., Laan, E., Spiering, M., Nilsson, T., Oomens, S., &Everaerd, W. (2008). Appetitive and aversive classicalconditioning of female sexual response. Journal of SexualMedicine, 5, 1386–1401.

Brom, M., Both, S., Laan, E., Everaerd, W., & Spinhoven, P.(2014). The role of conditioning, learning and dopamine insexual behavior: A narrative review of animal and humanstudies.Neuroscience& Biobehavioral Reviews, 38, 38–59.

Brom, M., Laan, E., Everaerd, W., Spinhoven, P., & Both, S.(2014). Extinction and renewal of conditioned sexualresponses. PLoS ONE, 9(8), e105955, http://dx.doi.org/10.1371/journal.pone.0105955

Cohen, J. (1988). Statistical power analysis for the behavioralsciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Cousijn, J., Goudriaan, A. E., & Wiers, R. W. (2011). Reachingout towards cannabis: Approach-bias in heavy cannabis userspredicts changes in cannabis use.Addiction, 106, 1667–1674.

Damasio, A. R. (1994).Descartes’ Error: Emotion, reason, andthe human brain. New York: Grosset/Putnam.

Delgado, M. R., Gillis, M. M., & Phelps, E. A. (2008).Regulating the expectation of reward via cognitive strategies.Nature Neuroscience, 11, 880–881.

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

Domjan, M. (2005). Pavlovian conditioning: A functionalperspective. Annual Review of Psychology, 56, 179–206.

Domjan, M. (2010). The principles of learning and behavior(6th ed.). Belmont, CA: Wadsworth Cengage.

Frijda, N. H. (2010). Impulsive action and motivation.Biological Psychology, 84, 570–579.

Gross, J. J. (1998). Antecedent- and response-focused emotionregulation: Divergent consequences for experience, expres-sion, and physiology. Journal of Personality and SocialPsychology, 74, 234–237.

Gross, J. J. (2002). Emotion regulation: Affective. Cognitive,and social consequences. Psychophysiology, 39, 281–291.

Gross, J. J. (2007).Handbook of emotion regulation.New York:Guilford Press.

Gross, J. J., & Thompson, R. A. (2007). Emotion regulation:Conceptual foundations. In J. J. Gross (Ed.), Handbook ofemotion regulation (pp. 3–24). New York: Guilford Press.

Gutiérrez, G., & Domjan, M. (1997). Differences in the sexualconditioned behavior of male and female Japanese quail(Coturnix japonica). Journal of Comparative Psychology111, 135–142.

Heatherton, T. F., & Wagner, D. D. (2011). Cognitiveneuroscience of self-regulation failure. Trends in CognitiveScience, 15, 132–139.

Hoffmann, H., Goodrich, D., Wilson,M., & Janssen, E. (2014).The role of classical conditioning in sexual compulsivity: Apilot study. Sex Addict Compulsivity, 21, 75–91. http://dx.doi.org/10.1080/10720162.2014.895460

Hoffmann, H., Janssen, E., & Turner, S. L. (2004). Classicalconditioning of sexual arousal in women and men: Effects ofvarying awareness and biological relevance of the condi-tioned stimulus. Archives of Sexual Behavior, 33, 43–53.

Janssen, E., Everaerd, W., van Lunsen, R. H. W., &Oerlemands, S. (1994). Visual stimulation facilitates penileresponses to vibration in men with and without erectiledisorder. Journal of Consulting and Clinical Psychology, 62,1222–1228.

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.

Klucken, T., Kruse, O., Wehrum-Osinsky, S., Hennig, J.,Schweckendiek, J., & Stark, R. (2014). Impact of COMTVal158Met-polymorphism on appetitive conditioning andamygdala/prefrontal effective connectivity. Human BrainMapping. http://dx.doi.org/10.1002/hbm.22688

Klucken, T., Schweckendiek, J., Christian, J., Merz, C. J.,Tabbert, K., Walter, B., Kagerer, S., Vaitl, D., & Stark, R.(2009). Neural activations of the acquisition of conditionedsexual arousal: effects of contingency awareness and sex.Journal of Sexual Medicine, 6, 3071–3085.

Kong, F., Zhen, Z., Li, J., Huang, L., Wang, X., Song, Y., &Liu, J. (2014). Sex-related neuroanatomical basis of emotionregulation ability. Plos One, 9, e97071.

Kuzma, J. M., & Black, D. W. (2008). Epidemiology, prevalence,and natural history of compulsive sexual behaviour.PsychiatricClinics of North America, 31, 603–611.

Laan, E., & van Lunsen, R. H. W. (2002). Orgasm latency,duration and quality in women: Validation of a laboratorysexual stimulation technique. Poster presented at 28thConference of the International Academy of Sex Research,Hamburg, Germany.

Laan, E., Everaerd, W., & Evers, A. (1995). Assessment offemale sexual arousal: Response specificity and constructvalidity. Psychophysiology, 32, 476–485.

Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2005).International Affective Picture System (IAPS): Affective

own-Regulation on the Expectation of Sexual Reward, Behavior

16 brom et al .

ratings of pictures and instruction manual (Tech. ReportA-6). Gainesville: University of Florida, Center for Researchin Psychophysiology.

van Lankveld, J. J. D. M., van den Hout, M. A., & Schouten,E. G. W. (2004). The effects of self-focused attention: Per-formance demand and dispositional sexual self-consciousnesson sexual arousal of sexually functional and dysfunctionalmen. Behaviour Research and Therapy, 42, 915–935.

LeDoux, J. E. (2000). Emotion circuits in the brain. AnnualReview of Neuroscience, 23, 155–184.

Mak, A. K. Y., Hu, Z.-G., Zhang, J. X. X., Xiao, Z., & Lee,T. M. C. (2009). Sex-related differences in neural activityduring emotion regulation.Neuropsychologia,47, 2900–2908.

McRae, K., Ochsner, K. N., Mauss, I. B., Gabrieli, J. J. D., &Gross, J. J. (2008). Gender differences in emotion regulation:An fMRI study of cognitive reappraisal. Group Processes &Intergroup Relations, 11, 143–162.

Pavlov, I. (1927).Conditioned reflexes.Oxford:OxfordUniversityPress.

Pfaus, J. G., Kippin, T. E., & Centeno, S. (2001). Conditioningand sexual behavior: a review.Hormones and Behavior, 40,291–321.

Ray, R. D., & Zald, D. H. (2012). Anatomical insights intothe interaction of emotion and cognition in the prefrontalcortex.Neuroscience&BiobehavioralReviews, 36, 479–501.

Rosen, R., Brown, C., Heiman, J., Leiblum, S., Meston, C.,Shabsigh, R., Ferguson, D., & D’Agostino, R. (2000). TheFemale Sexual Function Index (FSFI): A multidimensionalself-report instrument for the assessment of female sexualfunction. Journal of Sex & Marital Therapy, 26, 191–208.

Please cite this article as: Mirte Brom, et al., The Influence of Emotion DTherapy (2015), http://dx.doi.org/10.1016/j.beth.2015.01.005

Rosen, R. C., Riley, A., Wagner, G., Osterloh, I. H.,Kirkpatrick, J., & Mishra, A. (1997). The internationalindex of erectile function (IIEF): A multidimensional scalefor assessment erectile dysfunction of erectile dysfunction.Urology, 49, 822–830.

Salemink, E., & van Lankveld, J. J. D. M. (2006). Increasingdistraction inhibits genital but not subjective sexual arousalof sexually functional and dysfunctional women. Archivesof Sexual Behavior, 35, 179–190.

Ter Kuile, M. M., Brauer, M., & Laan, E. (2006). The femalesexual function index (FSFI) and the Female sexual distressscale (FSDS): Psychometric properties within a Dutchpopulation. Journal of Sex& Marital Therapy, 32, 289–304.

Webb, T. L., Miles, E., & Sheeran, P. (2012). Dealing withfeeling: A meta-analysis of the effectiveness of strategiesderived from the process model of emotion regulation.Psychological Bulletin, 138, 775–808.

Whittle, S., Yücel, M., Yap, M. B. H., & Allen, N. B. (2011).Sex differences in the neural correlates of emotion: Evidencefrom neuroimaging. Biological Psychology, 87, 319–333.

Wiers, R. W., Eberl, C., Rinck, M., Becker, E. S., &Lindenmeyer, J. (2011). Retraining automatic action ten-dencies changes alcoholic patients’ approach bias foralcohol and improves treatment outcome. PsychologicalScience, 22, 490–497.

RECEIVED: October 21, 2014ACCEPTED: January 29, 2015Available online xxxx

own-Regulation on the Expectation of Sexual Reward, Behavior