Abstract
Background
Most problematic pornography use (PPU) research relies on cross-sectional designs. Ecological momentary assessment (EMA) studies are needed to better understand PPU correlates. We evaluated how daily PPU and pornography use ratings were correlated with baseline PPU, pornography use frequency (over 12-months), alcohol problems, and depression in a community sample of men living in the United States.
Design
Participants (n = 113, mean age 37.95 [11.09]) completed a baseline survey battery followed by daily EMAs over 14 days (k = 1,240 observations). Data were gathered on the Prolific panel.
Findings
Screener thresholds suggested 32%, 25%, and 24% of sample were at-risk for PPU, depression, and alcohol use problems respectively. Baseline estimates suggested PPU to be weakly correlated with depression (r = 0.29) and past 12-month pornography use (r = 0.27), but not alcohol problems (r = 0.08). Multilevel modeling indicated aggregated pornography use amount as a robust correlate of aggregated pornography control problems over the observation period. Baseline PPU and pornography use indicators were significant predictors of aggregated control problems and pornography use amount respectively. Baseline depression and alcohol problem predictors were non-significant. Past 12-month pornography use frequency and depression scores exacerbated the relationships between daily pornography use amount and aggregated control problems. Sensitivity analyses specific to only those who used pornography across all observations were consistent with full model results (though effect strength changed).
Conclusions
With exception to the PPU and pornography use indicators, cross-sectional baseline estimates (e.g., alcohol problems and depression) were generally not predictive of aggregated pornography use and control problems in the following 14 days.
The International Classification of Diseases 11th Edition (ICD-11) recently recognized compulsive sexual behavior disorder (CSBD) as a psychiatric disorder. Operationally, CSBD is a persistent pattern of failure to control intense, repetitive sexual impulses or urges resulting in repetitive sexual behavior that causes marked distress or impairment in personal, family, social, educational, occupational, or other important areas of functioning (World Health Organization, 2022). Formal CSBD acknowledgement has coincided with increased scientific inquiry (Grubbs et al., 2020). Preliminary estimates suggest 8–13% of men and 5–7% of women may meet CSBD diagnostic criteria (Turner et al., 2022). Problematic pornography use (PPU) is arguably the most common CSBD presentation (Grubbs et al., 2020, 2024; Kraus et al., 2018). Recent findings demonstrate masturbation and pornography use as the two most prevalent concerns for U.S. community members who self-identify as “addicted to” sexual behaviors/perceiving their sexual behavior to be “out of control” (Grubbs et al., 2024). One recent international epidemiological study (Bőthe et al., 2024) involving 84,243 participants across 42 countries reported 16.6% of participants as above clinical threshold on a popular PPU assessment (the Brief Pornography Screen; Kraus et al., 2020), with additional screeners showing 9.8% and 3.2% above clinical thresholds respectively.
Despite increased interest (Grubbs et al., 2020) and documented clinical need (Ballester-Arnal, Castro-Calvo, Giménez-García, Gil-Juliá, & Gil-Llario, 2020; Borgogna, Owen, Johnson, & Kraus, 2023; Griffin, Way, & Kraus, 2021; Grubbs et al., 2020; Kraus & Sweeney, 2019; Turner et al., 2022), several important CSBD conceptual and empirical questions remain. For example, there is disagreement regarding CSBD diagnostics (Borgogna & Aita, 2022; Brand et al., 2022; Bőthe, Koós, & Demetrovics, 2022; Castro-Calvo et al., 2022; Gola et al., 2022; Rumpf & Montag, 2022; Sassover & Weinstein, 2022). Currently, CSBD is classified as an impulse control disorder. However, many have presented arguments supporting CSBD addiction models, and argue that CSBD (and PPU) are behavioral addictions like gambling addiction (Grubbs et al., 2020; Liberg et al., 2022; Love et al., 2015; Stark & Klucken, 2017). Others argue the evidence is insufficient to support an addiction model (Sassover & Weinstein, 2022), with others positing that CSBD presentations could follow both impulse control and/or addiction frameworks, depending upon underlying etiology (Borgogna & Aita, 2022).
Connected to these conceptual concerns is the necessity for improved research. In particular, the CSBD/PPU literature is heavily based on cross-sectional surveys (Grubbs et al., 2020). Multiple authors have suggested the need for additional longitudinal studies (Borgogna & Aita, 2022; Grubbs et al., 2023; Grubbs & Kraus, 2021). Longitudinal research is helpful for understanding the variation that occurs within participants across time. That is, pornography use is not static. Similarly, PPU, psychological distress, and all the sequalae that accompany psychopathology vary within persons across time. As such, longitudinal designs are necessary to help examine such phenomena.
To date, a handful of longitudinal studies examining CSBD and/or PPU-related outcomes have been published (Castro-Calvo, Ballester-Arnal, Giménez-García, García-Barba, & Gil-Llario, 2023; Chen, Jiang, Luo, Kraus, & Bőthe, 2022; Grubbs, Perry, Weinandy, & Kraus, 2021; McGraw et al., 2024; Rosansky, Borgogna, Kraus, & Grubbs, 2022; Rousseau, Bőthe, & Štulhofer, 2021). For example, post-traumatic stress symptomology and positive urgency were both shown to weakly predict hypersexual behavior scores in those with histories of gambling six months after baseline assessments, while adjusting for several relevant covariates (Rosansky et al., 2022). Male gender was shown to predict perceived addiction to pornography above and beyond baseline pornography use frequency and personality indicators at a one-year survey follow-up (Grubbs, Wilt, Exline, Pargament, & Kraus, 2018). One observational study of Spanish individuals confusingly demonstrated one-year post assessment CSBD scores to be significantly lower at follow-up, while scores on compulsive engagement in online sexual activities increased (Castro-Calvo et al., 2023).
This prior research represents an important first step; however, the documented longitudinal studies have only included a few waves (often as little as two). Additionally, almost all of the prior studies have involved larger waves between assessment gathering time periods, such as months or years (Castro-Calvo et al., 2023; Grubbs et al., 2021; Rousseau et al., 2021). Ecological momentary assessment (EMA) studies represent an important growth area in CSBD study. EMA longitudinal studies are inherently different than the extant PPU/CSBD longitudinal literature. They involve many observations over the course of relatively shorter time periods (days or weeks). Importantly, EMA studies could help researchers better understand PPU over the course of a few days/weeks instead of after many months/years from baseline.
To our knowledge, only a few EMA CSBD/PPU studies have been published. One involved a small case series design of n = 9 treatment seeking males (Wordecha et al., 2018). Several have explored sexual behaviors in sexual minority men, with one exploring condomless sex and sexual arousal behaviors in a large (n = 334) sample of gay and bisexual men (Rendina, Millar, Dash, Feldstein Ewing, & Parsons, 2018), with another exploring how sexual behavior was associated with affect in a sample (n = 47) of men-who-have-sex-with-men (Grov, Golub, Mustanski, & Parsons, 2010). A similar study examined the relationship between affect and sexual behaviors in men-who-have-sex-with-men (n = 39) with/without hypersexual symptoms (Miner, Dickenson, & Coleman, 2019). An additional study (n = 176, 64% female) examined potential withdrawal effects in participants who were asked to abstain from pornography for seven days relative to a control group, no main effects related to group were observed (Fernandez, Kuss, Justice, Fernandez, & Griffiths, 2023). Altogether the body of literature specifically examining daily PPU and/or daily pornography use frequency over the course of multiple weeks is very limited.
The present paper reports results from an EMA study that involved pornography use and PPU indicators. Specifically, we examined daily pornography use and perceptions of problematic use in a sample of men living in the United States that were monitored over the course of two weeks. Men were of particular interest, as most research demonstrates men experience PPU to a greater degree relative to women (Grubbs et al., 2020; Turner et al., 2022). Because EMA methodology often requires single item indicators to represent much more complex latent variables, we chose to examine one specific facet of PPU that we thought was representative: control difficulties associated with pornography use. This dimension we chosen because CSBD is classified as an impulse control disorder. While other dimensions are often associated with PPU, such as negative emotions associated pornography use and functional impairment (Kor et al., 2014; Kraus et al., 2020), they are secondary effects of out-of-control use. We also gathered baseline PPU, alcohol use problems, and depression indicators. Alcohol use problems and depression were chosen as comparison variables because PPU is often described to covary with substance use (Kraus et al., 2020, 2021) and mood issues (Borgogna, Duncan, & McDermott, 2018; Kraus et al., 2020; Maddock, Steele, Esplin, Hatch, & Braithwaite, 2019). We aimed to examine how baseline batteries (often employed in cross-sectional settings) predict pornography use and PPU indicators of the course of 14 days.
Method
Procedure/participants
Data came from a larger daily diary project examining men's vaccination status, masculinity, and social media use. This is the first paper to be published from the dataset. The survey was designed in Qualtrics and administered to men across the United States via the Prolific service panel in June 2024. All measures in the baseline assessment were randomized. Sampling was based on convenience, but participants had to self-identify as cisgender men as this was a recruitment criterion of the larger project. Initially, 162 cisgender men completed baseline assessments. We removed men who failed initial attention checks or failed to complete the baseline survey (n = 12). The day following baseline, participants (n = 150) were sent EMAs designed to briefly (in less than five minutes) measure past-day mood/social media engagement/pornography use. Each EMA was initiated at approximately 5:00 p.m. central U.S. time and remained open until 10:00 p.m. EMAs were administered daily for 14 days following baseline. Participants were informed of when the EMAs would be active. For the purposes of this paper, we only included measures relevant to our aims (see measures section). Participants were reimbursed with $4 for completing baseline, $1 for each EMA, and received a $10 bonus if they completed all EMAs. Following the example of previous EMA researchers with a similar sample size and design, we only analyzed participants who completed at least five EMA surveys (Nezlek, Newman, & Thrash, 2017). Altogether, k = 1,240 responses were obtained nested in n = 113 participants with data fit for analyses. For demographic information see Table 1. All procedures received institutional review board approval. Analyses were not preregistered and should be considered exploratory.
Demographics
| n (%) | M (SD) | Range | |
| Age | 113 (100%) | 37.95 (11.09) | 18–78 |
| AUDIT (Alcohol Problems) | 5.81 (7.15) | 0–30 | |
| Abstainer | 22 (20%) | ||
| Low Risk | 64 (56%) | ||
| Hazardous Consumption | 13 (12%) | ||
| Moderate-Severe Alcohol Use Disorder | 14 (12%) | ||
| BPS (Problematic Pornography Use) | 2.19 (2.50) | 0–10 | |
| *Above Threshold on the BPS (At Risk for PPU) | 36 (32%) | ||
| Below Threshold on the BPS | 77 (68%) | ||
| Past 12-Month Pornography Viewing Frequency | 4.89 (2.00) | 0–7 | |
| PHQ-9 (Depression) | 6.32 (5.43) | 0–22 | |
| **Above Threshold on PHQ-9 (At Risk for Depression) | 28 (25%) | ||
| Below Threshold on PHQ-9 | 85 (75%) | ||
| Race | |||
| African American/Black and non-Hispanic | 17 (15%) | ||
| Caucasian/White and non-Hispanic | 52 (46%) | ||
| Asian or Asian American | 13 (12%) | ||
| Caucasian/White and Hispanic | 21 (19%) | ||
| African American/Black and Hispanic | 3 (3%) | ||
| Multiracial | 5 (4%) | ||
| Prefer not to say | 2 (2%) | ||
| Education | |||
| Less than High School | 1 (1%) | ||
| High School Diploma or GED | 35 (31%) | ||
| Associate's Degree | 10 (9%) | ||
| Bachelor's Degree | 50 (44%) | ||
| Master's Degree | 15 (13%) | ||
| Post-Master's | 2 (2%) | ||
| Religion | |||
| Agnostic | 42 (37%) | ||
| Christian | 27 (24%) | ||
| Catholic | 17 (15%) | ||
| Atheist | 14 (12%) | ||
| Other | 13 (12%) | ||
| Sexual Orientation | |||
| Heterosexual | 91 (81%) | ||
| Gay | 5 (4%) | ||
| Questioning | 1 (1%) | ||
| Bisexual | 13 (12%) | ||
| Asexual | 1 (1%) | ||
| Self-Identify | 1 (1%) | ||
| Pansexual | 1 (1%) | ||
| Aggregated Controls Problems (Grand Mean) | 0.72 (1.54) | 0–10 | |
| Aggregated Pornography Use (Grand Mean) | 0.99 (1.47) | 0–7 | |
Note: *Participant with score ≥4 on the Brief Pornography Screen (BPS), **Participants with scores ≥10 on the Patient Health Questionnaire-9 (PHQ-9). AUDIT = Alcohol Use Disorders Identification Test.
Measures
Baseline
Baseline pornography use frequency
Baseline pornography viewing frequency was assessed with a single item: “Within the past 12 months, on average, how often have you intentionally accessed pornography?” Participants selected from 0 (“Haven't intentionally accessed pornography in the past 12 months”), 1 (“A few times in the past 12 months”), 2 (“About once a month”), 3 (“Multiple times a month, but less than once a week”), 4 (“About once a week”), 5 (“Multiple times a week, but not daily”), 6 (“About daily”), and 7 (“Multiple times a day”). Pornography was defined to participants as “Viewing sexual activity, organs, and experiences for the purpose of sexual arousal” (Borgogna & McDermott, 2018; Kalman, 2008).
Problematic pornography use
The Brief Pornography Screen (BPS; Kraus et al., 2020) was used to assess PPU. Participants were provided the following prompt, “In the last six months, have any of these situations happened to you in regard to your use of pornography?” and rated the five items with 0 (“Never”), 1 (“Occasionally”), or 2 (“Very often”). Example item: “You find yourself using pornography more than you want to”. A score of ≥4 is considered the clinical cut-off for being at risk for PPU. Eleven participants reported not viewing pornography in the past 12 months, accordingly they were assigned scores of “0” on the BPS. Internal consistency in the current sample was α = 0.898.
Depression
The Patient Health Questionnaire-9 (PHQ-9; Kroenke, Spitzer, & Williams, 2001) was used to assess depression. Participants were provided the following prompt, “Over the last 2 weeks, how often have you been bothered by any of the following problems?” and rated the five items with 0 (“Not at all”), 1 (“Several days”), 2 (“More than half the days”), and 3 (“Nearly every day”). Example item: “Little interest or pleasure in doing things”. A score of ≥10 is considered a clinical cut-off for being at risk for a depressive disorder. Internal consistency in the current sample was α = 0.892.
Alcohol use disorder
The Alcohol Use Disorder Identification Test (AUDIT; Saunders, Aasland, Babor, De La Fuente, & Grant, 1993) was used to assess alcohol use disorder. Participants were provided the following prompt, “The following questions are about your use of alcohol. Your answers will remain confidential so please be honest. Please select which option best describes your answer to each question.” and rated eight items from 0 (“Never”), 1 (“Monthly or less”), 2 (“2–4 times a month”), 3 (“2–3 times a week”), and 4 (“4 or more times a week”), with two items being score 0 (“No”), 2 (“Yes, but not in the last year), or 4 (“Yes, during the last year”). Example item: “How often during the last year have you needed a drink in the morning to get yourself going after a heavy drinking session?”. Thresholds delineated by the World Health Organization include: 0 “Abstainer”, 1–7 “Low risk”, 8–14 “Hazardous consumptions”, ≥15, and “Moderate-severe alcohol use disorder”. Internal consistency in the current sample was α = 0.916.
EMA
Daily hour pornography use
Participants were asked daily to answer “yes” or “no” to the question: “In the past 24 h, did you view pornography?”. Our pornography definition from baseline was also provided.
Daily pornography use amount
Participants who acknowledged past 24-hours pornography use were asked: “In the past 24 h, how much time did you spend viewing pornography?” and given the options 0 (“None”), 1 (“0–10 min”), 2 (“10–20 min”), 3 (“20–30 min”), 4 (“30–40 min”), 5 (“40–50 min”), 6 (“50–60 min”), and 7 (“More than an hour”).
Pornography use control problems
Participants who acknowledged past 24-hours (24-h) pornography were asked: “In the past 24 hours, how much were you able to control your pornography use?” and given a slide bar from 0–10 with iterative digit markers with “0 = Could not control my pornography use, 10 = Complete control over my pornography use”). Participants who reported that they did not view pornography were coded “10”. This scale was reverse coded for all analyses such that higher scores indicated more control problems.
Analytic plan
We calculated bivariate correlations across baseline measures. Effect sizes of r > 0.2 were determined as the minimum for practical significance (Ferguson, 2009). We then constructed multilevel models (MLMs) with observations nested within participants. Day level predictors were estimates that were aggregated over the 14 day observation period and modeled as random slopes. All slopes in MLMs were unstandardized, with grand mean centered predictors. All models were estimated with maximum likelihood estimation. Model fit comparisons were made via the Akaike Information Criterion (AIC), where models with lower values were considered more parsimonious and therefore preferable (Burnham & Anderson, 2004). First, we conducted a series of models predicting aggregated pornography use controls problems. In these models, each baseline measure (BPS, AUDIT, PHQ-9, and past 12-month pornography frequency) were held constant and iteratively probed in interaction with aggregated pornography use amount. The second series of models followed an identical format, though aggregated pornography use amount became the outcome and aggregated control problems became the interacting variable. Models were built hierarchically such that baseline measures were entered, followed by aggregated predictors, followed by interactions. SPSS v.29 was used to calculate all demographic, univariate, and bivariate statistics. R using the lme4, flexplot, and tidyverse packages (Bates, Mächler, Bolker, & Walker, 2015; Fife, 2024; Wickham et al., 2019) were used to estimate all MLMs.
Ethics
Study was approved by the University of South Alabama IRB.
Results
Preliminary statistics
Univariate statistics revealed that at baseline the BPS, PHQ-9, and past 12-month pornography use frequency were all normally distributed with no outliers. The AUDIT was also approximately normally distributed with some positive skew (1.75) and one outlier (z = 3.38). Threshold observations revealed 25% as at risk for depression, 24% indicating at least “hazardous” levels of alcohol consumption, and 32% as at risk for PPU.
Bivariate statistics revealed that the BPS was significantly correlated with the PHQ-9 (r = 0.29, p = 0.002) and past 12-month pornography use frequency (r = 0.27, p = 0.004), but not the AUDIT (r = 0.08, p = 0.433). Past 12-month pornography use frequency was not significantly correlated with either the PHQ-9 (r = 0.13, p = 0.179) or AUDIT (r = −0.06, p = 0.549). The PHQ-9 was significantly correlated with the AUDIT (r = 0.43, p < 0.001).
Primary results
Predicting aggregated control problems
Results from models predicting aggregated control problems are available on Table 2, those predicting aggregated pornography use are available on Table 3. The null aggregated control problems model suggested that mean levels of PPU were low (b0 = 0.72) on average across the 14 days, though heterogeneity was evident. Inclusion of the BPS, PHQ-9, AUDIT, and past 12-month pornography viewing improved model fit. Baseline BPS scores were associated with a significant increase in the aggregated control problems indicator, as was the past 12-month pornography viewing frequency indicator. The PHQ-9 and AUDIT failed to demonstrate significant associations with aggregated control problems. When including aggregated pornography use as a random effect, model fit improved substantially. Additionally, all baseline indicators became non-significant or notably shrank (BPS), as aggregated pornography use emerged as a significant predictor (b = 0.88 [95% CI:.63, 1.12]), qualified by the heterogeneity (τ2 = 1.15).
Multilevel models predicting aggregated control problems over 14 Days
| Daily PPU | b0 | b | 95% CI | τ2 | AIC |
| Null | 0.72 | 0.43, 1.02 | 2.38 | 4378.40 | |
| Intercept | 0.72 | 0.49, 0.46 | 1.48 | 4352.79a | |
| Baseline BPS (Problematic Pornography Use) | 0.34 | 0.24, 0.44 | |||
| Baseline PHQ-9 (Depression) | −0.03 | −0.08, 0.02 | |||
| Baseline AUDIT (Alcohol Problems) | −0.02 | −0.06, 0.02 | |||
| Baseline Past 12-Month Pornography Use Frequency | 0.17 | 0.04, 0.29 | |||
| Intercept | 0.95 | 0.69, 1.20 | 1.30 | 3264.61a | |
| Baseline BPS (Problematic Pornography Use) | 0.03 | 0.01, 0.06 | |||
| Baseline PHQ-9 (Depression) | −0.01 | −0.02, 0.01 | |||
| Baseline AUDIT (Alcohol Problems) | 0.00 | −0.01, 0.01 | |||
| Baseline Past 12-Month Pornography Use Frequency | 0.01 | −0.02, 0.04 | |||
| Aggregated Pornography Use | 0.88 | 0.63, 1.12 | 1.15 |
Note: a Significant model improvement relative to null. BPS = Brief Pornography Screen, PHQ-9 = Patient Health Questionnaire-9, AUDIT = Alcohol Use Disorder Identification Test, AIC = Akaike Information Criterion.
Multilevel models predicting aggregated pornography use over 14 Days
| Daily pornography use amount | b0 | b | 95% CI | τ2 | AIC |
| Null | 0.99 | 0.71, 1.27 | 2.15 | 3907.52 | |
| Intercept | 0.99 | 0.76, 1.21 | 1.42 | 3887.61 | |
| Baseline BPS (Problematic Pornography Use) | 0.09 | −0.01, 0.18 | |||
| Baseline PHQ-9 (Depression) | −0.04 | −0.08, 0.01 | |||
| Baseline AUDIT (Alcohol Problems) | 0.01 | −0.03, 0.05 | |||
| Baseline Past 12-Month Pornography Use Frequency | 0.41 | 0.29, 0.53 | |||
| Intercept | 1.23 | 1.01, 1.45 | 0.95 | ||
| Baseline BPS (Problematic Pornography Use) | −0.08 | −0.16, 0.00 | |||
| Baseline PHQ-9 (Depression) | −0.03 | −0.07, 0.01 | |||
| Baseline AUDIT (Alcohol Problems) | 0.02 | −0.01, 0.05 | |||
| Baseline Past 12-Month Pornography Use Frequency | 0.34 | 0.24, 0.43 | |||
| Aggregated Control Problems | 0.80 | 0.58, 1.03 | 0.66 | 3265.91a |
Note: a Significant model improvement relative to null. BPS = Brief Pornography Screen, PHQ-9 = Patient Health Questionnaire-9, AUDIT = Alcohol Use Disorder Identification Test, AIC = Akaike Information Criterion.
Predicting aggregated pornography use
The null model also revealed relatively low (b0 = 0.99) daily levels of pornography use amount. Entering baseline predictors was associated with improved model fit, though past 12-month pornography viewing frequency was the only significant baseline predictor. When aggregated control problems was entered into the model as random effect it became a significant predictor (b = 0.80 [95% CI:.58, 1.03]), qualified by heterogeneity (τ2 = 0.66). Past 12-month pornography use frequency also remained a significant, albeit more modest, predictor in the full model (b = 0.34 [95% CI:.24, 0.42]), with model fit improving substantially.
Exploratory interactions
All the interaction models are available in Supplementary Files 1 (aggregated control problems as the outcome) and 2 (aggregated pornography use as the outcome). Two significant interactions were evident in the prediction of aggregated control problems. First, baseline BPS scores interacted with aggregated pornography use to significantly exacerbate the slopes of between aggregated pornography use and aggregated control problems. That is, for every one unit increase on baseline BPS there was an associated 0.14 (95% CI = 0.05, 0.23) increase in the slope between aggregated variables. A similar, but weaker interaction was also evident between the PHQ-9 and aggregated pornography use 0.05 (95% CI = 0.01, 0.10) in the prediction of aggregated control problems. All other interactions were non-significant.
Sensitivity analyses
Because participants who reported not viewing pornography for a specific day were coded such that they had complete control over their use for that day, a scenario emerged in which statistical independence was compromised for k = 825 observations (66.5%). Accordingly, we replicated our analyses on the subsample that was not affected by this methodological note. Of note, these results are specific to observed days where pornography was used. The corrected AIC (or AICc) was used in these models as a fit indicator.
Predicting aggregated control problems
Results from models predicting aggregated control problems are available on Table 4, those predicting aggregated pornography use are available on Table 5. The null model suggested that mean levels of control problems were still low but higher than the full sample (b0 = 2.11) across the 14 days, this average is qualified by large heterogeneity (τ2 = 6.00). Inclusion of the BPS, PHQ-9, AUDIT, and past 12-month pornography viewing improved model fit. Baseline BPS scores were associated with a significant increase in the aggregated controls problem indicator (slightly stronger than the full sample), though the baseline past 12-month pornography use frequency predictor became non-significant. The PHQ-9 and AUDIT remained null predictors. When including aggregated pornography use as a random effect, model fit improved substantially. The BPS remained significant (though, again, slightly stronger than the full sample) and daily pornography use amount emerged as a significant predictor though with a much more modest effect (b = 0.25 [95% CI:.09, 0.24]), with a low amount of heterogeneity (τ2 = 0.09).
Multilevel models predicting aggregated control problems over 14 Days (sensitivity analyses)
| Past 24-hour PPU | b0 | b | 95%CI | τ2 | AICc |
| Null | 2.11 | 1.54, 2.69 | 6.00 | 1613.93 | |
| Intercept | 2.15 | 1.58, 2.71 | 3.81 | 1603.04a | |
| Baseline BPS | 0.57 | 0.38, 0.76 | |||
| Baseline PHQ-9 | 0.04 | −0.06, 0.14 | |||
| Baseline AUDIT | 0.00 | −0.08, 0.07 | |||
| Baseline Past 12-Month Pornography Use Frequency | −0.17 | −0.52, 0.18 | |||
| Intercept | 1.83 | 1.23, 2.43 | 3.95 | 1587.58a | |
| Baseline BPS | 0.51 | 0.31, 0.71 | |||
| Baseline PHQ-9 | 0.04 | −0.06, 0.14 | |||
| Baseline AUDIT | 0.00 | −0.08, 0.07 | |||
| Baseline Past 12-Month Pornography Use Frequency | −0.26 | −0.61, 0.09 | |||
| Aggregated Pornography Use | 0.25 | 0.09, 0.24 | 0.09 |
Note: a Significant model improvement relative to null. BPS = Brief Pornography Screen, PHQ-9 = Patient Health Questionnaire-9, AUDIT = Alcohol Use Disorder Identification Test, AICc = Akaike Information Criterion (Corrected).
Multilevel models predicting aggregated pornography use over 14 Days (sensitivity analyses)
| Past 24-hour pornography use amount | b0 | b | 95% CI | τ2 | AICc |
| Null | 2.63 | 2.29, 2.98 | 2.07 | 1351.58 | |
| Intercept | 2.35 | 1.95, 2.74 | 1.80 | 1364.14 | |
| Baseline BPS | 0.14 | 0.01, 0.27 | |||
| Baseline PHQ-9 | −0.03 | −0.09, 0.04 | |||
| Baseline AUDIT | 0.02 | −0.03, 0.07 | |||
| Baseline Past 12-Month Pornography Use Frequency | 0.29 | 0.05, 0.54 | |||
| Intercept | 2.24 | 1.86, 2.61 | 1.36 | 1334.62a | |
| Baseline BPS | 0.01 | −0.13, 0.15 | |||
| Baseline PHQ-9 | −0.04 | −0.10, 0.03 | |||
| Baseline AUDIT | 0.04 | −0.01, 0.09 | |||
| Baseline Past 12-Month Pornography Use Frequency | 0.24 | 0.01, 0.47 | |||
| Aggregated Control Problems | 0.17 | 0.06, 0.29 | 0.06 |
Note: a Significant model improvement relative to null. BPS=Brief Pornography Screen, PHQ-9= Patient Health Questionnaire-9, AUDIT=Alcohol Use Disorder Identification Test, AICc=Akaike Information Criterion (Corrected).
Predicting aggregated pornography use
The null model also revealed relatively low (b0 = 2.63) aggregated levels of pornography use, qualified by heterogeneity (τ2 = 2.07). Though, again, slightly higher than the full sample. Model fit did not improve when including the baseline predictors in the model. Fit improved when the aggregated control problems indicator was entered into the model as a random effect, where it emerged as a significant predictor (b = 0.17 [95% CI:.06, 0.29] τ2 = 0.06), though much weaker than the full sample. Past 12-month pornography use frequency also remained significant, albeit more modest than the full sample (b = 0.24 [95% CI:.06, 0.29]). The interaction models were also examined, which remained largely null. The original interaction between the PHQ-9 and aggregated pornography use predicting aggregated control problems was significantly attenuated (b = 0.03 [95% CI:.00, 0.06]), while the interaction between BPS and aggregated pornography use became non-significant altogether. Interestingly, a new interaction emerged between baseline past 12-month pornography use frequency and aggregated control problems predicting aggregated pornography use amount, but the interaction was quite weak (b = 0.08 [95% CI:.01, 0.15]), and should be interpreted with caution. Details of all interaction models from the sensitivity analyses are available in supplemental files 3 and 4.
Discussion
This is one of the first studies to approach PPU research using an EMA design. Clinical cut-off base rates illustrate the importance of PPU research. Indeed, 32% of the sample were above the BPS at-risk threshold, whereas only 25% were above threshold for depression, and 24% above threshold for alcohol use problems. Thus, PPU was a present issue in our community sample of men living in the United States. The baseline pornography frequency indicator suggested participants, on average, viewed pornography more than once a week.
Baseline PPU, as measured by the BPS, was weakly correlated with depression and past 12-month pornography use frequency scores. Baseline PPU was not significantly correlated with alcohol problems. These null findings could be attributed to the relatively non-clinical nature of the sample. Samples of patients struggling with addiction issues may be more likely to demonstrate stronger correlations across PPU, mood, and substance use indicators (Kraus, Popat-Jain, & Potenza, 2021). As discussed elsewhere (Bőthe, Tóth-Király, Potenza, Orosz, & Demetrovics, 2020), the cross-sectional baseline correlations were so weak that PPU could not be accurately deduced from past 12-month pornography use estimates.
The MLMs provided novel information that heretofore have not been documented and contextualize the accuracy of cross-sectional measurements in the prediction of short-term pornography use outcomes. For the full sample models, pornography use over the span of 14 days emerged as a strong predictor of control problems over 14 days, with baseline indicators becoming non-significant predictors. The grand mean of control problems was relatively small, meaning on average, across the sample, out-of-control pornography use was not something that most participants reported. When pornography use occurred, and as the amount of time spent viewing pornography increased, the perception that it was out-of-control also increased. More literally, our EMA pornography use scaling was anchored in 10-minutes intervals (on a 0–7 scale, with 7 indicating “more than an hour”). So, for every ten minutes spent looking at pornography, there was on average an associated 0.88 increase in control problem ratings (on a scale of 0–10).
On average, daily pornography use amount was around 0–10 min a day. However, individuals who did not view pornography on a given day were fixed to “0” for that day. Since not everybody used pornography every day, the grand mean is artificially low relative to when participants used pornography. This estimate can be contrasted with our sensitivity analysis, which indicated average daily use around 20–30 min. Though this estimate is specific to only observations in which a participant did use pornography. Together, it can safely be assumed that when participants viewed pornography it was generally in less than 30-min increments for a given a day.
Importantly, past cross-sectional PPU research has demonstrated that effect sizes between PPU and correlates of interest drastically weaken when including non-pornography viewing participants, at least when considering moral incongruence correlates (Borgogna, Johnson, Shegedin, & Brasil, 2024). In our case, participants viewed pornography on some (but not all) days assessed, which presents a new methodological consideration for how to generalize findings. If we were to only evaluate observations in which pornography use occurred (i.e., the sensitivity analysis), the relationships between aggregated use and control problems weakened, though they generally trended in the original (positive) directions.
In the full sample, aggregated pornography use amount was the only significant predictor of aggregated control problems. However, the effect strength was inflated due to the compromised statistical independence associated with how constructs were measured when use did not occur (i.e., people who did not view pornography on a given day were assumed to have control over their use for that day). In the sensitivity analysis, the BPS baseline score emerged as a significant, and stronger, predictor of aggregated control problems. Accordingly, pornography use over 14 days is a good indicator of controls problems. That said, if you only consider measurements where pornography use occurred, the cross-sectional BPS total score would be a better predictor of control problems.
Not surprisingly, past 12-month pornography viewing frequencies were correlated with aggregate daily use amounts. Those who viewed pornography more days of the year, tended also to view pornography for longer time increments. The relationship was weak though in both the full sample and the sensitivity analysis. Concurrently, aggregated control problem ratings corresponded directly to use amounts, as did baseline BPS score.
Taken together, our results demonstrate the importance of EMA methodology in CSBD research. Cross-sectional baseline estimates, and the daily diary results, illustrate nuanced stories. Baseline estimates alone would indicate a weak relationship between past 12-month pornography use frequency and PPU. They would also suggest that PPU would be modestly associated with depression. This is informative and broadly consistent with the extant, heavily cross-sectional literature (Grubbs et al., 2020). However, the cross-sectional findings are contextualized by the addition of our EMA results. For example, in the EMA, depression problems were a very poor predictor, which is interesting given it reflected a more proximal measurement relative to aggregated control problems than baseline BPS. Similarly, baseline pornography use frequency failed to predict aggregated pornography control problems (both the full sample and sensitivity analyses). One place where cross sectional estimates and EMA findings were 100% consistent was regarding the role of alcohol problems, which never correlated with control problems or aggregated pornography use regardless of analysis.
Review of the interactions adds further light to these relationships. Broadly most of the interactions were non-significant across models. A few exceptions emerged. Baseline BPS scores were associated with an exacerbation effect between daily pornography use amount and aggregated control problems indicators. In other words, the more PPU someone perceived themselves as experiencing at baseline, the stronger the relationship between aggregated pornography use amount and aggregated control rating over the observation period. Though this interaction was not significant in the sensitivity analyses. A similar interaction was observed between baseline PHQ-9 scores and daily pornography use amount. As depression scores increased, they were associated with a strengthening of the relationship between agregated pornography amount and aggregated control problem ratings. Interestingly, this effect remained significant in the sensitivity analyses (though was attenuated). In line with the self-perceived problematic pornography use literature (Grubbs & Perry, 2019; Grubbs, Perry, Wilt, & Reid, 2019), those who consider themselves as having pornography problems, might more readily classify their use as problematic. Similarly, and consistent with a cognitive model of depression (Beck, 2002, 2008), those who experience more depression problems may be more likely to describe their use as problematic.
Additionally, in the sensitivity analyses an interaction emerged between baseline pornography use and aggregated control problems predicting aggregated pornography use such that as control problems increased the relationship between baseline pornography use and aggregated pornography use also increased. However, this effect was quite small, and baseline pornography use was such a weak predictor of aggregated pornography use that the boost largely does not matter. Indeed, all the interaction results should be interpreted cautiously given their relative size and susceptibility to type 1 error. We present them as a potential springboard for future research directions.
Limitations
Since we did not use an experimental design, firm casual conclusions regarding the identified relationships cannot be made. That said, the use of EMA is a strength relative to historical cross-sectional only designs. Our study also suffers from traditional biases associated with self-report and self-selection. We took steps to minimize these issues (e.g., measure randomization at baseline), but elements of method effect biases cannot be ruled out. Our results only generalize to relatively healthy men living in the United States. It should also be noted that while the sample was non-clinical by design, the amount of those responding at potentially clinical levels of PPU, depression, and alcohol problems were all higher than what might be expected in a normative community sample. Particularly, the BPS threshold score may be too liberal (and is therefore overrepresenting the prevalence of baseline PPU in our sample). We suggest future researchers continue to examine the BPS threshold score and refine as needed. Readers should be mindful to observe the different scalings associated with our measurements. For instance, the BPS reflects problems over the past six months, whereas our EMA was assessing past 24-h pornography use and control problems, which was then aggregated over 14 days. While we believe wording from both measures reflect “PPU”, it is important to highlight that they involve different time frame considerations during interpretations, which could be associated with some of the heterogeneity in our findings between the cross-sectional results and the MLM results. Notably, the daily control problems and daily pornography use amount measures were single items. While brief assessments are needed in EMA research, they are only capturing small pieces of otherwise complex constructs. For instance, if we had used a PPU measure anchored in the moral incongruence literature such as a “perceived addiction to pornography” item (Grubbs, Exline, Pargament, Hook, & Carlisle, 2015), we may have found different results.
Conclusions
PPU is not a static construct. Examining various facets of PPU, such as control problems, over multiple days will yield findings that are different from cross-sectional mainline batteries. Importantly, cross-sectional estimates could only weakly predict pornography use and control problems in the near term (over the 14 days following baseline). In the case of depression and alcohol problems, they were not predictive at all. Future researchers should continue employing EMA/longitudinal designs to explore these relationships in more depth with clinical populations. It is likely that our current understanding of CSBD and PPU, which is based on a strong foundation of cross-sectional literature (Grubbs et al., 2020), will improve as we study these constructs utilizing more complex designs.
Funding sources
Drs. Borgogna, Brasil, Levant, and McDermott all received funding from the American Psychological Association and the Centers for Disease Control and Prevention Award# 6NU87PS004366-03-02 to support the data used in this paper.
Authors' contribution
NCB conceived the aims and wrote initial drafts. JV and TO conducted analyses. KMB, SWK, RFL, JV, TO, and RCM all reviewed and edited the manuscript. NCB, KMB, RFL, and RCM obtained funding and designed the initial study that the data was derived from.
Conflict of interest
No conflicts of interest.
Supplementary material
Supplementary data to this article can be found online at https://doi.org/10.1556/2006.2025.00008.
References
Ballester-Arnal, R., Castro-Calvo, J., Giménez-García, C., Gil-Juliá, B., & Gil-Llario, M. D. (2020). Psychiatric comorbidity in compulsive sexual behavior disorder (CSBD). Addictive Behaviors, 107, 106384. https://doi.org/10.1016/J.ADDBEH.2020.106384.
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67, 1–48. https://doi.org/10.18637/jss.v067.i01.
Beck, A. T. (2002). Cognitive models of depression. In R. Leahy (Ed.), Clinical Advances in cognitive psychotherapy: Theory and application (pp. 29–61).
Beck, A. T. (2008). The evolution of the cognitive model of depression and its neurobiological correlates. American Journal of Psychiatry, 165, 969–977. https://doi.org/10.1176/APPI.AJP.2008.08050721/ASSET/IMAGES/T411F2.JPEG.
Borgogna, N. C., & Aita, S. L. (2022). Another failure of the latent disease model? The case of compulsive sexual behavior disorder • Commentary to the debate: “Behavioral addictions in the ICD-11”. Journal of Behavioral Addictions, 11(3), 615–619. https://doi.org/10.1556/2006.2022.00069.
Borgogna, N. C., Duncan, J., & McDermott, R. C. (2018). Is scrupulosity behind the relationship between problematic pornography viewing and depression, anxiety, and stress? Sexual Addiction & Compulsivity, 25, 293–318. https://doi.org/10.1080/10720162.2019.1567410.
Borgogna, N. C., Johnson, D. A. L., Shegedin, M., & Brasil, K. M. (2024). Taking another look at the pornography problems due to moral incongruence model: Does gender play a role? Archives of Sexual Behavior, 53, 703–714. https://doi.org/10.1007/S10508-023-02743-Y/TABLES/3.
Borgogna, N. C., & McDermott, R. C. (2018). The role of gender, experiential avoidance, and scrupulosity in problematic pornography viewing: A moderated-mediation model. Sexual Addiction & Compulsivity, 25, 319–344. https://doi.org/10.1080/10720162.2018.1503123.
Borgogna, N. C., Owen, T., Johnson, D., & Kraus, S. W. (2023). No magic pill: A systematic review of the pharmacological treatments for compulsive sexual behavior disorder. The Journal of Sex Research, 61, 1328–1341. https://doi.org/10.1080/00224499.2023.2282619.
Bőthe, B., Koós, M., & Demetrovics, Z. (2022). Contradicting classification, nomenclature, and diagnostic criteria of compulsive sexual behavior disorder (CSBD) and future directions: Commentary to the debate: “Behavioral addictions in the ICD-11.”. Journal of Behavioral Addictions, 11, 204–209. https://doi.org/10.1556/2006.2022.00030.
Bőthe, B., Nagy, L., Koos, M., Demetrovics, Z., Potenza, M. N., & Kraus, S. W. (2024). Problematic pornography use across countries, genders, and sexual orientations: Insights from the International Sex Survey and comparison of different assessment tools. Addiction, 119, 928–950. https://doi.org/10.1111/add.16431.
Bőthe, B., Tóth-Király, I., Potenza, M. N., Orosz, G., & Demetrovics, Z. (2020). High-frequency pornography use may not always be problematic. Journal of Sexual Medicine, 17, 793–811. https://doi.org/10.1016/j.jsxm.2020.01.007.
Brand, M., Rumpf, H.-J., Demetrovics, Z., Müller, A., Stark, R., King, D. L., … Potenza, M. N. (2022). Which conditions should be considered as disorders in the International Classification of Diseases (ICD-11) designation of “other specified disorders due to addictive behaviors”. Journal of Behavioral Addictions, 11, 150–159. https://doi.org/10.1556/2006.2020.00035.
Burnham, K. P., & Anderson, D. R. (2004). Multimodel inference: Understanding AIC and BIC in model selection. Sociological Methods and Research, 33, 261–304. https://doi.org/10.1177/0049124104268644.
Castro-Calvo, J., Ballester-Arnal, R., Giménez-García, C., García-Barba, M., & Gil-Llario, M. D. (2023). Natural course of compulsive sexual behavior (CSB): A 1-year follow-up study. International Journal of Mental Health and Addiction, 1–21. https://doi.org/10.1007/S11469-023-01061-7/TABLES/5.
Castro-Calvo, J., Flayelle, M., Perales, J. C., Brand, M., Potenza, M. N., & Billieux, J. (2022). Compulsive sexual behavior disorder should not be classified by solely relying on component/symptomatic features: Commentary to the debate: “Behavioral addictions in the ICD-11”. Journal of Behavioral Addictions, 11, 210–215. https://doi.org/10.1556/2006.2022.00029.
Chen, L., Jiang, X., Luo, X., Kraus, S. W., & Bőthe, B. (2022). The role of impaired control in screening problematic pornography use: Evidence from cross-sectional and longitudinal studies in a large help-seeking male sample. Psychology of Addictive Behaviors, 36, 537–546. https://doi.org/10.1037/ADB0000714.
Ferguson, C. J. (2009). An effect size primer: A guide for clinicians and researchers. Professional Psychology: Research and Practice, 40, 532–538. https://doi.org/10.1037/a0015808.
Fernandez, D. P., Kuss, D. J., Justice, L. V., Fernandez, E. F., & Griffiths, M. D. (2023). Effects of a 7-day pornography abstinence period on withdrawal-related symptoms in regular pornography users: A randomized controlled study. Archives of Sexual Behavior, 52, 1819–1840. https://doi.org/10.1007/S10508-022-02519-W.
Fife, D. (2024). Flexplot: Graphically based data analysis using “flexplot”. In R package version 0.20.6.
Gola, M., Lewczuk, K., Potenza, M. N., Kingston, D. A., Grubbs, J. B., Stark, R., & Reid, R. C. (2022). What should be included in the criteria for compulsive sexual behavior disorder? Journal of Behavioral Addictions, 11, 160–165. https://doi.org/10.1556/2006.2020.00090.
Griffin, K. R., Way, B. M., & Kraus, S. W. (2021). Controversies and clinical recommendations for the treatment of compulsive sexual behavior disorder. Current Addiction Reports, 8, 546–555. https://doi.org/10.1007/S40429-021-00393-5/TABLES/2.
Grov, C., Golub, S. A., Mustanski, B., & Parsons, J. T. (2010). Sexual compulsivity, state affect, and sexual risk behavior in a daily diary study of gay and bisexual men. Psychology of Addictive Behaviors, 24, 487–497. https://doi.org/10.1037/A0020527.
Grubbs, J. B., Exline, J. J., Pargament, K. I., Hook, J. N., & Carlisle, R. D. (2015). Transgression as addiction: Religiosity and moral disapproval as predictors of perceived addiction to pornography. Archives of Sexual Behavior, 44, 125–136. https://doi.org/10.1007/s10508-013-0257-z.
Grubbs, J. B., Hoagland, C., Lee, B. N., Grant, J. T., Davison, P. M., Reid, R. C., & Kraus, S. W. (2020). Sexual addiction 25 years on: A systematic review of empirical literature on compulsive sexual behavior and an agenda for future research. Clinical Psychology Review, 82, 101925. https://doi.org/10.1016/j.cpr.2020.101925.
Grubbs, J. B., & Kraus, S. W. (2021). Pornography use and psychological science: A call for consideration. Current Directions in Psychological Science, 30, 68–75. https://doi.org/10.1177/0963721420979594/ASSET/IMAGES/LARGE/10.1177_0963721420979594-FIG1.JPEG.
Grubbs, J. B., Lee, B., Floyd, C. G., Bőthe, B., Jennings, T., & Kraus, S. W. (2024). What is the “sex” in sex addiction? Problem behaviors reported among those endorsing compulsive sexual behavior. Sexual Health & Compulsivity, 31, 351–367. https://doi.org/10.1080/26929953.2024.2386518.
Grubbs, J. B., & Perry, S. L. (2019). Moral incongruence and pornography use: A critical review and integration. Journal of Sex Research, 56, 29–37. https://doi.org/10.1080/00224499.2018.1427204.
Grubbs, J. B., Perry, S. L., Weinandy, J. T. G., & Kraus, S. W. (2021). Porndemic? A longitudinal study of pornography use before and during the COVID-19 pandemic in a nationally representative sample of Americans. Archives of Sexual Behavior, 51, 123–137. https://doi.org/10.1007/S10508-021-02077-7.
Grubbs, J. B., Perry, S. L., Wilt, J. A., & Reid, R. C. (2019). Pornography problems due to moral incongruence: An integrative model with a systematic review and meta-analysis. Archives of Sexual Behavior, 48, 397–415. https://doi.org/10.1007/s10508-018-1248-x.
Grubbs, J. B., Reid, R. C., Bőthe, B., Demetrovics, Z., Coleman, E., Gleason, N., … Kraus, S. W. (2023). Assessing compulsive sexual behavior disorder: The development and international validation of the compulsive sexual behavior disorder-diagnostic inventory (CSBD-DI). Journal of Behavioral Addictions, 12, 242–260. https://doi.org/10.1556/2006.2023.00005.
Grubbs, J. B., Wilt, J. A., Exline, J. J., Pargament, K. I., & Kraus, S. W. (2018). Moral disapproval and perceived addiction to internet pornography: A longitudinal examination. Addiction, 113, 496–506. https://doi.org/10.1111/add.14007.
Kalman, T. (2008). Clinical encounters with internet pornography. The Journal of the American Academy of Psychoanalysis and Dynamic Psychiatry, 36, 593–618. https://doi.org/10.1521/jaap.2008.36.4.593.
Kor, A., Zilcha-Mano, S., Fogel, Y. A., Mikulincer, M., Reid, R. C., & Potenza, M. N. (2014). Psychometric development of the problematic pornography use scale. Addictive Behaviors, 39, 861–868. https://doi.org/10.1016/j.addbeh.2014.01.027.
Kraus, S. W., Gola, M., Grubbs, J. B., Hoff, R. A., Lew-Starowicz, M., Martino, S., … Potenza, M. N. (2020). Validation of a brief pornography screen across multiple samples. Journal of Behavioral Addictions, 9, 259–271. https://doi.org/10.1556/2006.2020.00038.
Kraus, S. W., Krueger, R. B., Briken, P., First, M. B., Stein, D. J., Kaplan, M. S., … Reed, G. M. (2018). Compulsive sexual behaviour disorder in the ICD‐11. World Psychiatry, 17, 109–110. https://doi.org/10.1002/WPS.20499.
Kraus, S. W., Popat-Jain, A., & Potenza, M. N. (2021). Compulsive sexual behavior and substance use disorders. In R. Balon, & P. Briken (Eds.), Compulsive sexual behavior disorder: Understanding, assessment, and treatment (pp. 21–34). American Psychiatric Association. www.webofmedical.com.
Kraus, S. W., & Sweeney, P. J. (2019). Hitting the target: Considerations for differential diagnosis when treating individuals for problematic use of pornography. Archives of Sexual Behavior, 48, 431–435. https://doi.org/10.1007/s10508-018-1301-9.
Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16, 606–613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x.
Liberg, B., Görts-Öberg, K., Jokinen, J., Savard, J., Dhejne, C., Arver, S., … Abé, C. (2022). Neural and behavioral correlates of sexual stimuli anticipation point to addiction-like mechanisms in compulsive sexual behavior disorder. Journal of Behavioral Addictions, 11, 520–532. https://doi.org/10.1556/2006.2022.00035.
Love, T., Laier, C., Brand, M., Hatch, L., Hajela, R., Love, T., … Hajela, R. (2015). Neuroscience of internet pornography addiction: A review and update. Behavioral Sciences, 5, 388–433. https://doi.org/10.3390/bs5030388.
Maddock, M. E., Steele, K., Esplin, C. R., Hatch, S. G., & Braithwaite, S. R. (2019). What is the relationship among religiosity, self-perceived problematic pornography use, and depression over time? Sexual Addiction & Compulsivity, 26, 211–238. https://doi.org/10.1080/10720162.2019.1645061.
McGraw, J. S., Weinandy, J. T. G., Floyd, C. G., Hoagland, C., Kraus, S. W., & Grubbs, J. B. (2024). Problematic pornography use and suicidal thoughts: Results from cross-sectional and longitudinal analyses. Psychology of Addictive Behaviors, 38, 728–738. https://doi.org/10.1037/ADB0000996.
Miner, M. H., Dickenson, J., & Coleman, E. (2019). Effects of emotions on sexual behavior in men with and without hypersexuality. Sexual Addiction & Compulsivity, 26, 24–41. https://doi.org/10.1080/10720162.2018.1564408.
Nezlek, J. B., Newman, D. B., & Thrash, T. M. (2017). A daily diary study of relationships between feelings of gratitude and well-being. Journal of Positive Psychology, 12, 323–332. https://doi.org/10.1080/17439760.2016.1198923.
Rendina, J. H., Millar, B. M., Dash, G., Feldstein Ewing, S. W., & Parsons, J. T. (2018). The somatic marker hypothesis and sexual decision making: Understanding the role of Iowa gambling task performance and daily sexual arousal on the sexual behavior of gay and bisexual men. Annals of Behavioral Medicine, 52, 380–392. https://doi.org/10.1093/ABM/KAX006.
Rosansky, J. A., Borgogna, N. C., Kraus, S. W., & Grubbs, J. B. (2022). Cross-sectional and longitudinal associations between posttraumatic stress symptoms and hypersexual behaviors among individuals who have gambled in their lifetimes. The Journal of Sexual Medicine, 19, 1813–1823. https://doi.org/10.1016/J.JSXM.2022.09.003.
Rousseau, A., Bőthe, B., & Štulhofer, A. (2021). Theoretical antecedents of male adolescents’ problematic pornography use: A longitudinal assessment. The Journal of Sex Research, 58, 331–341. https://doi.org/10.1080/00224499.2020.1815637.
Rumpf, H.-J., & Montag, C. (2022). Where to put compulsive sexual behavior disorder (CSBD)? Phenomenology matters: Commentary to the debate: “Behavioral addictions in the ICD-11.”. Journal of Behavioral Addictions, 11, 230–233. https://doi.org/10.1556/2006.2022.00039.
Sassover, E., & Weinstein, A. (2022). Should compulsive sexual behavior (CSB) be considered as a behavioral addiction? A debate paper presenting the opposing view. Journal of Behavioral Addictions, 11, 166–179. https://doi.org/10.1556/2006.2020.00055.
Saunders, J. B., Aasland, O. G., Babor, T. F., De La Fuente, J. R., & Grant, M. (1993). Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption-II. Addiction, 88, 791–804. https://doi.org/10.1111/J.1360-0443.1993.TB02093.X.
Stark, R., & Klucken, T. (2017). Neuroscientific approaches to (online) pornography addiction. In C. Montag, & M. Reuter (Eds.), Internet addiction (pp. 109–124). https://doi.org/10.1007/978-3-319-46276-9_7.
Turner, D., Briken, P., Grubbs, J., Malandain, L., Mestre-Bach, G., Potenza, M. N., & Thibaut, F. (2022). The World Federation of Societies of Biological Psychiatry guidelines on the assessment and pharmacological treatment of compulsive sexual behaviour disorder the World Federation of Societies of Biological Psychiatry guidelines on the assessment and pharm. Dialogues in Clinical Neuroscience, 24, 10–69. https://doi.org/10.1080/19585969.2022.2134739.
Wickham, H., Averick, M., Bryan, J., Chang, W., McGowan, L., François, R., … Yutani, H. (2019). Welcome to the tidyverse. Journal of Open Source Software, 4, 1686. https://doi.org/10.21105/joss.01686.
Wordecha, M., Wilk, M., Kowalewska, E., Skorko, M., Łapiński, A., & Gola, M. (2018). “Pornographic binges” as a key characteristic of males seeking treatment for compulsive sexual behaviors: Qualitative and quantitative 10-week-long diary assessment. Journal of Behavioral Addictions, 7, 433–444. https://doi.org/10.1556/2006.7.2018.33.
World Health Organization (2022). International statistical Classification of Diseases and related Health problems (11th ed.).
