Prevalence of first- and second-wave digisexualities in Germany and their relation to Compulsive Sexual Behavior: Findings from a National Online Survey

Abstract

Background

Problematic use of online pornography is considered a hallmark symptom of Compulsive Sexual Behavior Disorder (CSBD), but the role of other digisexual behaviors in CSBD remains largely unknown. Digisexuality encompasses a range of technologies mediating sexuality, such as online pornography and sexting (first-wave digisexualities), and simulating sexuality, such as virtual reality (VR-)pornography and highly realistic sex dolls/robots (second-wave digisexualities). The prevalence of these evolving behaviors across different generations and the relation to CSBD is still unclear.

Method

We conducted an online survey in 2023 with a sample of N = 3,564 individuals from Germany, selected to represent the German population by age, gender and regional distribution. The aim of the study was to investigate the prevalence and frequency of digisexual behaviors and their relationship with CSB assessed with the CSBD-19 scale.

Results

First-wave digisexualities showed prevalences ranging from 19% (sexting erotic pictures) to 66.1% (pornography consumption). Second-wave digisexualities showed lower prevalences, with 5.3% for VR-pornography use and 3.9% for the use of highly realistic sex dolls/robots. Emerging and technologically advanced digisexualities were predominantly reported by younger generations. Additionally, all surveyed digisexualities showed medium (r = 0.308; pornography use) to strong (r = 0.529; casual sex via dating apps) correlations with CSBD-19 scores.

Conclusions

These findings suggest that interventions and research for CSBD need to extend beyond online pornography and include a broader range of digisexual behaviors. As technologized sexual practices continue to evolve, new opportunities and challenges arise, underscoring the need for future research and tailored therapeutic approaches to address potential risks.

Introduction

Compulsive Sexual Behavior Disorder (CSBD) is increasingly recognized in clinical practice, with problematic online pornography consumption identified as one of its most prevalent symptoms (Wizła & Lewczuk, 2024). However, in recent years, other digital stimuli or devices for sexual stimulation have been emerging. While problematic pornography use is well-documented as a feature of CSBD, the prevalence and the relevance for CSBD of other emerging digisexual behaviors - such as the use of virtual reality (VR-)pornography or highly realistic sex dolls/robots - has not been studied so far. Recent research by Grubbs et al. (2024) underscores the shifting landscape of sexual behaviors associated with CSBD, emphasizing the importance of contemporary, population-based studies to better understand these dynamics.

Compulsive Sexual Behavior Disorder – Diagnostic criteria and conceptual debates

CSBD is characterized by persistent patterns of excessive, out-of-control sexual thoughts, urges, or behaviors that lead to significant distress or impairment in personal, social, or occupational functioning (WHO, 2022). Individuals with CSBD often struggle to regulate their sexual behavior despite experiencing severe consequences, such as strained relationships, financial difficulties, or legal repercussions (Kraus et al., 2018). Historically, concerns over out-of-control sexual behavior have been documented under terms like “nymphomania” or “satyriasis” long before the condition was formally conceptualized and digital media played any role (Fuss, Briken, & Klein, 2015). The medicalization of out-of-control sexual behaviors began in the 19th century with Krafft-Ebing's Psychopathia Sexualis (1886). The conceptualization of CSBD as a distinct clinical disorder is relatively recent and has evolved through ongoing debates about its classification and phenomenology, highlighting its shared impulsive, addictive, and compulsive features – traits that have been central to related concepts such as hypersexual disorder or sexual addiction (Fuss et al., 2019a, 2019b).

Prevalence estimates for CSBD range from 3 to 10% in the general population (Bőthe et al., 2020; Briken et al., 2022; Dickenson, Gleason, Coleman, & Miner, 2018; Lewczuk et al., 2022), with higher rates among men than women (Briken et al., 2022; Bőthe et al., 2020; Dickenson et al., 2018). The strong association between problematic online pornography consumption and other problematic behavior (Wizła & Lewczuk, 2024) is unsurprising, as online pornography is the most widely used digital medium for sexual stimulation. However, as technology advances, other accessible, affordable, and anonymous options will likely emerge, further facilitating digital sexual engagement. This development could become increasingly relevant for CSBD, particularly in clinical and diagnostic contexts.

The expansion of digisexual behaviors and their categorization

The internet provides a vast range of opportunities to engage in online sexual behaviors (Griffiths, 2012). With advancing technology, increasingly high-quality devices and innovations such as teledildonics (i.e. sex toys controlled by a human partner), AI-generated pornography, chatbots and dating apps further expand the possibilities for expressing sexual behavior. Interactions mediated by technology offer qualities like constant availability, affordability, and (at least partial) anonymity. Additionally, some of these options are for free and accessible to anyone with an internet connection, making them available to a wide audience. While this offers the possibility to access sexual stimuli easily, it may also carry the risk of facilitating out-of-control sexual behavior (Griffiths, 2012).

With the rapid advancement of digital technologies, researchers have introduced various umbrella terms to categorize sexuality-related, technology-driven behaviors, each with its own strengths and limitations. Döring et al. (2021) discuss conceptual frameworks that help structure different types of sexual interactions in digital contexts, offering a comprehensive overview. For instance, Technology-Mediated Sexual Interaction (TMSI) specifically addresses interpersonal sexual interactions facilitated by technology, such as video sexting. In contrast, digisexuality provides a broader framework for understanding how technology intersects with sexual behavior, encompassing a wide range of digital innovations in both sexual and relational contexts. Within this framework, researchers further differentiate between first-wave and second-wave digisexualities, representing distinct ways in which technology interacts with human sexuality (McArthur & Twist, 2017).

The concept of first- and second wave digisexualities

The core element of first-wave is that technology mediates and depicts the sexual relationship between human partners (McArthur cited in CBC radio, 2019; McArthur & Twist, 2017). This is understood by academics to include communication technologies that were not developed directly for sexual contacts (such as video call apps), but also popular internet dating websites and dating-apps. It also includes teledildonics and technologies where the human partner is not present or aware of the user, such as pornography, live camera interactions or live sex-chat websites.

Second-wave digisexualities are merely defined by their immersivity and the simulation of sex (McArthur & Twist, 2017). Human partners do not need to be present or involved, as their presence can be simulated through VR-pornography (Dekker, Wenzlaff, Biedermann, Briken, & Fuss, 2021) or replicated using highly realistic sex dolls and sex robots. Although scholars state that humanoids cannot yet adequately mimic the experience of sex with a real human partner (McArthur & Twist, 2017), studies on these simulations show a more nuanced picture (Dekker et al., 2021; Desbuleux & Fuss, 2023a, 2023b). Regular users of highly realistic human-like objects report anthropomorphizing their doll, and about half of the users report to be in a romantic and/or sexual relationship with their doll (Desbuleux & Fuss, 2023a, 2023b). VR-pornography users experience intimacy and interaction in the virtual reality, enhancing their sense of realism and presence through immersive visual cues (Dekker et al., 2021; McArthur & Twist, 2017).

Researchers suggest that many individuals have used at least some of the mentioned technologies (McArthur & Twist, 2017). As technological advancements continue to expand device usage, reliable data on engagement with these emerging technologies as a form of digisexuality remains limited, with most studies focusing on online pornography. That said, online pornographic images remain the most commonly used form of digisexual behavior (Gesselman, Kaufman, Marcotte, Reynolds, & Garcia, 2022). Previous studies have indicated that the prevalence of lifetime pornography use for sexual purposes ranges from 70 to 94% across Australia, North America, and Europe (as reported in Bőthe, Nagy, Koós, Demetrovics, & Potenza, 2024; Herbenick et al., 2020). Specifically, 84–85% of males and 54–57% of females reported lifetime use (Zarate, Allen, Kannis-Dymand, Karimi, & Stavropoulos, 2023). Regarding generational differences, a study by Price, Patterson, Regnerus, and Walley (2016) reported that younger cohorts exhibited higher lifetime prevalence rates of pornography use compared to older generations. Regarding the usage of dating apps, studies (Sawyer, Smith, & Benotsch, 2018; Smith, 2016) indicate that, as of 2015, 15% of adults had used at least one online dating website or mobile dating application. Among individuals aged 55–65 years, online dating usage doubled over the previous two years (increasing from 6% to 12%), while usage nearly tripled among those aged 18–24 (from 10 to 27%; Smith, 2016 as cited in Sawyer et al., 2018). For sexting (sending sexually explicit messages, images, or videos), most work focused on adolescents. However, a meta-analysis found that for Americans aged 18–29 years, 38% had sent sexts and 42% received sexts (Mori et al., 2020).

Research questions

While previous research has examined the prevalence of certain digisexual behaviors, newer forms of digisexuality remain understudied. With the rapid emergence of various digisexual behaviors, there is a notable gap in understanding the comparative prevalence and especially the interrelationships of these behaviors. This exploratory study aims to assess the prevalence of diverse digisexual behaviors within a sample designed to reflect key demographic characteristics of the German population (RQ I). Additionally, we seek to examine generational differences in the uptake and utilization patterns of these behaviors (RQ II), as shifts in technology may result in varying degrees of engagement across age groups. Given the potential transition from conventional pornography consumption (i.e., encompassing all common forms of pornography from traditional VHS-pornography and magazines to online pornography) to other forms of digisexuality, this study is particularly timely for understanding the possible links between these emerging behaviors and CSBD. Specifically, we aim to examine whether digisexual behaviors –such as virtual reality pornography, sexting, or the use of sex dolls/robots—may also function as expressions of or risk factors for CSBD (RQ III). By addressing these research questions, this study will highlight evolving trends in sexual health. Given the early stage of research on digisexuality and CSBD, this study was exploratory in nature and did not involve pre-registered predictions or a priori hypotheses.

Methods

Procedure

The survey was distributed in partnership with USUMA,¹ an independent survey and analysis service provider, and the professional online panel provider KANTAR. Various methods, including opt-in emails, e-newsletters, and internal and external social media platforms, were utilized for distribution. Data collection occurred between May 17 and June 1 2023 with the survey taking approximately 17 min to complete. Participants did not receive financial compensation for their participation. For this study, panelists from across Germany were recruited, using a controlled quota sample proportional to age, gender, and regional population at the federal state level. Due to systematic underrepresentation of those aged 75 and older in online panels, an adjustment weighting was applied to account for age, gender, and region. However, the sample cannot be considered fully representative of the adult German-speaking population, as random selection was not used in drawing the sample.

Participants

A total of 5,573 individuals expressed interest in the study by clicking on the survey link, though 178 did not give informed consent. Consequently, 5,395 participants declared that they were at least 18 years old, gave informed consent and proceeded to start the survey. Of these, 971 participants discontinued participation, with most (n = 735) exiting after the initial questions. Additionally, we excluded 333 participants who did not complete the quality control questions satisfactorily. Due to closed regional quotas, another 527 participants were excluded, resulting in a final sample of N = 3,564 participants who completed the survey. The sample was structured and controlled to proportionally reflect the population distribution by age, gender, and region at the federal state level in Germany.

Participants had an average age of 50.76 years (SD = 18.15). 48.6% identified as male, 50.7% as female. The majority identified as heterosexual (85.6%), 4.0% as bisexual and 2.7% as gay or lesbian. Most participants reported being in a relationship (62.7%). 44.0% indicated they were not religious, while Christianity was the largest religious affiliation, encompassing 48.3% of participants. For more information about sociodemographic data, see Table 1.

Table 1.

Sample characteristics

	M (SD)	n (%)
Age	50.76 (18.15)
Sex assigned at birth
Male		1,744 (48.9)
Female		1,818 (51.0)
Not specified/no response		2 (0.1)
Gender
Male		1,732 (48.6)
Female		1,808 (50.7)
Non-binary, gender fluid, gender queer		20 (0.6)
Other		3 (0.1)
Not specified/no response		1 (0.0)
Sexual orientation
Heterosexual		3,050 (85.6)
Gay or lesbian		95 (2.7)
Heteroflexible		63 (1.8)
Homoflexible		9 (0.3)
Bisexual		143 (4.0)
Queer		9 (0.3)
Pansexual		18 (0.5)
Asexual		32 (0.9)
I don't know yet		30 (0.8)
Not specified/no response		113 (3.2)
Relationship status
Single		1,328 (37.3)
In a relationship		2,235 (62.7)
Not specified/no response		1 (0.0)
Religion
Not religious		1,569 (44.0)
Spiritual		95 (2.7)
Christianity		1,722 (48.3)
Islam		110 (3.1)
Judaism		16 (0.4)
Buddhism		17 (0.5)
Hinduism		13 (0.4)
Other		22 (0.7)
Not specified/no response		1 (0.0)
Educational level
No formal qualification (yet)		17 (0.5)
Lower secondary school certificate		566 (15.9)
Intermediate secondary school certificate		1,507 (42.3)
Higher secondary school certificate		1,454 (40.8)
Not specified/no response		21 (0.6)
Political orientation	3.07 (4.73)
Size of current place of residence
Metropolis (over 1 million inhabitants)		393 (11.0)
Large city (between 100,000 and 999,999 inhabitants)		856 (24.0)
City (between 1,000 and 99,999 inhabitants)		1,848 (51.9)
Village (less than 1,000 inhabitants)		467 (13.1)

Note. Total N = 3,564. Further differentiated religious groups were summarized due to the small group size.

The information on the education level is based on the German education system and has been translated into English without corresponding to an English-speaking school system. Participants were asked for their highest educational degree.

Mean value for political orientation is based on a 7-point scale, ranging from 1 = politically left to 7 = politically right.

Reported sociodemographic data is weighted proportionally according to age, gender and regional distribution in Germany.

Measures

The questionnaire included 37 single questions along with additional question sets grouped by topic. Various topics concerning human sexuality were gathered for analysis in separate studies.

Sociodemographics

In addition to other variables, participants were asked to indicate their sex assigned at birth, their gender identity and their sexual orientation. Participants were also asked to describe their relationship status and their religious affiliation (“What is your current religion?”). Due to the small sample sizes for some religious groups, we combined them into a single category labeled “Other”. Age was indicated by years (“How old are you?”). We also asked participants for their highest level of education they had completed. Participants were asked to indicate their political view on a 7-point scale from (1) left-wing to (7) right-wing and the size of their current residency.

Generations

We created categories to classify participants according to their generation. Based on their age we calculated the year of birth and formed the following categories: Silent Generation (born between 1928 and 1945; 2.7%), Boomers (born between 1946 and 1964; 30.1%), Generation X (born between 1965 and 1980; 28.9%), Millennials (born between 1981 and 1996; 25.6%) and Generation Z (born from 1997 to 2012; 12.7%). This division is based on previous publications (Pew Research Center, 2019).

Digisexualities

When selecting the behaviors for our study, we aimed to include both first- and second-wave digisexualities. According to McArthur and Twist (2017), the most important second-wave digisexual behaviors are VR-pornography and the use of highly realistic sex robots or their predecessors, such as sex dolls. Given the wide range of first-wave digisexual behaviors, we had to make a strategic selection to keep the questionnaire manageable. To provide a baseline reference, we included conventional pornography, the most widely studied form of digital sexual behavior. Additionally, we selected sexting (both fantasies and image-based sexting) and dating app use, as preliminary studies suggest these behaviors are relatively common (Gesselmann et al., 2022; Mori et al., 2020). By focusing on behaviors with higher prevalence, we aimed to assess their potential role in CSBD more effectively.

Following previous publications (Bőthe et al., 2023), we asked participants whether they had ever used the particular digisexuality, asking about lifetime prevalence (e.g., „Have you ever used pornography?”). If participants answered “yes”, we then asked them to specify how often they had used the technology in the past 12 months (e.g., “How often have you used pornography in the past year (in the last 12 months)?”). Here, the participants could choose between: 1 = never, 2 = once in the last year, 3 = 2–6 times in the last year, 4 = 7–11 times in the last year, 5 = once a month, 6 = 2–3 times per month, 7 = once per week, 8 = 2–3 times per week, 9 = 4–5 times per week, 10 = 6–7 times per week and 11 = more than 7 times per week. We used this question format for pornography use [“Pornography”], use of virtual reality pornography [“VR-pornography”], use of highly realistic sex dolls or sex robots [“sex doll/robot”], sexting of erotic images of one's own body [“sexting pics”], and sexting of sexual fantasies or topics [“sexting fantasies”]. We also asked about the use of dating apps or dating portals such as Tinder, Joyclub or Grindr [“Dating App”]. If they answered “yes” to the latter, we asked them two questions: how often they had used the apps in the past 12 months and how often they had met people for casual sex via the apps in the past 12 months. We defined pornography, sexting and dating apps as first-wave digisexualities and the use of highly realistic sex dolls and/or sex robots as well as VR-pornography as second-wave digisexualities.

CSBD

Compulsive sexual behavior was assessed using the CSBD-19 scale (Bőthe et al., 2020). The validated 19-item Scale measures compulsive sexual behavior based on the ICD-11 diagnostic guidelines for CSBD): control (three items; i.e., failure to control CSB), salience (three items; i.e., CSB being the central focus of one's life), relapse (three items; i.e., unsuccessful efforts to reduce CSB), dissatisfaction (three items; i.e., experiencing less or no satisfaction from sexual behaviors), and negative consequences (seven items; i.e., CSB generating clinically significant distress or impairment) (Bőthe et al., 2020). Participants were asked to answer each item on a scale ranging from 1 (Completely disagree) to 4 (Completely agree). The general score is obtained from the sum of the items with a minimum of 19 and a maximum of 76 points. A cut-off score of 50 or higher indicates a high risk for CSBD. At this threshold, the sensitivity was 98.5%, and the specificity was 99.1% (Bőthe et al., 2020). The scale shows very good internal consistency (α = 0.94, ω = 0.94; see for example Wizła & Lewczuk, 2024).

Statistical analysis

Statistical analyses were conducted using IBM SPSS Statistics 29.0 (IBM Corp., Armonk, NY). Descriptive statistics were used to present the total number and percentage of various sociodemographic variables, as well as prevalence measures of digisexual behaviors. For these prevalence measures, means were calculated for the entire sample and separately for males, females, and each generation. Spearman correlation analysis was then conducted to examine the associations between the CSBD total score and the frequency of pornography use, VR-pornography use, dating app use, casual sex via dating apps, sex dolls/robots use, and sending erotic images and fantasies (sexting). Statistical significance was set at a p-value of <0.01 (two-sided). Following Cohen's (1988) guidelines, we interpret Spearman's correlation coefficients (r) as follows: small effect (r = 0.10–0.29), moderate effect (r = 0.30–0.49), and large effect (r ≥ 0.50).

Ethics

The study procedures were conducted following the Declaration of Helsinki. The ethics committee of the Psychotherapeutenkammer Hamburg approved the study design. All participants were informed about the study, provided informed consent to participate in the study, and to have the data used for scientific publication.

Results

Prevalences of first- and second-wave digisexualities (RQ I)

Pornography use had the highest lifetime prevalence (66.1%), with gender differences between men (84.9%) and women (47.9%). Among the other digisexualities, the prevalence of lifetime use was less frequent but more prevalent for first-wave digisexualities, including sexting of erotic fantasies (30.1%), the use of dating apps (28.7%), and the sending of erotic pictures (19.0%). Second-wave digisexualities were the least frequent, including the use of VR-pornography (5.3%) and the use of highly realistic sex dolls or robots (3.9%). All digisexualities, but particularly those of the second-wave, were more prevalent among men than women. Overall, 257 participants (7.2%) scored 50 or higher on the CSBD-19 scale, indicating a risk of CSBD. For more information see Table 2.

Table 2.

Lifetime prevalence for digisexual behaviors/digisexualities

	Total sample	Male	Female	Gen. Z	Millennials	Gen. X	Boomers	Silent G.
	% (n)
Pornography	66.1% (2,354)	84.9% (1,481)	47.9% (871)	70.0% (323)	72.5% (644)	68.8% (647)	57.5% (636)	60.9% (103)
VR-pornography	5.3% (189)	8.9% (155)	1.8% (33)	16.6% (76)	9.1% (81)	2.7% (25)	0.5% (6)	0.6% (1)
Dating App	28.7% (1,024)	33.0% (575)	24.7% (449)	50.2% (230)	47.9% (425)	26.5% (250)	10.0% (111)	5.3% (9)
Sex doll/robot	3.9% (138)	6.5% (113)	1.4% (25)	9.6% (44)	7.4% (66)	1.7% (16)	0.8% (9)	1.2% (2)
Sexting pics	19.0% (677)	22.0% (383)	16.2% (294)	38.2% (175)	33.7% (300)	15.0% (141)	5.1% (57)	3.0% (5)
Sexting fantasies	30.1% (1,073)	35.5% (618)	25.0% (455)	52.1% (238)	49.7% (441)	27.2% (256)	11.4% (126)	7.1% (12)

Note. Total N = 3,564.

Generations based on the following: Silent Generation (born between 1928 and 1945; 4.7%), Boomers (born between 1946 and 1964; 31.1%), Generation X (born between 1965 and 1980; 26.4%), Millennials (born between 1981 and 1996; 24.9%), Generation Z (born from 1997 to 2012; 12.9%).

%-values and n-values indicating participants agreeing on the items (“Yes”).

Percentages always refer to the subgroup of the population. The percentage therefore indicates how many of the people in a subgroup (male, female, Generation Z – Silent Gen.) responded in the affirmative (= “Yes”) to the question.

Reported data is weighted proportionally according to age, gender and regional distribution in Germany.

Generational differences in the use of digisexualities (RQ II)

Regarding generational differences, Millennials reported the highest prevalence of pornography use at 72.5%. However, overall pornography use appears to be relatively stable across generations, with the lowest prevalence of 57.5% (Boomers). In contrast, the frequency of use for other technologies generally demonstrated a declining trend with increasing age across generations. For example, while 50.2% of Generation Z reported a use of Dating Apps, this was only reported by 5.3% of the Silent Generation. For second-wave digisexualities, prevalences decreased from 16.6% (VR-pornography; Gen Z) to 0.5% (Boomers). For the use of highly realistic sex dolls or sex robots, the prevalence decreased from 9.6% (Gen Z) to 0.8% (Boomers).

Digisexual behaviors correlate with CSBD-19 score (RQ III)

The mean CSBD-19 score among participants was 29.37 (SD = 11.35). The frequency of engagement with different digisexual behaviors varied. Conventional pornography use had a mean score of 4.42 (SD = 2.9), indicating that the majority of participants fell within the range of occasional to moderate usage, with an average frequency of approximately once per month. The mean score for virtual reality pornography (VR-pornography) was 4.69 (SD = 2.86). This demonstrates that although the number of users differs significantly, with conventional pornography being used by a considerable proportion (66.1%) and VR-pornography by a relatively small number (5.3%), the frequency of use among those who engage with each behavior is approximately similar, with comparable average usage values. The mean frequency for dating app usage was 3.60 (SD = 3.01), with casual sex via dating apps showing a lower mean of 2.32 (SD = 2.31). Sex doll use exhibited a mean score of 5.21 (SD = 3.11), suggesting occasional engagement (2–3 times a month) among a limited subset of participants (n = 138). These results suggest a broad spectrum of engagement in digisexual behaviors, with pornography use being a widely reported behavior, while emerging technologies such as VR-pornography and sex doll use were less common.

For Spearman Correlation analysis, we correlated the CSBD-19 score with the frequency of different digisexual behaviors within the past 12 months. The correlation analyses revealed significant positive associations between the CSBD-19 score and various digisexual behaviors. Specifically, CSBD-19 scores were moderately correlated with the frequency of conventional pornography use (r = 0.308, p < 0.01), dating app usage (r = 0.312), sexting fantasies (r = 0.401), VR-pornography use (r = 0.420), sexting pictures (r = 0.421) and sex doll use (r = 0.452). CSBD-19 also highly positively correlated with the frequency participants had met others for non-committal sex via the apps in the past 12 months (r = 0.529).

Furthermore, the frequency of the use of one digisexuality was positively correlated with the frequency of the use of another digisexual device. A strong correlation was found between the second-wave digisexualities VR-pornography use and sex doll use (r = 0.827), suggesting a tendency for individuals engaging in one form of advanced digisexuality to also engage in others. The frequency of non-VR-pornography use was also highly correlated with sex doll use (r = 0.588) and moderately positively correlated with, e.g., VR-pornography frequency (r = 0.465), suggesting overlaps in consumption patterns among these behaviors. Additionally, high correlations were observed between sexting pictures and sexting fantasies (r = 0.773), indicating that participants engaging in one form of sexting were likely to engage in the other. For more information see Table 3. An overview of the correlations between various digisexual behaviors and CSBD, divided by sex, can be found in Table 4.

Table 3.

Correlation between frequency of digisexualities and compulsive sexual behavior

	Range	M	SD	Mdn.	1	2	3	4	5	6	7	8
1. CSBD	19–76	29.37	11.35	25.0	1	0.308** [0.259–0.355]	0.420** [0.246–0.568]	0.312** [0.237–0.384]	0.529** [0.468–0.585]	0.452** [0.252–0.615]	0.421** [0.334–0.5	0.401** [0.332–0.466]
2. Pornography	1–11	4.42	2.91	4.0		1	0.465** [0.285–0.613]	0.386** [0.306–0.461]	0.304** [0.22–0.385]	0.588** [0.399–0.729]	0.364** [0.268–0.453]	0.416** [0.343–0.483]
3. VR-pornography		4.69	2.86	4.0			1	0.640** [0.48–0.759]	0.605** [0.435–0.733]	0.827** [0.712–0.899]	0.708** [0.559–0.812]	0.556** [0.378–0.695]
4. Dating App		3.60	3.01	2.0				1	0.592** [0.537–0.643]	0.701** [0.543–0.811]	0.571** [0.483–0.647]	0.556** [0.48–0.623]
5. Casual sex		2.32	2.31	1.0					1	0.629** [0.446–0.762]	0.631** [0.553–0.699]	0.562** [0.487–0.629]
6. Sex doll		5.21	3.11	5.0						1	0.802** [0.683–0.880]	0.761** [0.629–0.85]
7. Sexting (pics)		3.27	2.47	3.0							1	0.773** [0.726–0.813]
8. Sexting (fantasies)		3.37	2.54	3.0								1

Note.

Correlations show Spearman Correlation Coefficients.

** indicates a p < 0.01.

The scale measuring frequencies of digisexualities ranges from a minimum of 1 to a maximum of 11. 1: never, 2: once in the past year, 3: 2–6 times in the past year, 4: 7–11 times in the past year, 5: monthly, 6: 2–3 times a month, 7: weekly, 8: 2–3 times a week, 9: 4–5 times a week, 10: 6–7 times a week, 11: more than 7 times a week.

Sample size (n) varies, as frequency was calculated only for participants who confirmed having ever used the devices or engaged in the specified behaviors; weighted n: n₁ = 3,564, n₂ = 2,345, n₃ = 189, n₄ = 1,023, n₅ = 1,024, n₆ = 138, n₇ = 677, n₈ = 1,073.

Reported data is weighted proportionally according to age, gender and regional distribution in Germany.

Confidence intervals (CI) are reported at the 99% confidence level.

Table 4.

Correlation between frequency of digisexualities and compulsive sexual behavior by sex

	Male									Female
	M (SD)	1	2	3	4	5	6	7	8	M (SD)	1	2	3	4	5	6	7	8
1. CSBD	32.72 (12.15)	1	0.239**	0.423**	0.278**	0.550**	0.536**	0.397**	0.395**	26.18 (9.54)	1	0.235**	0.438**	0.185**	0.374**	0.202	0.338**	0.297**
2. Pornography	5.34 (2.91)		1	0.439**	0.325**	0.195**	0.572**	0.292**	0.346**	2.87 (2.15)		1	0.686**	0.332**	0.279**	0.748**	0.363**	0.411**
3. VR-pornography	4.85 (2.97)			1	0.637**	0.616**	0.822**	0.698**	0.512**	3.98 (2.19)			1	0.646**	0.526**	0.893**	0.824**	0.835**
4. Dating App	4.35 (3.18)				1	0.553**	0.719**	0.549**	0.541**	2.72 (2.48)				1	0.541**	0.600**	0.491**	0.477**
5. Casual Sex	2.83 (2.59)					1	0.674**	0.640**	0.553**	1.68 (1.69)					1	0.491*	0.542**	0.480**
6. Sex doll	5.27 (3.18)						1	0.824**	0.744**	4.95 (2.83)						1	0.710**	0.877**
7. Sexting (pics)	3.71 (2.63)							1	0.749**	2.7 (2.13)							1	0.780**
8. Sexting (fantasies)	3.8 (2.67)								1	2.78 (2.23)								1

Note.

Correlations show Spearman Correlation Coefficients.

** indicates a p < 0.01.

The scale measuring frequencies of digisexualities ranges from a minimum of 1 to a maximum of 11. 1: never, 2: once in the past year, 3: 2–6 times in the past year, 4: 7–11 times in the past year, 5: monthly, 6: 2–3 times a month, 7: weekly, 8: 2–3 times a week, 9: 4–5 times a week, 10: 6–7 times a week, 11: more than 7 times a week.

Sample size (n) varies, as frequency was calculated only for participants who confirmed having ever used the devices or engaged in the specified behaviors.

Reported data is weighted proportionally according to age, gender and regional distribution in Germany.

Discussion

The objective of this study was to ascertain the prevalence of first- and second-wave digisexualities in Germany and to investigate any potential associations with CSBD. This study was not intended to pathologize sexual activities in digital/artificial contexts, but rather to understand if these technologies and behaviors are relevant for people struggling with their out-of-control sexual behavior.

As the three main results we observed first that conventional pornography use remains the most prevalent digisexual behavior, as reported by 66.1% of participants, with greater gender differences than in previous studies (Herbenick et al., 2020). Secondly, the strongest correlation regarding the CSBD-19 score was observed for casual sexual encounters mediated through dating apps. Thirdly, there is a notable interconnectivity between emerging digisexual behaviors, particularly the use of VR-pornography and highly realistic sex dolls/robots, with a strong correlation observed between these activities.

Generational and sex differences in digisexuality

The findings in our study further indicate that men consistently report higher prevalence rates for digisexual behavior than women, with particularly pronounced differences in the consumption of VR-pornography, conventional pornography and the use of highly realistic sex dolls/robots. These differences may reflect gender norms regarding sexual behaviors or potential biases in how these topics are discussed (Petersen & Hyde, 2010). Although these differences between men and women are robust, it is important to consider that the pornography industry primarily targets men in its offerings (Grubbs, Wright, Braden, Wilt, & Kraus, 2019). A similar pattern can be observed in online shops for sex dolls, which also appear to be predominantly marketed toward male consumers.

Furthermore, younger generations generally reported higher prevalence rates of digisexual behaviors compared to older generations. This trend may be due to younger people's greater openness to technology and familiarity with navigating technological devices, as many have grown up with these. This finding aligns with previous research suggesting that younger individuals are often early adopters of new technologies (Cormier & O’Sullivan, 2021) and seem to be more interested in sexual actions (Beutel, Stöbel-Richter, & Brähler, 2008).

The widespread use of sexting among younger generations exemplifies the shifts in sexual communication in the digital age. The differentiation between sexting images and sexting fantasies provides valuable insights into the varying levels of digital intimacy. In a previous study on online sexual behavior in Germany (Döring & Mohseni, 2018), 41% of adults engaged in at least one type of sexting, which aligns with our findings of 19% respectively 30.1%. However, while our study found a clear trend of increasing sexting engagement among younger participants, other studies (Courtice & Shaughnessy, 2017; Döring & Mohseni, 2018) report higher sexting rates among adults compared to adolescents. Notably, our study did not include adolescents, which may account for this difference.

Potential risks and benefits of digisexual behaviors

Although the debate about potential risks for young people arising from their use of 2D online pornography is still ongoing (Malamuth, 2018; Owens, Behun, Manning, & Reid, 2012; Peter & Valkenburg, 2016), many experts fear that usage could lead to risks such as greater permissive sexual attitudes, stronger gender-stereotypical beliefs, earlier experiences of sexual intercourse, more casual sex and more sexual aggression (for a review, see Peter & Valkenburg, 2016). The high prevalence rates for other digisexualities (50.2% reported using online dating apps; 50.1% reported sexting fantasies) in Gen Z could reinforce such fears. Nevertheless, it is also noteworthy that other studies have indicated that pornography can have beneficial effects, such as facilitating sexual education (see for example, Brown et al., 2006) or enhancing sexual agency (Klein, Šević, Kohut, & Štulhofer, 2022). A study measuring both positive and negative outcomes of online sexual behavior among adults in Germany (Döring & Mohseni, 2018) found that negative consequences were less pronounced than positive ones. The authors also noted that greater engagement in online sexual activities surely increases exposure to potential personal influences. However, as new behaviors emerge, it might be essential to observe and describe them systematically before engaging in polarized debates about their impact.

It is similarly important to identify potential risks and opportunities for clinical practice in this context. In light of the evidence indicating the pervasive prevalence of diverse forms of digisexualities across generations, it is important for clinicians and the general population to consider these behaviors when it comes to assessing sexual behavior.

While our data indicate that older generations barely use such objects, it is noteworthy that sex dolls are frequently discussed as potential aids for the elderly or individuals with disabilities (Fosch-Villaronga & Poulsen, 2020). Practical factors such as the size and weight of the dolls may make them less accessible to people with physical limitations. Furthermore, these high-cost technologies require extensive equipment, which may make them less likely to be used.

Digisexual behaviors and their association with CSBD

All digisexual behaviors examined in this study correlated on a moderate to strong level with CSBD-19 scores. The strongest correlation was observed with the use of dating apps for casual sex, followed by the use of highly realistic sex dolls/robots. Nonetheless, while these behaviors are still less common in the general population than others (28.7% for dating app use; 3.9% for sex doll/robot use), they are more prevalent among younger generations. Furthermore, while the utilization of sex dolls/robots is rare, individuals who engage in it tend to do so several times a month and more often than engaging with other digisexualities. These findings might have significant implications for understanding and dealing with CSBD because sex dolls/robots may serve as a harm-reduction tool, offering a controlled outlet for individuals struggling with compulsive sexual behaviors without involving real-life partners. However, their physical realism and frequent use could also reinforce compulsivity by creating a self-sustaining cycle of solitary sexual engagement, potentially diminishing the motivation for interpersonal intimacy.

The strong links between various digisexual behaviors and compulsive tendencies reveal potential targets for early intervention and prevention strategies. CSBD is characterized by out-of-control sexual behavior that persists despite negative consequences in personal, social or professional spheres (Bőthe et al., 2020; Fuss, Briken, et al., 2019; Grubbs et al., 2023). Among the digisexual behaviors examined, dating app use for casual sex is particularly notable, as it is the only one that inherently involves direct interpersonal contact. Interestingly, recent cross-cultural research has not only identified a consistent association between CSBD and the frequency of dating app use across diverse cultural contexts but identified it as the most consistent factor related to CSBD (Lewczuk et al., 2024).

This relationship also echoes findings in studies on chemsex—the use of psychoactive substances to enhance or prolong sexual encounters—which has been linked to both dating app use and CSBD (Jennings, Gleason, Nieblas, Borgogna, & Kraus, 2025; Malandain & Thibaut, 2023, 2025). Chemsex is further associated with serious health risks, including depression, anxiety, and increased risk of HIV transmission (Amundsen, Muller, Reierth, Skogen, & Berg, 2024). Taken together, these findings suggest a need for future research to explore the potential intersection of CSBD, dating app-facilitated casual sex, and chemsex, and to determine whether these behaviors reinforce each other or represent overlapping expressions of compulsive sexual behavior.

The data suggest that problematic pornography use, a core symptom of CSBD (Wizła & Lewczuk, 2024), may be accompanied by newer digisexual behaviors, as these technologies become more accessible and affordable. Our findings are also echoing results by Grubbs et al. (2024), who found diverse behavioral expressions among individuals reporting problematic sexual behavior. While our study identified age and gender as relevant factors shaping the expression of different behaviors, Grubbs et al. (2024) further emphasized the role of religiosity as an additional influence and highlighted a more nuanced picture regarding sexual orientation. This highlights the need to understand how emerging technologies may exacerbate or facilitate problematic behaviors in individuals with CSBD and how they will shape the future phenomenology of CSBD. Furthermore, the exceptionally strong correlation between digisexualities suggests that these behaviors often co-occur. Clinicians and scholars should be aware of these behavioral clusters when treating maladaptive sexual habits.

Limitations

The main limitation of this work is that, although we tried to achieve a representative sample of German society through a weighting factor, an online panel recruitment cannot be considered entirely representative, as people who are less likely to use the internet may be underrepresented, and recruitment does not follow a randomized approach.

Although we selected the digisexualities based on existing literature, there are many other ways of living sexuality through technology (e.g., teledildonics, various social media, AI-generated pornography, etc.). This means that we have not covered all presentations of digisexualities in our survey. Also, we assessed pornography consumption without differentiating between digital and non-digital formats such as magazines. However, as contemporary pornography use is overwhelmingly digital (Döring, Daneback, Shaughnessy, Grov, & Byers, 2017; Miller, Raggatt, & McBain, 2020; Price et al., 2016), we assume that our findings primarily reflect online consumption, aligning with our conceptualization of first-wave digisexualities.

Conclusion and future directions

With this study we reveal that the use of digisexualities is prevalent among the general population and associated with compulsive sexual behavior particularly among younger generations. As research on CSBD and its symptoms advances, it is crucial to incorporate a broader spectrum of digisexual behaviors, including online dating, sexting and the use of highly realistic sex dolls/robots. With the increasing presence of generative AI and customizable erotic technologies, one might ask whether we are witnessing the beginning of a new, third wave of digisexuality—one that could once again shift how (compulsive) sexual behavior is experienced and expressed (Döring, Le, Vowels, Vowels, & Marcantonio, 2025; Lapointe, Dubé, Rukhlyadyev, Kessai, & Lafortune, 2025). These rapid technological advances highlight the need for diagnostic frameworks like CSBD to stay attuned to changing sexual norms. Yet, this need for timely conceptual updates is often at odds with the methodologically demanding process of developing and validating clinical instruments, creating a key challenge for researchers and practitioners alike.

A key next step could be the development of robust measurement tools to assess various forms of problematic digisexuality—akin to the Problematic Pornography Use Scale (Bőthe et al., 2018), but tailored for other digital sexual behaviors. However, given the methodological scope of this study, we did not explore the potential positive effects of these technologies. Future research should ensure that these aspects are not overlooked, fostering a comprehensive and balanced understanding of digisexuality.