Online sexual activities (OSAs) encompass engagement in various activities on the internet that cater to sexual gratification. These activities include searching for sex-related information, viewing pornography, seeking potential sexual partners, engaging in sexual chatting, and purchasing sexual services online [1]. In the Chinese social context, official propaganda considers pornography harmful; and the dissemination of pornography is considered a crime in China. However, the internet offers accessibility, affordability, and anonymity, providing individuals with a secure space for sexual experiences [2]. Furthermore, pornography consumption can serve as a substitute for casual sex [3]. As a result, pornography use is still common in China. A significant percentage (approximately 79.58%) of Chinese college students have engaged in at least one type of OSAs within the past year, with the highest proportion being those who have ever viewed pornography online [4]. Similarly, nearly 89% of Chinese adults have engaged in OSAs [5]. These findings aligned with studies conducted among Western population [6,7]. As mentioned, the pornography industry is forbidden in China. The Chinese culture tends to be conservative regarding sexual issues [8]. Thus, the potential adverse effects of extensive OSA within a conservative sexual culture is of concern.

In most cases, engaging in OSAs can have positive effects on individuals such as improving their sex life, providing stress relief, alleviating boredom, and enhancing sexual knowledge [9]. However, OSAs can become excessive and uncontrollable for a small group of individuals, affecting their physical and mental health and possibly even causing sexual violence and crime [10]. The problematic OSAs is associated with negative consequences and functional impairment [11] characterized by repetitively or excessively engaging in OSAs; persistent desires for OSAs or unsuccessful efforts to stop, reduce, or control online sexual behaviors [12]. Studies have shown that the frequency and duration of pornography use determined the quantity of pornography use [13], and the severity of pornography use leads to problematic pornography use (PPU) [14]. Moreover, PPU is more closely related to frequency of use than duration of pornography use [15,16]. Excessive pornography use was deemed an essential prerequisite for the emergence of PPU [17]. Hence, we were inclined to assess the frequency of pornography use. Although OSAs cover a broader range than pornography use, the definition of problematic OSAs suggests that high frequency of engaging in OSAs is a necessary and obvious symptom of problematic OSAs. Nevertheless, the threshold for frequently engaging in OSA that classifies it as problematic use has not been explored.

Therefore, what is considered as excessive (problematic) use of OSAs? The exploration of this question will facilitate the early identification and diagnosis of PPU and problematic use of OSAs. Recent studies have found a strong positive correlation between OSAs and problematic use, frequency of OSAs are strong predictors of problematic use in the community sample [18]. To prevent problematic OSAs, attention ought to be given to individuals’ engagement in cybersex in addition to their frequency of pornography use [19]. Research has found an association between craving pornography and the frequency of engaging in cybersex, as watching pornography has been identified as a gateway to cybersex and vice versa [20]. An established cutoff value for the frequency of OSAs could help individuals to consciously assess whether their engagement is problematic and seeking help is necessary. Without the cutoff, continuous excessive involvement in OSAs can cause physical and psychological harm. Establishing the threshold for excessive engagement in OSAs (frequency) can prompt initial screening for OSAs’ problematic use and cause timely preventive measures to be implemented.

There are more than 20 scales for assessing PPU and problematic OSAs from various perspectives, such as the Problematic Pornography Use Scale (PPUS) [16], Problematic Pornography Consumption Scale (PPCS) [15], Brief Pornography Screen (BPS) [21], and Cyber Pornography Use Inventory-9 [22]. However, most of these scales assess the consumption of online or offline pornography, which is categorized as solitaryarousal activities. These scales do not measure other OSAs such as engaging in sexual chats, seeking or participating in sex webcams, or pursuing sexual partners, which are partneredarousal OSAs [23,24]. Meanwhile, both partnered- and solitaryarousal activities can manifest problematic patterns [25,26]. In addition, the short internet addiction test adapted to OSAs (s-IAT-sex) assesses problematic OSAs. It measures the loss of control in browsing internet sex sites but lacks a threshold for excessive OSAs. Therefore, research must establish a threshold for flagging OSAs with problematic use.

Research has demonstrated that individuals seeking help for pornography-related issues exhibit distinct characteristics that deviate from the norm. Specifically, these individuals demonstrate heightened attention towards sexual stimuli, an increased propensity for sexual arousal and pornography-related cravings, and an overall preoccupation with pornography that significantly impacts their thoughts, emotions, and behaviors [27]. This suggests that the subclinical sample (those who have lost control and sought help regarding pornography use) displays a greater severity of pornography use compared to the community sample. Moreover, according to the two-stage model of addiction, the initial stage is characterized by compulsive reliance and loss of control over a substance or behavior, while the subsequent stage is marked by withdrawal symptoms and an inability to terminate the addiction despite severe negative consequences [28]. To observe the progression of the two stages, we selected two samples, namely community and subclinical samples, to exhibit varying levels of engagement in OSAs.

Based the following considerations 1) the widespread use of OSAs and individuals’ excessive engagement in them, which can result in problematic behavior; 2) the lack of a threshold to measure what extent of excessive OSA is classified as problematic pornography consumption; and 3) the frequency of engaging in OSAs varies between the community and subclinical samples, we aimed to determine the cutoff for the excessive engagement in OSAs by estimating the frequency of engaging in OSAs in community and subclinical samples. Compared to women, men generally have a higher frequency of OSAs [5,20] and are more likely to encounter PPU [7,29]. Thus, adult male participants were recruited for this study. Furthermore, the questionnaire for OSAs classifies sub-types of online sexual behavior according to the definition of OSA [26,30] and is widely employed in Chinese research [5,8,27,29,31-33]. In this study, the frequency of OSAs was measured by the frequency of viewing sexually-explicit materials (SEM), seeking sexual partners, engaging in cybersex, and participating in flirting and sexualrelationship maintenance, as suggested by Shaughnessy et al. [23] and Zheng and Zheng [29]. We used these four sub-types of OSA as profile indicators to identify subgroups within the two samples. PPU can manifest as frequently engaging in various forms of OSA, intense cravings for pornography, compulsive sexual behaviors [12], impaired control [34], and negative outcomes such as poor mental health [35]. Consequently, the quality and characteristics of PPU considered in this study include sexual compulsion [36], loss of control over pornography [21], craving for pornography [37], and overall health status [38]. These variables were investigated to ascertain the disparity between two distinct male groups, determined by the cutoff of the OSA questionnaire.

In this study, community and subclinical adult males were recruited, and their data were analyzed separately. Cut-point analyses showed that an OSAs score of 45 points was the cutoff for excessive OSAs in the community sample, but 60 points was the cutoff in the subclinical sample. The significance of the OSA cutoff varies in regard to screening community and subclinical samples for PPU. In the community group, the OSAs cutoff categorized individuals into two distinct groups: the LO-NR group and the HO-IC group. Similarly, the subclinical sample was segregated into two groups: HO-PPU and the HV-IC individuals. In a broader context, excessive OSAs can be a valuable screening tool for detecting out-of-control in OSAs within the Chinese populace. Furthermore, the implementation of more stringent criteria for excessive OSAs within subclinical samples can aid in distinguishing the PPU from the purely impaired control in viewing SEM. By synthesizing the results obtained from the two samples, it is evident that two distinct subcategories of loss of control (HO-IC and HV-IC) exist.

There are two types of OSAs: solitary- and partnered-arousal activities. Partnered-arousal activities encompass activities such as seeking sexual partners, engaging in cybersex, flirting and maintaining sexual relationships. Solitary-arousal activities, such as viewing SEM, do not involve interpersonal interaction. The HO-IC (community sample) and HO-PPU (subclinical sample) groups scored higher on all four subtypes of OSAs, including both solitary- and partnered-arousal activities. In contrast, the HV-IC group in the subclinical sample engaged more frequently in viewing SEM than other collaborative activities. These results suggest that those who frequently view SEM may be more prone to overuse of pornography and have negative consequences. People browse pornography more due to its ease and convenience compared to other forms of OSAs. The intentional use of SEM is most often coupled with solo-masturbation [41]. Among men who masturbated frequently, 70% used pornography at least once a week [42]. In general, higher pornography consumption was primarily associated with increased masturbation frequency. Another notable finding was that both women and men with frequent pornography use were more likely to report high masturbation and sexual satisfaction than those with low masturbation [43]. In sum, viewing SEM facilitates arousal and solo-masturbation and can fulfill individuals’ sexual desires and enhance their overall sexual contentment. Consequently, this can intensify users’ reliance on viewing SEM, ultimately resulting in a loss of control or PPU. Additionally, our finding supports the notion that browsing pornography is the most prevalent OSA [4,26,30].

The threshold of excessive OSAs vary among the public population and those who exhibit out-of-control pornography use. There are individuals who need to be watchful of the frequency of their engagement in OSAs, because they may experience impaired control and compulsive use (HO-IC group, OSA≥45, n=69). In the subclinical sample, a notable percentage of individuals (HV-IC group, OSA<60, n=509) did not meet the criteria for salience, tolerance, mood modification, and withdrawal of PPU symptoms but had more serious negative consequences (greater sexual compulsivity and poorer health) from a loss of control in viewing SEM. The addiction theory proposed by Griffiths [39] posits that addiction comprises six fundamental components: salience, emotional regulation, conflict, tolerance, repetition, and withdrawal symptoms. Many researchers have highlighted the importance of tolerance and withdrawal symptoms in distinguishing addictive behaviors from those that are out of control [44]. Additionally, a recent study found that withdrawal and mood modification were core features of PPU, and as a coping mechanism, mood modification may be at a greater risk of developing into problematic use [45]. Therefore, the group of users with a lower tendency regarding withdrawal and mood modification was also labeled as having impaired control [27]. Researchers have suggested that compulsive use and unsuccessful efforts to reduce pornography use despite negative consequences is the first step of problematic PPU [46]. At this stage, individuals engage in online pornographic information or other OSAs to gain satisfaction after perceiving an internal need and external cue of relevance [45,47]. The PPU stage is characterized by difficulty to withdraw, repeated failures, and the inability to stop even when experiencing strong negative consequences. While impaired control can predict PPU over a period of six months and play an essential role in screening PPU, it is insufficient to indicate PPU itself [34]. Therefore, the males in the HV-IC group were also considered as having impaired control, although they scored higher than the PPCS cutoff.

When comparing the HO-IC and the HV-IC groups, the former had a higher frequency of all four sub-types of OSAs than the latter. However, regarding the indicators of loss control and PPU, the latter is more serious than the former. Concerning the qualitative indicators for PPU (six dimensions of PPU, SCS, PCQ, and GHQ), there is a difference between the two groups. This distinction lies in the conflict dimension, as the HV-IC scored higher than the cutoff of conflict, while the HO-IC group scored lower than the cutoff. The results indicate that the HV-IC group exhibited heightened conflict characteristics. Conflict refers to the adverse consequences derived from impaired control, including interpersonal conflict and efficiency reduction due to porn use—knowing the activity is causing problems but feeling unable to cut down or cease its use [15]. Thus, their conflict may be the motivation to seek help. The HV-IC group was identified in the subclinical sample. The sample was selected from those who sought help on the website for their pornography use. This finding is in line with extant studies: compared to the quantity of pornography use, negative symptoms such as out-of-control use or worsening intimate relationships, could be better predictors of patients’ willingness to seek treatment [48].

Hence, a significant finding from this study suggests that the loss of control encompasses not only the inability to regulate pornography consumption but also the difficulty in managing other OSAs. Although some studies have revealed that high frequency of OSAs use was not necessarily a problem [17] and at the edge of the symptom network in impaired control and PPU groups [49], the role of OSA frequency in screening for control disorders and PPU is not ruled out. The frequency of engaging in OSAs was important for identifying male adolescents with self-perceived addiction to pornography but actually without behavioral dysregulation [27]. In the community sample, a score of 45 indicated excessive OSAs, while in the subclinical sample, a score of 60 was the cutoff. Excessive OSAs can be a screening tool to identify impaired control in all four dimensions of OSAs in the general public. However, stricter criteria for identifying excessive OSAs (≥60) within subclinical samples are necessary to accurately distinguish between PPU and purely impaired control in viewing SEM. It is advisable to look at the frequency of OSAs as an auxiliary index for the screening and diagnosis of impaired control and problematic use [32].

This study has some limitations. Firstly, all measures were self-report questionnaires with the assumption that participants were aware of and willing to truthfully report their OSAs. The reliability of the results relied on the candor of the participants. Secondly, our data collection relied on network-based self-report measures, which could have introduced certain biases, often overrepresenting younger, urban, and better educated individuals. Moreover, the participants who were interested in completing our questionnaire on pornography use might have been personally troubled by its use, although they did not seek help. Thirdly, because the sample consisted of only adult males, including a small number of non-heterosexual individuals, it was not possible to examine whether the cutoff of the OSAs differed across gender and sexual orientations. To advance future research, the extent of excessive OSAs could be further corroborated by self-reports and clinical diagnoses, thus broadening its application.