The Effect of Time Spent on Online Gaming on Problematic Game Use in Male: Moderating Effects of Loneliness, Living Alone, and Household Size

Article information

Psychiatry Investig. 2024;21(2):181-190

Publication date (electronic) : 2024 February 22

doi : https://doi.org/10.30773/pi.2023.0027

Kyeongwoo Park¹, Hyein Chang^,1, Jin Pyo Hong^2,3, Myung Hyun Kim³, Sohee Park³, Jin Young Jung³, Dahae Kim³, Bong-Jin Hahm^4,5, Ji Hyun An^,2,3

¹Department of Psychology, Sungkyunkwan University, Seoul, Republic of Korea

²Department of Psychiatry, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea

³Department of Psychiatry, Samsung Medical Center, Seoul, Republic of Korea

⁴Department of Psychiatry and Behavioral Science, Seoul National University College of Medicine, Seoul, Republic of Korea

⁵Department of Neuropsychiatry, Seoul National University Hospital, Seoul, Republic of Korea

Correspondence: Hyein Chang, PhD Department of Psychology, Sungkyunkwan University, 25-2 Seonggyungwan-ro, Jongno-gu, Seoul 03063, Republic of Korea Tel: +82-2-760-0490, E-mail: hichang@skku.edu

Correspondence: Ji Hyun An, MD Department of Psychiatry, Samsung Medical Center, 81 Irwon-ro, Gangnam-gu, Seoul 06351, Republic of Korea Tel: +82-2-3410-3588, Fax: +82-2-3410-0050, E-mail: sereinjh@gmail.com

Received 2023 February 2; Revised 2023 October 25; Accepted 2023 December 1.

Abstract

INTRODUCTION

The popularization of smartphones and the growth of the mobile gaming have increased the variety and availability of video games [1], leading to a surge in global gaming [2], particularly in countries like Korea with high internet penetration [3]. In 2021, the game usage rate of Koreans was 71.3%, and among game users, the mobile game usage rate (90.9%) highly exceeded the PC game (57.6%) or console game (21.0%) usage rate [4]. As the number of game players increase, and the media and content of games become more diversified, game use has surged not only as a form of amusement but also as an alternative way to fulfill various psychological needs, such as social engagement, accomplishment, emotional avoidance, and stress release [5]. This may be more evident in online and multiplayer games with social components [6]. The number of online gamers has exceeded that of offline games, and this gap has widened during the coronavirus disease-2019 (COVID-19) pandemic, when offline activities were highly restricted [7,8].

However, overuse of games may have negative effects on users, in the form of compromised physical/mental health, academic/professional performance, and interpersonal relationships as a result of failure to control their game use and their tendency to prioritize gaming over other duties/activities [9,10]. Such tendencies are referred to as problematic gaming or problematic game use (PGU) [6,11], and if they persist over an extended period (e.g., more than 12 months), they can be diagnosed as a gaming disorder in the International Classification of Diseases 11th Revision [12]. Although the prevalence of gaming disorder varies considerably across regions, it is known to be particularly high in East Asia, including Korea [13]. In addition, although controversial, some studies have argued that increased PGU severity may cause neurological changes similar to those in substance use disorders [10,14].

From this perspective, chronic PGU is often considered a subtype of behavioral addiction, underscoring the importance of early screening and intervention to minimize psychological and socioeconomic damage. To achieve this goal, it is essential to identify risk factors and reveal how they may conjointly have varying effects on individuals with PGU vulnerability. Male gamers can be considered an important target population for this endeavor, given that PGU is generally more prevalent in male than in female. According to a recent meta-analysis, male are more than twice as likely to experience PGU as female [15]. In addition, several studies have shown that male and female also differ in major risk factors of PGU and in the relationship between the factors [16,17]. These findings suggest that an approach to investigate male and female separately may be useful in studying the underlying mechanisms of PGU.

Excessive gaming is a major risk factor for PGU. However, it is important to note that quantitative indicators (e.g., time and frequency of engagement) of digital media use may only partially correlate with problematic use [18,19], and this may also be the case for PGU [20,21]. Latent profile analysis studies conducted in Europe and Asia have consistently identified two subgroups within gamers, with differences in PGU severity but no significant differences in gaming time [22,23]; that is, those who have socio-psychological issues and those who do not may be differentiated, even among individuals who commit a great deal of time to gaming. Heavy gamers who are not maladjusted and are actively engaged in social activities, such as gaming YouTubers and professional gamers, exemplify that excessive gaming does not always lead to PGU, which is also indirectly supported by the ineffectiveness of gaming control legislations in many Asian countries, which focus on restricting usage time [24,25]. This evidence suggests that examining gaming time alone may not be sufficient to assess PGU risk, and that it may be necessary to investigate the conditions under which excessive game use leads to PGU.

Loneliness is a subjective emotional experience of perceived isolation that is independent of the actual lack of social relationships [26], and may function as a “railroad switch” connecting excessive gaming with PGU, especially in male. Studies have found that loneliness shows a weak or no significant relationship with game usage but nevertheless demonstrates a consistent positive correlation with PGU [27-29]. In addition, according to Stockdale and Coyne’s research comparing problematic gamers and low-risk gamers, the two groups did not vary in perceived levels of social support networks (e.g., the number of regular interactions), but the PGU group experienced more feelings of social isolation such as loneliness [30]. While evidence is still scarce, some researchers have suspected that loneliness may reduce responsiveness to positive social interactions (which is a protective factor for PGU), resulting in mistrust of others and social withdrawal [31]. In a similar context, gamers who suffer severe loneliness tend to have relatively low social competence and/or entertainment sought motivation (e.g., playing games to interact with friends or having fun) and high escape and fantasy motivation (e.g., using games to escape the real world or break away from their usual identity) [17,28].

Interestingly, both the main and indirect effects (through game use motivations) of loneliness on PGU were found to be significant only in male [17]. This differential effect of loneliness by gender is consistent with previous studies on other addictive behaviors (e.g., pathological gambling and problematic social media use) [32,33]. Considering that the game market tends to be more tuned towards men’s preferences [34], and self-efficacy about the game is generally higher in male [35], the above result may reflect that for female, playing games has a relatively lower incentive than other behavioral options and is not expected to be a useful tool for addressing their loneliness.

Additionally, we explored the possibility that the potential moderating effect of loneliness on the association between gaming time and PGU differed depending on the presence or absence of a housemate. As stated above, because loneliness is an emotion based on subjective experience, individuals may experience severe loneliness even if they live with others. Indeed, it has been suggested that loneliness and living arrangements should be considered as close but independent factors that may determine individual outcomes [36,37], and previous studies have shown that their correlation is inconsistent and influenced by factors such as age and relationship quality with housemates [36-38]. Nevertheless, the presence of housemates may serve as a minimal safeguard against PGU resulting from loneliness and excessive game use. In addition to monitoring maladaptive digital media use [39], housemates may prevent loneliness from lasting too long by decreasing the time spent completely alone [38]. The presence of a housemate may also mitigate the risk of other mental health issues such as depression and stress, which can increase vulnerability to PGU [40,41].

However, even if the risk of PGU is lower in households with housemates than in solitary ones, it is uncertain whether this effect changes as a function of household size (i.e., the number of housemates). Several studies have found that persons living with more than three or four housemates are more likely to feel lonely [42], perceive external information more negatively [43], and be more susceptible to environmental stress than those living with fewer housemates [44]. Living with too many housemates can lead to decreased social support, privacy invasion, limited alone time, and more negative interactions [44,45]. Therefore, we believe that the possible role of household size in PGU should be examined from a more exploratory perspective. Nevertheless, given that gaming is mostly done at home, regardless of age [4], we anticipated that the greater the number of housemates who can continually monitor and moderate gaming behavior, the less likely lonely game overusers are to experience PGU.

Based on these findings, we hypothesized the following:

H1. The relationship between gaming time and PGU is moderated by loneliness. In other words, the positive effect of gaming time on PGU is greater when loneliness is high.

H2. The interaction effect between gaming time and loneliness on PGU differs depending on whether participants live with housemate(s). In other words, the moderating effect of loneliness on the relationship between gamers’ gaming time and PGU is greater for individuals living alone.

H3. Among participants living with housemate(s), household size moderates the interaction between gaming time and loneliness. Specifically, the moderating effect of loneliness on the relationship between gaming time and PGU is greater for gamers living with fewer housemates.

METHOD

RESULTS

Sample characteristics

Demographic characteristics are summarized in Table 1. The age range of the participants was 18 to 72 years (M_age= 34.62, SD_age=10.84), and the majority (89.4%) were under the age of 50. In terms of employment, 504 (67.8%) participants were full-time workers, 100 (13.5%) were temporary workers, and 139 (18.7%) were unemployed or were students. Regarding the level of education, 11 (1.5%) graduated from middle school or lower, 280 (37.7%) graduated from high school, 451 (60.7%) graduated from college or had higher degrees, and 1 (0.1%) did not respond. A total of 194 (26.1%) participants were classified as high-risk for PGU based on the cutoff points, and 148 (19.9%) reported living alone. Of those living with housemates, 137 (18.4% of the total) had one housemate, 235 (31.6% of the total) had two, 190 (25.6% of the total) had three, and 33 (4.4% of the total) had four or more.

Table 1.

Demographic information (N=743)

Descriptive statistics and bivariate correlations

Descriptive statistics and bivariate correlations are presented in Table 2. The average daily gaming time of the participants was 1.98 hours (approximately 1 hour and 59 min, SD=1.50). The range was 0.29 to 9.43 hours, but 65.9% of participants reported 2 hours or less per day and 85.1% reported 3 hours or less. The mean scores for PGU and loneliness were 36.96 (SD=10.09) and 0.81 (SD=1.27), respectively. In addition, the mean household size was 2.77 (SD=1.19). PGU was negatively correlated with age (r=-0.155, p<0.001) and positively correlated with loneliness (r=0.211, p<0.001) and gaming time (r=0.241, p<0.001). This indicates that younger individuals, those experiencing higher levels of loneliness, or those who spent more time playing games were more likely to experience PGU. The correlation between age and gaming time was also significant (r=-0.207, p<0.001). In addition, loneliness was positively correlated with age (r=0.162, p<0.001) and living alone (r=0.158, p<0.001), and negatively correlated with household size (r=-0.119, p=0.001), indicating that younger participants, those who lived alone, or those with fewer housemates were more likely to feel lonely.

Table 2.

Descriptive statistics and bivariate correlations (N=743)

Main and interaction effects of gaming time and loneliness on PGU

PROCESS macro Model 1 was used to test the moderating effect of loneliness on the relationship between gaming time and PGU. As presented in Table 3 and Figure 1, the main effects of gaming time (B=1.268, 95% confidence interval [CI]=[0.797–1.739]) and loneliness (B=1.735, 95% CI=[1.184–2.286]) and their interaction effects (B=0.325, 95% CI=[0.031–0.619]) were significant. The model explained 12.4% of the total PGU variance (F [4,738]=26.184, p<0.001).

Table 3.

Main, interaction, and conditional effects of gaming time and loneliness on PGU (N=743)

Figure 1.

Moderation effect of loneliness on the relationship between gaming time and PGU. The model included age as a covariate. *p<0.05; ***p<0.001.

To probe the two-way interaction, the conditional effect of gaming time on PGU was analyzed by dividing the groups based on the mean and ±1 SD points of loneliness. The positive effect of gaming time on PGU was significant, regardless of the level of loneliness, but the effect for the condition with high loneliness (B=1.679, 95% CI=[1.119–2.240]) was stronger than the average (B=1.268, 95% CI=[0.797–1.739]) or low (B=1.005, 95% CI=[0.451–1.559]) loneliness conditions.

Main and interaction effects of gaming time, loneliness, and living alone on PGU

PROCESS macro Model 3 was used to examine the three-way interaction effects of game time, loneliness, and living alone on PGU. As shown in Table 4, the main effects of gaming time (B=0.805, 95% CI=[0.297–1.312]) and loneliness (B=2.199, 95% CI=[1.556–2.842]) were significant. In terms of interaction effects, as shown in Figure 2, the two-way interaction effects of gaming time×living alone (B=2.718, 95% CI=[1.497–3.940]) and loneliness×living alone (B=-1.359, 95% CI=[-2.561–-0.157]), and the three-way interaction effect of gaming time×loneliness×living alone (B=0.757, 95% CI=[0.039–1.475]) on PGU were both statistically significant. The model explained 16.1% of the total PGU variance (F [8,734]=17.567, p<0.001).

Table 4.

Main, interaction, and conditional effects of gaming time, loneliness, and living alone on PGU (N=743)

Figure 2.

Three-way interaction effect of gaming time, loneliness, and living arrangement on PGU. The model included age as a covariate. *p<0.05; **p<0.01.

To probe the three-way interaction, the conditional effect of the interaction term of gaming time×loneliness on PGU was analyzed by dividing groups based on living arrangements (living alone vs. living together). The moderating effect of loneliness on the relationship between gaming time and PGU was significant only for participants living alone (B=0.900, F [1,734]=7.617, p=0.006) and not for those living with housemate(s) (B=0.143, F [1,734]=0.756, p=0.386). Furthermore, as shown in Figure 3, in the living alone group, gaming time positively predicted PGU at all levels of loneliness, but the effect for the condition of high loneliness (B=4.663, 95% CI=[3.402–5.923]) was larger than the average (B=3.523, 95% CI=[2.408–4.637]) or low (B=2.796, 95% CI=[1.489–4.103]) loneliness conditions.

Figure 3.

Conditional effects of gaming time on PGU depending on the level of loneliness based on living arrangements. The model included age as a covariate. The units for PGU are consistent with the original scale scores. Both the gaming time and loneliness scores have been standardized. *p<0.05; **p<0.01; ***p<0.001. SD, standard deviation.

Main and interaction effects of gaming time, loneliness, and household size on PGU

Using data from participants living with housemate(s) (n=595), we tested the three-way interaction effect once again by entering the household size variable into the model in place of the variable of living alone. The results are presented in Table 5. The main effects of gaming time (B=0.826, 95% CI=[0.294–1.355]) and loneliness (B=2.173, 95% CI=[1.519–2.827]) were significant, but none of the interaction effects, for either the two- or three-way interaction, were significant. The model explained 11.9% of the total PGU variance (F [8,586]=9.923, p<0.001).

Table 5.

Main and interaction effects of gaming time, loneliness, and household size on PGU (N=595)

DISCUSSION

This study aimed to investigate the underlying mechanism of male gamers’ PGU in the context of increased social concerns regarding problematic digital media use during the COVID-19 pandemic. Specifically, we examined whether the relationship between excessive gaming and PGU differs based on the level of loneliness and whether this interaction effect changes depending on the presence or absence of a housemate or household size. The results indicated that the relationship between gaming time and PGU was moderated by loneliness, such that the positive effect of gaming time on PGU in male gamers was greater when the level of loneliness was high. In addition, the interaction effect of gaming time and loneliness on PGU differed depending on whether the participants lived with housemate(s). The moderating effect of loneliness on the relationship between gaming time and PGU was significant only for participants living alone. Consequently, Hypotheses 1 and 2 of this study were supported. However, Hypothesis 3 was not supported. In other words, household size did not moderate the interaction between gaming time and loneliness among male gamers living with housemates.

As expected, loneliness seems to serve as a “railroad switch” that drives intense gaming to become associated with PGU. Compared to less lonely gamers, lonely male gamers may show a steeper rise in the likelihood of reporting PGU symptoms, such as dysfunction or loss of control, as game usage time increases. This could be attributed to factors such as wanting to ‘escape’ from the real world and identity crises [17,28], which are known to increase vulnerability to PGU [21,41].

However, the role of housemates in the relationship between loneliness, gaming time, and PGU seems complex. Our findings show that the presence of housemates may be a protective factor that decreases the risk of PGU among lonely gamers with excessive gaming time. For gamers living with housemates, the level of loneliness did not change the relationship between gaming time and PGU, although it increased the risk of PGU. Moreover, the presence of housemates seems to buffer the effect of excessive game use on PGU. This is probably because housemates can prevent gamers from being left alone for too long [38], and monitor their behavior [39]. By doing so, housemates can help gamers maintain certain levels of daily function, even if they spend excessive time playing games. Contrary to our expectations, in the living together group, having a greater number of housemates did not change the moderating effect of loneliness in the relationship between gaming time and PGU, due to the potential conflicts arising from having more people in the household [44,45], which can offset the positive effects of having housemates.

Interestingly, as shown in Table 4, the effect of loneliness is rather buffered by living alone. We conducted a supplementary analysis to clarify this relationship. The results of PROCESS macro Model 1 revealed that the effect of loneliness on PGU was significant only in the living together group (B=2.070, 95% CI=[1.415–2.725]) and not in the living alone group (B=0.539, 95% CI=[-0.488–1.566]). This suggests that individuals who experience high levels of loneliness despite living with housemates may have PGU vulnerabilities that cannot be explained by excessive game use alone. Even if gaming time is somewhat modulated by the presence of housemates, lonely gamers may repeatedly experience loss of self-control or negative consequences while immersing themselves in the game whenever they have a chance. In other words, loneliness seems to play an important role in increasing male gamers’s risk of PGU, regardless of living arrangements, but may show group differences in underlying mechanisms (in particular, whether loneliness produces a synergistic effect with excessive game use) or symptom patterns. Thus, it could be useful to investigate whether there are differences in the PGU patterns of male gamers, depending on the presence of housemates.

This study has a few limitations. First, it was impossible to evaluate the direction of causality in the relationship between the variables using a cross-sectional design. For example, although our model posits that loneliness precedes PGU, PGU may also be a risk factor for loneliness due to increased social isolation. Second, due to data constraints, we were unable to analyze the effects of potential factors that may be involved in the PGU mechanism, such as mood symptoms, gaming media, and preferred genre [10,16,17,20,29,41,51]. Future studies that include these factors may further clarify the relationship between the variables. The generalizability of findings is also a concern. Our results, obtained from a Korean sample, may have been influenced by the country’s sociocultural characteristics. In addition, the criteria for PGU vary slightly across scales, as evidenced by the inconsistent terminology (e.g., game addiction, gaming disorder, internet gaming disorder). Thus, replication is needed to confirm if the findings are consistent in other cultural contexts or when using different PGU scales. Furthermore, we cannot exclude the possibility that the relationship between playing time and PGU, and the factors involved, may vary in certain segments of gaming time (e.g., differences in gaming time may no longer be predictive of PGU risk within individuals above a certain level of gaming time), and future studies should examine the relationship between variables in a more systematic and multifaceted manner, for example, using nonlinear regression or latent profile analysis.

Despite these limitations, this study expands the literature on PGU and offers practical implications for developing intervention strategies to address the increasing problems in modern society. Our findings suggest that loneliness, living arrangements, and gaming time may be important factors to consider when screening high-risk individuals for PGU as well as potential targets for intervention. For example, among male who live alone, the level of loneliness can be a useful criterion for determining the risk status of PGU among game overusers. However, the importance of usage time in screening and intervention may be relatively low in male gamers who feel lonely despite living with housemates. In such cases, it would be better to pay more attention to other symptoms of PGU (e.g., impaired self-control, high priority given to gaming, and functional impairments) rather than focusing on excessive game use. Intervention strategies that can be applied to both groups may include alleviating loneliness, exploring adaptive coping strategies, and creating alternative social support systems. The acquisition of adaptive coping strategies for negative affect such as loneliness is an important component of cognitive behavioral therapy, and its effectiveness on PGU has been proven in previous studies [52]. Recently, online interventions based on digital technology have been proposed as an alternative for reducing loneliness and related mental health problems [53,54]. Our findings provide empirical evidence supporting this direction of intervention.

Article information Continued