Psychological distress and aggression among adolescents with internet gaming disorder symptoms

,


Introduction
As of December 2020, the number of Chinese netizens reached 989 million, accounting for 70.4 % of the total population (CNNIC, 2020).However, excessive internet use, especially online gaming, have become a significant concern.The American Psychiatric Association (APA) included internet gaming disorder (IGD) in section III of the Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition-Text Revision (DSM-5-TR) (O'Brien et al., 2022).The World Health Organization (WHO) also included gaming disorder (GD) in the 11th revision of the International Classification of Diseases (ICD-11) and recently recognized it as a diagnosis code (WHO, 2019).Previous studies have shown that the prevalence of IGD ranges from 0.7 % to 27.5 % (Mihara and Higuchi, 2017).However, a recent meta-analysis indicated that the worldwide prevalence of gaming disorder was 3.05 % (Stevens et al., 2021).
Another meta-analysis revealed similar results, with a prevalence of 3.3 % (Kim et al., 2022;Mihara and Higuchi, 2017).In addition, multiple studies have found that the prevalence of IGD among Chinese adolescents ranges from 3.5 % to 17 % (Long et al., 2018;Yu et al., 2021).Moreover, there are multiple factors that contribute to the development and maintenance of adolescent IGD, especially the mental health problems experienced during the pandemic (Macur and Pontes, 2021;Paulus et al., 2018;Przybylski et al., 2017).Therefore, there is still a need for more comprehensive and larger sample studies on adolescent IGD.Some studies have found that excessive online gaming over a long period of time can cause physical and mental health problems in teenagers and even trigger IGD (CNNIC, 2020;Chia et al., 2020).In recent years, violent crimes have occurred among teenagers who are addicted to online games, but few studies have conducted an in-depth analysis of the relationship between IGD symptoms and aggressive behavior.In fact, aggressive behavior is also common in adolescents and is a characteristic of many mental disorders in this age group (Buchmann et al., 2014).Therefore, it is still unclear how aggressive behavior in individuals with IGD differs from that in individuals with other mental disorders.
Aggressive behavior takes various forms in organisms and has varying meanings, but unmanaged aggression often leads to negative outcomes (WHO, 2023).Aggression is a complex, heterogeneous, and multifactorial phenomenon.Presently, two distinct forms of human aggression are widely acknowledged: impulsive aggression (IA), also referred to as reactive aggression, and premeditated (PM) aggression, alternatively known as instrumental or proactive aggression (Barratt et al., 1997;Cruz et al., 2019;Stanford et al., 2003).Additionally, there are other classifications, which encompass subtypes such as direct versus indirect (Buss, 1961), physical versus verbal (Buss, 1961), and rational versus manipulative aggression (Björkqvist et al., 1992).The primary distinction between impulsive and premeditated aggression lies in the level of motivation and behavioral control exhibited in aggressive events (Cruz et al., 2019).IA tends to involve highly spontaneous, uncontrolled, and unpredictable reactions to a perceived threat.Conversely, PM corresponds to planned actions designed to achieve specific predefined goals (Cruz et al., 2019).To assess IA and PM, researchers at Stanford University (Stanford et al., 2003) developed the Impulsive/-Premeditated Aggression Scale (IPAS).The IPAS is currently being employed in various studies to evaluate aggressive behavior.
Researchers have been continuously exploring the relationship between adolescent online gaming and aggressive behavior (Joshi et al., 2022;Li et al., 2023).However, few studies have directly associated IGD with aggressive behavior.In the early 21st century, many researchers focused on the influence of the media on adolescent aggressive behavior (Browne and Hamilton-Giachritsis, 2005;Hollingdale and Greitemeyer, 2014;Li et al., 2023).With the evolution of online gaming, researchers have investigated the potential link between online gaming and violent or aggressive behavior, but the results are inconsistent and mixed (Coyne et al., 2018;Kowalski et al., 2014;Li et al., 2023).A randomized clinical trial revealed that habitual exposure to violent media is a risk factor for dangerous behavior around firearms (Chang and Bushman, 2019).Additionally, students with longer (versus shorter) gaming durations are more likely to report weapon-carrying and feeling unsafe in school (Zhai et al., 2020).However, some studies reported a weak correlation between playing video games and aggression (Ferguson, 2015), indicating that these correlations were not significant over a 5-year period (Coyne et al., 2018).Moreover, aggression is complex, heterogeneous, and multifactorial (Elhai et al., 2019).The lack of investigation into the relationship between internet gaming-related factors and aggression limits the prevention of potential aggressive behavior among adolescents with IGD.
It is well known that one of the psychological characteristics of adolescents is emotional instability and impulsive behavior.Previous studies have suggested that teenagers who excessively and habitually play internet games are at a high risk of performing violent acts such as suicide and aggressive acts.First, adolescents are prone to addictive behaviors (Hammond et al., 2014;Romer et al., 2017).The increased activity of the nucleus accumbens during adolescence may be related to the increase in impulsive and risky behaviors observed during this period of development (Casey et al., 2008).During the juvenile stage, testosterone levels are elevated compared to those in the early stages of development, promoting aggression and anger (Lee et al., 2017).In addition, according to the frustration-aggression hypothesis (Breuer and Elson, 2017), maladaptive smartphone use is linked to increased annoyance and negative affect and may increase violent tendencies (Bersani et al., 2022;Kim and O'Brien, 2018).Some researchers believe that IGD may share the same neural networks and neurobiological mechanisms as aggressive behavior (Hahn and Kim, 2014;Mohammadi et al., 2020).Two other studies demonstrated a correlation between the severity of IGD and the levels of depression, anxiety, and distress in teenagers (Li et al., 2021;Wu et al., 2018).A recent study (Zhao et al., 2020) also showed that internet addiction exacerbates depressive symptoms and aggressive behavior over time.
Therefore, we hypothesized that IGD symptoms in adolescents would be associated with their aggressive behavior and that some risk factors for IGD symptoms could increase the aggressive tendencies of adolescents.To verify the above hypothesis, we first investigated the prevalence of IGD in a large sample of adolescents, determined the risk factors for IGD symptoms, and then analyzed the relationship between IGD symptoms and aggressive behavior.Finally, we utilized multiple linear regression analysis to explore the impact of IGD-related factors on adolescent aggression.

Study design and participants
From August to October 2020, a cross-sectional survey of junior and senior high school students from southern, southwestern, central, and eastern China was conducted.All students from selected classes were invited to participate in the study.A total of 9364 questionnaires were collected (with a response rate of 100 %).However, due to the missing rate of more than 20 % of the response items, 72 questionnaires were excluded from the data analysis.The final dataset included responses from 9214 students with a mean age of 15.80 ± 1.46 years old, which were deemed suitable for subsequent statistical analysis (see Table 1).The questionnaires were anonymous to ensure the confidentiality and reliability of the data.
The self-report questionnaires were designed to survey demographic information, gaming-related factors, and mental health via the WeChatbased survey program Questionnaire Star, which is openly accessible online to students.Participants were allowed to terminate the survey at any time they desired.A scripted set of instructions was used to introduce the questionnaires, including a statement that there were no right or wrong answers to the questions being asked and that the purpose was to identify how the participants felt about or experienced the topic under investigation.The consent form and all study procedures were approved by the Human Subjects Committee at Zhongshan School of Medicine, Sun Yat-sen University.Written informed consent was obtained from all subjects before the study.

Survey instruments
The questionnaires covered sociodemographic characteristics, gaming-related factors, IGD symptoms, psychological distress, and aggression.Participants provided basic personal information, including sex, age, place of residence, school grade, and other relevant details.

Gaming-related factors
Games were divided into 13 categories (CNNIC, 2020), of which multiplayer online arena games (MOBAs) were the most popular (66 %).Participants also needed to provide the characteristics of the online games they played, the time they spent playing games, and parental interference.Gaming frequency, gaming expense, game activities, and love for gaming were also rated by respondents on a 5-point scale ranging from 0 ("not at all") to 4 ("extremely") (the specific parameter concepts are detailed in Table S1).
A favorite video game list approach was used to measure violent game exposure (VGE) (Anderson and Dill, 2000).First, participants were needed to write the names of their three favorite video games.Then, they reported how frequently they played each game using a five-point rating scale (e.g., 1 = sometimes; 5 = very often) and how violent each game was using a five-point rating scale (e.g., 1 = not at all; 5 = very much).Game violence was calculated by multiplying the violence content rating and frequency rating.The average score of these three video games was used to measure the exposure to violent video games.This measurement method has been widely used in previous studies (Anderson and Dill, 2000;Fikkers et al., 2017), including those conducted among Chinese adolescents (Teng et al., 2017;Teng et al., 2019).For the scores of the three video games, the internal consistency coefficient was 0.760.

IGD symptoms
IGD symptoms were assessed using the Internet Gaming Disorder Test (IGD-20 test) (Pontes et al., 2014).This test reflects the nine criteria for IGD as stated in the DSM-5 and incorporates the theoretical framework of the components model of addiction for assessing IGD tendencies.This is a validated and well-established scale that has been used in many studies and translated into Spanish, Portuguese, Arabic, and Chinese (Fuster et al., 2016;Shu et al., 2019).Participants rate the 20 items on a 5-point Likert scale (1 -strongly disagree to 5 -strongly agree).In the present study, the scale had excellent internal validity (Cronbach's alpha = 0.93).

Psychological distress
The Kessler Psychological Distress Scale (K10) was designed by Professor Ronald C. Kessler from Harvard University as part of the mental health component of the US National Health Interview Survey (Andrews and Slade, 2001).The questionnaire consists of 10 questions designed to measure the level and severity associated with psychological symptoms in population surveys (Krueger, 1999;Vollebergh et al., 2001).The K10 measures the level of anxiety and depressive symptoms and has been used to assess psychological distress across multiple settings and populations, including adults, adolescents, and older adults, from Australia, South Africa, France, New Zealand, and China.In the present study, the scale had excellent internal validity (Cronbach's alpha = 0.95).

Aggression
Aggression was assessed using the Impulsive/Premeditated Aggression Scale (IPAS) (Stanford et al., 2003), which is a self-reported, self-administered scale designed to assess impulsiveness and premediated aggressive behavior (Cruz et al., 2019).The IPAS comprises 30 items rated on a 5-point Likert scale (1 -strongly disagree to 5 -strongly agree) and measures aspects of planning and control during aggressive acts.The scale can be scored either dimensionally or categorically.The primary classification used is the bimodal classification, in which aggressive behaviors are classified into two distinct subtypes: impulsive aggression (IA) and premeditated aggression (PM).Discrete categories (impulsive vs. premeditated) are obtained by a categorical approach, in which only the percentage of positive items (5 -strongly agree or 4agree) for each aggression scale is calculated.Therefore, not all participants may be accounted for in a categorical scoring, but rather only those scoring 4 or 5 on the items of each subscale (Cruz et al., 2019).The IA and PM scales have demonstrated high internal consistency across studies.In the present study, the scale had excellent internal validity (Cronbach's alpha = 0.94).

Statistical analysis
The data were analyzed using SPSS V.23.0.Statistical significance was defined as a two-tailed p value less than 0.05.We examined the distribution of the variables and applied transformations to normalize the distributions when needed.Univariate normality was checked by examining skewness and kurtosis values.Values within < |1| suggested the absence of severe violations of normality (Tabachnick and Fidell, 2007).The variance inflation factor (VIF) was utilized to examine the multicollinearity issue (1 < VIF < 3).If the data did not follow a normal distribution, we applied a ln-transformation to the VGE score.Chi-square tests and analysis of variance (ANOVA) were utilized to analyze the impact of the variables on IGD, assuming that the data were normally distributed.Furthermore, logistic regression was conducted to examine the factors associated with the development of IGD symptoms in adolescent gamers, while multiple linear regression was performed to investigate the effect of psychological distress and IGD symptoms on aggression.

Demographic characteristics and gaming-related factors among different gamers
Among the 9,364 students who completed the questionnaire survey, 150 students with missing data were excluded, leaving 9214 students in the valid sample.Table 1 shows the demographic information of the participants.Of the included students, 7318 students (79.42 %) reported playing games, while 1,896 students (20.58 %) did not.A probable IGD diagnosis was found in 164 (1.78 %) students.The prevalence of probable IGD was significantly different between junior and senior high school students (χ2 =15.95, p < 0.001).
All students were first divided into two groups: those who did not play games (the nongaming group, NG group) and those who did (the gaming group).Within the gaming group, participants were further divided into three subgroups based on their IGD-20 test scores: the casual gamer group (CG), risk gamer group (RG), and disordered gamer group (DG).The grouping criteria were as follows: CG group (score less than 50), RG group (score greater than 51 but less than 70), and DG group (score greater than or equal to 70) (Hawi et al., 2018).
As shown in Table 2, there were significant differences in all the variables among the different gamer groups (all p < 0.05), including gaming-related factors such as the importance of games and the duration of gaming.Significant sex differences were observed (χ2 = 364.68,p < 0.001).A total of 32 females (19.5 %) and 132 males (80.5 %) were included in the DG group.The RG group included 334 females (25.8 %) and 959 males (74.2 %).Age differed significantly among the three groups, with the RG and CG groups being older than the IG group, suggesting that students with IGD tend to be younger.Regarding parental interference, pairwise comparisons indicated significant differences between groups.Notably, the DG group (20.7 %) had the least parental interference among the three gaming groups.Moreover, there was a significant difference in VGE among the three gamer groups (p < 0.001).Pairwise comparisons showed significant differences among the groups, indicating that students with IGD were more likely to be exposed to violent games than other students.

Factors associated with the IGD symptoms in adolescent gamers
The independent variables entered into the stepwise binary logistic regression equation included sex (OR=0.376;95 % CI: 0.235-0.602),degree of gaming interest [not really (OR=0.167;95 % CI: 0.031-   3).Logistic regression analysis showed that these factors were closely associated with the development of IGD symptoms.

Influence of IGD symptoms on the development of psychological distress and aggression
In addition to obvious differences in the severity of IGD symptoms, the three gamer groups (CG, RG, and DG groups) experienced different degrees of psychological distress and aggression.ANOVA and pairwise comparison (Bonferroni) showed that the DG group had more severe IGD symptoms and higher levels of depression and anxiety than the CG and RG groups (p < 0.05).Interestingly, it was observed that the CG group, rather than the NG group, demonstrated the lowest scores for both psychological distress and aggression compared to other groups, a difference that was statistically significant (p < 0.05).Concurrently, the DG group exhibited the highest levels of aggression, as evidenced by their elevated IPAS scores, including both the IPAS-IA and IPAS-PM (as in Table 4).

Discussion
The present study is the first large-scale survey of IGD among Chinese adolescents.We investigated the risk factors for IGD and analyzed the factors related to aggressive behavior.The results showed that the prevalence of IGD symptoms was 1.78 % among Chinese adolescents.The adolescents in the disordered gamer group had the most severe IGD symptoms and the highest levels of psychological distress and aggression.Moreover, higher aggression was significantly associated with younger age, male sex, more severe psychological distress and IGD symptoms, and more violent game exposure.Our results further suggested that excessive online gaming may contribute to serious psychological distress and aggression in adolescents.Apart from male sex and younger age, severe IGD symptoms and psychological distress are the most important predictors for the development of aggressive behavior.Interestingly, appropriate leisure gaming may have a positive impact on the psychosomatic health of adolescents.

The prevalence of IGD among adolescents
Our results showed that the prevalence of IGD symptoms among Chinese youth was 1.78 %, but the prevalence of IGD symptoms among adolescent gamers was 2.2 %, which is slightly lower than that in a previous report.A recent meta-analysis of all regions of the world showed that the prevalence of IGD was 3.05 % (CI: [2.38,3.91]).If only studies that met more stringent sampling criteria were included, the prevalence would drop to 1.96 % (Stevens et al., 2020).A survey before the COVID-19 pandemic found that the prevalence of IGD among Chinese adolescent gamers was 17.0 % (Liao et al., 2020).However, the average age of the participants was 19.67 years old, including some college students who had more leisure time to play online games than middle-and high-school students devoted to studying for the entrance examination.In some Chinese cities, such as Macau, the estimated prevalence of probable IGD among teenagers was 2.0 %, which is very close to our findings (Wu et al., 2018).Significant differences were found in age and educational attainment between disordered gamers and casual gamers (Wu et al., 2018), indicating that junior school students are more likely to develop this disorder.

Factors associated with IGD symptoms
Our study found that the degree of gaming interest, frequency of game consumption, frequency of participation in online game activities, duration of playing games, average daily game time, violent game exposure, and psychological distress are significantly associated with the development of IGD symptoms.The results of several previous studies are inconsistent, possibly due to differences in brain development, increased gonadal hormones, and differences in developmental environments (Casey et al., 2008;Gentile et al., 2017).Demographic differences are mainly reflected in sex.The prevalence of IGD in males is approximately four times that in females.Neuroimaging studies have shown that internet gaming triggers more brain responses related to Notes: Chi-square value =601.335,p < 0.001.Abbreviations: 95 % CI=the 95 % confidence interval for the estimated odds ratio; SE=estimated precision of the coefficients.
X. Deng et al. craving in males than in females (Wang et al., 2019).In the current study, male sex was regarded as an important predictor of IGD, which has also been reported in several previous studies (Lopez-Fernandez and Kuss, 2020;Mannikko et al., 2015;Teng et al., 2020).Several studies have found that certain gaming-related factors contribute to the development of IGD, particularly the duration of playing games and average daily game time (Vally, 2020;Zajac et al., 2020), which may imply that paying attention to the time-related aspects of adolescent gaming behavior may be more helpful in preventing the occurrence of IGD.
Other studies have demonstrated that a considerable number of adolescents are exposed to violent content through online gaming and that this exposure is associated with IGD (Cuong et al., 2021;Hou et al., 2022).Our study also reached similar conclusions, which may be due to the different prevalence of violent content in many games, and more stimulating information is more likely to attract adolescents in the midst of puberty.It is worth noting that MOBAs are the most popular games among Chinese teenagers (66 %), probably because social demand is one of the reasons why teenagers play online games (Demetrovics et al., 2011).In China, the majority of adolescents predominantly engage in playing "Honor of Kings".These types of MOBA games usually contain mild violence and have strong social elements.IGD is more likely to develop among adolescents who experience intense social anxiety (Marino et al., 2020).T 'Ng et al. (2022) believes that MOBA players who encounter frustration in meeting their basic needs may become succumb to uncontrollable gaming behaviors (T' Ng et al., 2022).Given the context of mobile MOBA games, it is necessary to further study the potential relationship between MOBA games and IGD in order to reduce the negative impacts of online gaming on adolescents.

IGD and aggression
We found that compared with students in the CG and RG groups, those in the DG group had more severe IGD symptoms and psychological distress (depression and anxiety) and the highest level of aggression (IA and PM).The link between IGD and psychological distress may be bidirectional.
According to the Person-Affect-Cognition-Execution (I-PACE) model, adolescents may engage in addictive behaviors under stress, which can lead to psychological distress (Brand et al., 2019).Conversely, recent longitudinal studies have suggested that internet addiction may also contribute to psychological distress (Chen et al., 2020;Yu and Shek, 2018).This may be because adolescents with IGD often become addicted to gaming, resulting in problems such as insufficient sleep (Lin et al., 2021).However, the mechanisms behind this phenomenon remain unclear.Several studies have indeed revealed that individuals with IGD have a higher risk of developing aggressive and violent behavior (Teng et al., 2020;Zhu et al., 2020).
We also found that compared with students in the CG and RG groups, those in the DG group had the highest level of aggression.The latest meta-analysis by Li et al. (2023) found a moderate relationship between IGD and aggression (r = 0.300, 95 % CI [0.246, 0.353]).Certainly, social learning theory posits that people acquire aggressive behaviors through observation (Bandura and Walters, 1977).Therefore, certain studies believe that popular MOBA games often include violent content, which may lead escalated aggression among players.However, it is also plausible that individuals inclined toward impulsive/aggressive behavior might have a higher affinity for games with substantial violent content (Lee et al., 2017).More notably, the decreased impulse control among individuals with IGD might contribute to higher levels of aggression.It is also noteworthy that individuals with IGD have reduced cognitive control (Lee et al., 2021).These defects might contribute to the higher levels of aggression in individuals with IGD.
Interestingly, the CG group (but not the NG group) had the lowest psychological distress and aggression scores, indicating that appropriate leisure gaming may have positive significance for the psychosomatic health of adolescents.Casual gaming could alleviate psychological stress and reduce the level of aggression among adolescents.The catharsis theory of gaming holds that individuals use online gaming to alleviate stress or fulfill unmet control needs from real life (Ferguson and Olson, 2013).This theory may explain the findings from previous research by Agarwal et al. (2017), which indicated a negative correlation between aggression and online gaming addiction.Nevertheless, with an increased risk of developing IGD, once a certain threshold is surpassed, adolescents tend to exhibit higher levels of psychological stress and aggression.Hence, determining this 'threshold' is not solely needed for the diagnosis of IGD but is also a significant concern for researchers to address.

IGD-related factors associated with aggression
To further investigate the relationship between IGD-related factors and aggression, we performed linear regression and found that a higher level of aggression was significantly associated with younger age (β = -0.35),male sex (β = 0.858), severe psychological distress (β = 0.425), severe IGD symptoms (β = 0.311), and violent game exposure (β = 0.059).
Age is a moderating factor in the relationship between IGD and aggression (Li et al., 2023).The correlation between IGD and age is highest among elementary school students, followed by college students, and the lowest among middle school students.Previous studies did not emphasize this correlation among middle school students, and our results also demonstrated a similar age-related effect.Within the middle school cohort, both IGD and aggression display a negative correlation with age.Increased self-control increase, mental maturation, and brain development in adolescence contribute to the decline of IGD and aggression levels.Furthermore, due to the college entrance examination, senior high school students face much greater academic stress than junior high school students, resulting in reduced engagement in online gaming.Some researchers have proposed that decreased exposure to violent factors might reduce the degree of aggression.However, our results do not seem to support this view because video game exposure (VGE) is relatively weak in predicting aggressive behavior (β = 0.059).
Predictably, there was a strong correlation between sex and adolescent aggression, particularly in males (β= 0.858).The prevailing trend was observed across most studies (Joshi et al., 2022;Kim et al., 2018;Li et al., 2023), consistent with the predominance of males in both aggressive behavior and IGD-related research.It is unclear whether this is due to a higher male preference for gaming or whether most online games, including MOBAs (Multiplayer Online Battle Arena Games) and ACT (Action Games), primarily target male audiences.
The level of psychological distress is moderately associated with aggression (β = 0.425).Moreover, the CG group exhibited lower psychological distress and aggression, and their scores were even lower than those of students who did not play games.There was a similar trend in psychological distress and aggression.Magnetic resonance imaging studies also revealed that there may be a common neural mechanism between IGD and aggression, such as the fronto-limbic-striatal circuit (Ko et al., 2015;Yang et al., 2017).In addition, IGD has been shown to impair emotional regulation and increase impulsivity, both of which independently increase vulnerability to engaging in aggressive behaviors (Davidson et al., 2000).Moreover, functional abnormalities associated with emotion regulation can increase the propensity for aggression (Gilam et al., 2018;Lin et al., 2021;Yang et al., 2017).
In conclusion, in this study, we found that excessive online gaming not only contributed to psychological distress in adolescents but also increased the level of aggressive behavior.Apart from male sex and younger age, severe IGD symptoms and psychological distress are the most important predictors of the development of aggressive behavior.In addition, although the sample size of this study was relatively large, we adopted a cross-sectional survey strategy to collect data on Chinese adolescents, which may lack much important information from longitudinal research.Moreover, due to anonymous investigation, we also did not interview every subject who met the scoring criteria for IGD and received a clinical diagnosis.These limitations need to be addressed in future longitudinal clinical studies.

Table 1
Demographic characteristics of the participants (n = 9214).
Note: n: number; M: mean; SD: standard deviation; y: years X.Deng et al.

Table 2
Comparisons of gaming-related factors among the different gamer groups.

Table 3 .
Logistic regression model of IGD symptoms development among adolescent gamers.

Table 4
Differences in IGD symptoms, psychological distress, and aggression among the groups.