Current Addiction Reports

https://doi.org/10.1007/s40429-020-00339-3

GAMING (H-J RUMPF, SECTION EDITOR)

Gaming Disorder Across the Lifespan: a Scoping Review

of Longitudinal Studies
Jérémie Richard 1 & Caroline E. Temcheff 1 & Jeffrey L. Derevensky 1

Accepted: 6 October 2020

# Springer Nature Switzerland AG 2020

Abstract
Purpose of review Gaming disorder is a mental disorder characterized by impaired control over gaming behaviors resulting in the
escalation of gaming despite significant negative consequences and functional impairments. This scoping review aims to
synthesize empirical evidence for the development of gaming disorder based solely on extant findings from longitudinal studies.
Recent findings Although research efforts investigating gaming disorder are expanding, there is a need to conceptualize this
disorder from a developmental framework in order to identify trajectories of this disorder across the lifespan while accounting for
both risk factors and consequences of GD.
Summary The scoping review identified 57 research papers identifying antecedents (n = 33), consequences (n = 24), and vari-
ables having reciprocal relationships with GD (n = 9). The temporal stability of GD differed based on age and study duration,
ranging from 20 to 84%. Commonly reported risk factors for GD include emotion dysregulation and negative self-esteem, with
depressive symptoms, inattentive symptoms, and social isolation being reciprocally associated with GD. Consequences of GD
include increases in anxiety and poor parental relationships and decreases in life satisfaction and academic performance. Overall,
various psychosocial factors interact to influence the development of GD across the lifespan, while certain consequences may
maintain GD symptomatology.

Keywords Development . Longitudinal studies . Gaming disorder . Problem gaming . Review . Risk factors

Introduction of the DSM-5 [2], a substantial amount of research has gone

into understanding GD and has resulted in the diagnostic
Gaming disorder (GD) is an officially recognized mental dis- criteria as outlined in the ICD-11 [1]. According to the
order in the eleventh edition of the of the International WHO [1], GD is defined as a pattern of persistent or recurrent
Classification of Diseases (ICD-11; World Health video gaming behavior over a period of at least 12 months that
Organization [WHO]) [1]. GD was first introduced as a disor- is characterized by an impaired control over gaming behav-
der requiring further research in the Diagnostic and Statistical iors, increasing priority given to gaming, and a continuation or
Manual of Mental Disorders, Fifth Edition (DSM-5; escalation of gaming despite negative consequences or func-
American Psychiatric Association [APA]) [2] with the pro- tional impairments. A review of epidemiological studies sug-
posed criteria for GD (referred to as “Internet gaming disor- gests that the prevalence of GD ranges from 0.7 to 27.5%
der”) being based on the diagnostic criteria for substance use globally, with European and North American estimates rang-
disorders and gambling disorder [2–4]. Since the publication ing from 1 to 10% [5]. Higher rates of GD have been reported
among children and adolescents when compared to adults
[6–9], with males reporting higher rates of GD than females
This article is part of the Topical Collection on Gaming
[10].
* Jérémie Richard
A majority of past research investigating GD has been
[email protected] cross-sectional, with the purpose of identifying factors associ-
ated with problematic and disordered gaming. In a recent
1 meta-analysis of GD research across 20 countries, Cheng
International Centre for Youth Gambling Problems and High-Risk
Behaviors, McGill University, 3724 McTavish Street, and colleagues [11] reported a moderately strong positive as-
Montreal, QC H3A 1Y2, Canada sociation between GD symptoms, psychological problems,
 Curr Addict Rep

and interpersonal problems. Furthermore, in a comprehensive Fourth, Brand and colleagues [25, 26] developed the
review, González-Bueso and colleagues [12] reported that the Interaction of Person-Affect-Cognition-Execution (I-PACE)
largest observed associations by effect size between GD and model of addictive behaviors. The authors argue that various
comorbid psychopathology were for depression, anxiety, and predisposing characteristics interact with affective, cognitive,
attention-deficit hyperactivity disorder. Other systematic re- and executive processes to predict the development and main-
views [5, 6, 13] have reported associations between GD and tenance of GD [26]. Additionally, the I-PACE model specifies
certain personality traits, motivations for playing, psychiatric the role of environmental and media-specific factors, while
conditions, and psychosocial consequences. Beyond cross- differentiating between the mechanisms involved in the early
sectional studies, researchers have advocated the need for lon- and later stages of addiction [25].
gitudinal studies [5, 12, 13] while conceptualizing GD from a Given the partial overlap, strengths, and limitations of the
developmental framework [6, 14–17]. As GD has been de- abovementioned models and the need to understand the de-
scribed as being progressive in nature [18], it is important to velopmental course of GD across the lifespan, a review and
consider how the disorder may present itself over time, while integrative synthesis of longitudinal studies is warranted.
linking gaming to normal and abnormal developmental pro- Moreover, although all four models claim to be specific to
cesses [16, 19–21]. Indeed, the time at which an individual the development of GD, their propositions are not solely based
begins developing symptoms of GD may be related to the on findings from longitudinal research. As such, the goal of
presence of certain psychosocial vulnerabilities. As such, ex- the present review is twofold: (1) to identify and evaluate
amining developmental trajectories while measuring both dis- prospective evidence for the temporal trajectories, risk factors,
tal and proximal risk and protective factors is necessary when and consequences of GD across the lifespan; and (2) evaluate
attempting to understand this disorder. existing theoretical models for the development of GD based
To date, four theoretical models have been proposed to on empirical evidence. For the first goal of the present review,
explain the development of GD. First, Benarous and col- research questions include the following: (i) What is the tem-
leagues [22] outlined an internalized and an externalized path- poral stability of GD?; (ii) What are risk factors or antecedents
way to GD based on case reports. According to this model, the of GD?; (iii) What are the consequences or outcomes of GD?;
internalized pathway leads to GD via attachment issues during and (iv) What factors have reciprocal relationships with GD
infancy, internalizing symptoms during childhood and adoles- across time? For the second goal of the review, the four de-
cence, and behavioral avoidance and social withdrawal as a velopmental models will be evaluated based on the existing
young adult. Alternatively, the externalized pathway leads to evidence and a comprehensive conceptual model of longitu-
GD via externalizing problems, severe family conflicts and dinal findings will be proposed.
school difficulties during childhood, poor emotion regulation
as an adolescent, and aggressive behaviors and poor social
competencies as a young adult.
Second, Lee, Lee, and Choo [23] proposed a typology of Method
GD based on the pathways model of problem gambling [24].
Lee and colleagues [23] described three types of problem Literature Search
gamers based on the salience of biological, psychological,
and social vulnerabilities: (1) impulsive/aggressive problem Search Strategy A scoping review was conducted to identify
gamers (individuals playing games to release their aggressive longitudinal studies investigating the temporal stability of GD,
impulses, seek sensations, and alleviate boredom), (2) emo- in addition to the predictors and consequences of problem and
tionally vulnerable problem gamers (individuals playing disordered gaming over time. Peer-reviewed journal articles
games for escapism or mood-modification purposes), and (3) were collected primarily through a search of four databases:
socially conditioned gamers (individuals playing games to PsycINFO, Medline, PubMed, and Scopus. Keywords for the
meet their social needs and reduce feelings of loneliness). database searches were developed through the screening of
Third, Paulus and colleagues [13] developed an inte- articles investigating GD and various biopsychosocial factors
grated model of GD based on a summary of etiological (see Table 1 for keywords). The following simplified search
knowledge whereby both internal (e.g., structural brain string with appropriate Boolean operators (modified to reflect
deficits, inattention, impulsivity) and external (e.g., social the search logic of each database) was used to identify studies:
factors, game-related factors, parental influences) factors (gaming keywords) AND (biological keywords) OR (neuro-
predict GD. Although they do not specify the temporal logical keywords) OR (cognitive keywords) OR (psycholog-
sequencing of these problems, they provide a second ical and emotion-related keywords) OR (addiction and sub-
model outlining the pathway from gaming fun to addicted stance use keywords) OR (social and environmental key-
gaming that is characterized by a loss of control and pre- words) OR (demographic keywords) OR (psychopathological
occupation over gaming. and temporal keywords) OR (method keywords).
Curr Addict Rep

Table 1 Keywords for database

searches Grouping terms Keywords

Gaming “video game”; “problem* use of video games”; “problem* video game use”; “problem*
gaming”; “excessive gaming”; “internet gaming disorder”; “gaming disorder”; “video
game addiction”; “gaming addiction”
Biological genetic; “twin stud*”; hereditary; puberty; “pubertal timing”
Neurological EEG; fMRI; MRI; “diffusion tensor imaging”; “fractional anisotopy”; “brain structure”;
“orbitofrontal cortex”; “dorsolateral prefrontal cortex”; amygdala; “anterior cingulate
cortex”; “nucleus accumbens”; “white-matter integrity”
Cognitive craving; “impulse control”; tolerance; withdrawal; “executive control”; “time
perspective”; “repetitive behavioral patterns”; compulsion; “cognitive function”;
“cognitive salience”; regret; “behavio*ral salience”; reward
Psychological and psychosocial; “mental health”; psychopathology; wellbeing; “well being”; well-being;
emotion-related pathology; depression; inattention; hyperactivity; “attention deficit hyperactivity
disorder”; loneliness; anxiety; self-esteem; “social anxiety”; personality; comorbidity;
“emotion regulation”; “social skills”; “social deficiencies”; aggressi*; delinquen*;
“conduct problems”; “conduct disorder”; “post-traumatic stress”; antisocial;
anhedonia; motivation; hostility
Addiction and “substance use”; “substance abuse”; alcohol; smoking; cannabis; gambling; “behavioral
substance use addiction”
Social and parenting; “parent-child interaction”; “parental monitoring”; “school functioning”; peer;
environmental “peer victimization”; “deviant peer affiliation”; “early childhood adversity”; “adverse
childhood experience”; trauma; “father-child relationship”; “mother-child
relationship”; “social integration”; “school performance”; “academic performance”;
“quality of life”; “social support”; “prosocial behavior”; “psychosocial development”;
“interpersonal dependency”; hobbies; “leisure time”; occupation; sport; exercise
Demographic sex; gender; race; ethnicity; “socio-economic status”; “family situation”; “parental
education”; “residential area”; siblings; age; education
Psychopathological etiology; consequences; outcomes; harms; distress; dysfunction; concurrent; predict*;
and temporal distal; proximal; “spontaneous remission”; “spontaneous recovery”; treatment;
“treatment response”; “longitudinal course”; “natural course”
Method longitudinal; “longitudinal study”; “longitudinal analysis”; prospective; “developmental
trajectory”; cross-lagged; “autoregressive cross-lagged modelling”; “long-term
evaluation”; “latent growth model”

Note. “;” in the table represents “OR” Boolean operator

Included studies were published between January 2000 publication was English or French. Studies were excluded if
and January 2020. The initial search strategy produced they did not involve original research (e.g., literature reviews,
4578 articles (PsychINFO [2074], MEDLINE [2314], study protocols) and if variables were only measured at a
PubMed [127], and Scopus [27]). Overall, 4215 articles single time-point (e.g., cross-sectional studies, case-control
remained after the removal of duplicates. The titles and studies). Treatment studies were also excluded as the purpose
abstracts of the remaining articles were reviewed for rele- of the present review was not to investigate the effectiveness
vancy and to remove publications that did not contain orig- of various treatments. Studies were excluded if they did not
inal research. Following this initial review, 4095 articles directly measure problem or disordered gaming (e.g., screen
were excluded, with 120 articles remaining for the full text time, sedentary behavior, problematic Internet use) or only
review. Out of these 120, 63 were excluded and an addi- measured gaming frequency or type of video game played.
tional five articles were added after consulting the reference If studies investigated Internet addiction, they were only in-
section of relevant review and empirical articles. The final cluded if they had a measure specific to problem or disordered
collection included 57 articles. The literature review pro- gaming.
cess is outlined in a PRISMA diagram (Fig. 1).
Data Extraction Extracted information from the final articles
Study Selection and Inclusion and Exclusion Criteria Studies included the research questions, study type, duration, mea-
were selected for the review if they included longitudinal or surement points, sample size and characteristics (i.e., age,
prospective data with more than one time point and included a sex, country), measurement tool for problem or disordered
measure of problem or disordered gaming. Populations of all gaming, other measured variables, type of analysis, and the
ages from all countries were included if the language of central results of the study.
 Curr Addict Rep

PRISMA 2009 Flow Diagram

Identification

Records identified through database Additional records identified

searching through other sources
(n = 4578) (n = 0)

Records after duplicates removed

(n = 4215)
Screening

Records screened Records excluded

(n = 4215) (n = 4095)

Full-text articles assessed Full-text articles excluded, with

Eligibility

for eligibility reasons

(n = 120) (n = 63)

Review articles (n = 22)

Consultation of review Treatment articles (n = 28)
article reference list General Internet use/addiction (n = 8)
(n = 4) Gaming without GD measure (n = 2)
Questionnaire development (n = 1)
Consultation of included Combination of other behavioral
article reference list addictions as DV (n = 1)
(n = 1)
Included

Neurological comparison (n = 1)

Studies included in
qualitative synthesis
(n = 57)

Fig. 1 PRISMA flow diagram of study selection for empirical review

Results samples, 23 having over 65% male samples, and 29 having

a representation of both males and females (ranging between
Samples and Demographics 35 and 65% for each sex). No study had a majority female
sample. Studies generally included younger individuals, with
Table 2 summarizes each study included in the systematic two studies focusing solely on children, nine on children and
review. Out of the 57 included articles, there were a total of adolescents, 17 on adolescents, eight on adolescents and
43 different samples. Sample demographics were quite di- adults, and 21 studies including emerging and older adults.
verse, with samples from 13 different countries including
China (n = 12), the Netherlands (n = 9), Australia (n = 8), Study Design and Follow-up Periods
USA (n = 5), Singapore (n = 5), Germany (n = 5), Norway
(n = 4), Switzerland (n = 3), South Korea (n = 2), Canada Although all included studies had a longitudinal design, the
(n = 1), Czech Republic (n = 1), Sweden (n = 1), and Taiwan time period and number of measurement points differed be-
(n = 1). Within the included articles, a larger percentage of tween studies. The most commonly employed longitudinal
males were recruited with five studies having male only designs are as follows: 1-year time range with two
Table 2 Summary of longitudinal articles investigating various factors associated with the development of gaming disorder

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points
Curr Addict Rep

Adams et al. Australia June 2016–September Emerging adults Internet Gaming Anxiety Anxiety at T1 was a risk factor for GD at T3 accounting
[14] 2016 T1: N = 61 Disorder Scale-Short Family cohesion for 8.9% of the variance in GD (β = .1). The
T1 = June 2016 T2: N = 56 Form 9 (9-item) Video game genre interaction between anxiety (T1) and balanced family
T3 = September 2016 T3: N = 43 cohesion (T2) positively moderated GD symptoms
77% male; 23% female (T3) accounting for 17.88% of the variance in GD at
Mage = 23.02 (SD = 3.42, T3 (β = − .04).
range 18–29) The ceiling point for values of balanced family cohesion
was 29.74. After this point, the moderation effect at
T2 on the association between anxiety (T1) and GD
(T3) reached its maximum protective effect (β = .06).
Baggio et al. Switzerland 2010/12–2012/13 Emerging adults Game Addiction Scale Physical health GD symptoms at T1 predicted depression (β = .11),
[28] T1 = 2010–2012 N = 4813 (7-item short form) Mental health physical health/somatic problems (β = − .244), and
T2 = 2012–2013 100% male Depression low mental health-wellbeing (β = − 1.079) at T2.
Mage = 20 (SD = N/A, Video game play Heavy use of video games over time was
range N/A) non-significant in predicting GD.
Brunborg et al. Norway 2010–2012 Adolescents Game Addiction Scale Depression Increases in GD symptoms from T1 to T2 (10%)
[29] T1 = 2010 N = 1928 (7-item short form) Academic achievement predicted increases in depression (2.5%), lower
T2 = 2012 44.5% male; 55.5% Heavy episodic drinking frequency average grade (− 1.7 points) and increases in serious
female Conduct problems (serious, covert, (i.e., theft, vandalism) (3.3%), aggressive (i.e.,
Mage = N/A (SD = N/A, aggressive) violence, swearing) (5.9%) and covert (i.e., truancy)
range 13–17) Video game play conduct problems (5.9%) at T2. Increases in GD
symptoms were not associated with frequency of
heavy episodic drinking.
Burleigh et al. Australia June 2016–September Emerging adults Internet Gaming Depression Higher GAR moderated the relationship between
[30] 2016 T1: N = 61 Disorder Scale-Short Gamer-Avatar Relationship (GAR) depression and IGD whereby GAR exacerbates the
T1 = June 2016 T2: N = 56 Form 9 (9-item) effect of depression on GD (β = 0.02). This model
T3 = September 2016 T3: N = 43 accounted for 47.76% of variance of GD.
77% male; 23% female Moreover, 8% of variance in GD scores at T3 were
Mage = 23.02 (SD = 3.42, explained by depression scores at T1 (β = 0.21) and
range 18–29) 26% of variance in GD scores at T3 were predicted by
GAR at T1 (β = 0.27).
Chang & Lin, Taiwan 2012–2014 Emerging adults Problematic Internet Use Online gaming motives High-engagement gamers had higher escapism motives
[31] T1 = 0 months T1: N = 387 Scale (14-item) (advancement, escapism and (η2 = .663, p < .001) and achievement motives
T2 = 6 months T2: N = 333 socializing) (η2 = .663, p < .001) than all other groups.
T3 = 12 months T3: N = 315 Depression High-engagement gamers had significantly higher
T4 = 18 months T4: N = 308 Academic performance depression scores (η2 = .087, p < .001), academic
T5 = 24 months T5: N = 292 Academic and peer-related stress stress scores (η2 = .071, p < .001), and peer-related
71.8% male; 28.2% Video game play stress (η2 = .079, p < .001) compared to all other
female groups of gamers across all time points.
Mage = 19.4 (SD = N/A,
range N/A)
Singapore 2006–2009 Children, adolescents Relationship between parent and child
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points

Charlie et al. T1 = 2006/7 T1: N = 2527 Pathological Gaming Perception of family environment Those with GD had a significantly different score in
[32] T2 = 2007/8 T3: N = 1363 Scale (10-item) Video game play perception of family environment compared to
T3 = 2008/9 76.5% male; 23.5% non-GD over all 3 years (p < .001).
female Youth with GD had poorer relationships with their
Mage = 12.88 (SD = 2.03, parents than youth without GD. Although this was
range N/A) non-significant at T1, it was significant by T3
(p < .001).
Chen et al. [33] China N/A Adolescents, adults Smartphone Game Intention to decrease play time In those that wanted to resolve symptoms of GD,
T1 = 0 weeks N = 381 Addition Screening Ubiquity thought suppression increased intrusive thoughts
T2 = 6 weeks 48.4% male; 51.6% (4-item) Intrusive thoughts about gaming (β = .506) and impulsive playing (β = .172), which
female Thought suppression led to greater perceived difficulties in quitting
Mage = N/A (92.9% of the Impulsive playing (β = .719) and reduced self-efficacy (β = .332) to
sample 18–40) Perceived difficulty decrease playing time (β = − .457).
Deficient self-efficacy Overall, the model explained 30.5% of the variance in
Video game play decreased playing among players with self-reported
GD.
Choo et al. Singapore N/A Children, adolescents Pathological Gaming Parent-child closeness Higher parent-child closeness at T1 had a significant
[34] T1 = 0 year T1: N = 2974 Scale (10-item) Parental restriction of child gaming main effect, decreasing the number of GD symptoms
T2 = 1 year T2: N = 2601 at T2 (β = − .14). This effect was significantly
72.6% male; 27.4% stronger for boys than girls and for girls, greater GD
female symptoms was associated with greater parent-child
Mage = 11.2 (SD = 2.05, closeness.
range N/A) Parental restrictions of child-video gaming at T1 had no
main effect on GD symptoms at T2.
Coyne et al. USA 2011–2012 Adolescents, emerging Pathological Gaming Respiratory sinus arrhythmia (RSA) Poor self-esteem (β = − .26) and emotion dysregulation
[35] T1 = 2011 adults Scale (11-item) Galvanic Skin Response (GSR) (β = − .28) at baseline predicted GD symptoms
T2 = 2012 N = 374 Both measured during tasks: 1) 1 year later.
48.4% male; 51.6% solving a Rubik’s Cube and 2) Lower RSA predicted GD symptoms (β = .19).
female discussing a family problem Greater GSR predicted GD symptoms in females
Mage = 15.29 (SD = 1.05, (β = − .20), but not males.
range 13–18)
Dang et al. China 2016–2017 Emerging adults DSM-5 criteria for Trait emotional intelligence Emotional intelligence had an indirect protective effect
[36] T1 = April to N = 282 Internet Gaming Coping flexibility on GD symptoms. Depression at T2 fully mediated
May 2016 39.4% male; 60.6% Disorder (9-item; 0–1 Depression the relationship between emotional intelligence at T1
T2 = April to female response) and GD and coping flexibility at T2. Specifically, trait
May 2017 Mage = 20.47 (SD = 1.15, emotional intelligence at T1 predicted greater coping
range 18–29) flexibility at T2 (β = .37), which predicted reduced
depressive symptoms at T2 (β = − .32). Trait
emotional intelligence at T1 predicted reduced
depressive symptoms at T2 (β = − .13) which in turn
predicted greater GD symptoms at T2 (β = .29).
China 2016–2017 Emerging adults
Curr Addict Rep
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points
Curr Addict Rep

Dong et al. T1 = 2016 T1: N = 149 Young’s Internet fMRI scanning: at baseline and 1 year Individuals not meeting the criteria for GD at T1 but
[37] T2 = 2017 T2: N = 103 (N = 23 with Addiction Test & later meeting the criteria at T2 had greater activation of the
GD) Internet Gaming Cue-elicited craving task bilateral lentiform nucleus directly following gaming
69.6% male; 30.4% Disorder Interview but not during gaming. Activation in right lentiform
female (IGDI) gyrus was also correlated with self-reported craving in
Mage = 22.34 (SD = 2.10, those that developed GD (r = .429, p = .04).
range N/A) Deprivation was related to relatively reduced anterior
cingulate cortex activation in both groups.
Dong et al. China 2016–2017 Emerging adults Young’s Internet fMRI scanning: at baseline and 1 year Among individuals no longer meeting the criteria for
[38] T1 = 2016 N = 29 Addiction Test & later GD, decreased activation was identified in the
T2 = 2017 82.8% male; 17.2% Internet Gaming Cue-elicited craving task lentiform nucleus and other reward-related brain
female Disorder Interview regions (e.g., bilateral ACC) after recovery.
Mage = 21.73 (SD = 1.91, (IGDI) Furthermore, increased ACC-lentiform and
range N/A) lentiform-ACC connectivity was found in those that
no longer met criteria for GD.
Ferguson & USA N/A Adolescents Pathological Gaming Attention problems Attention problems (β = 0.19) and male sex (β = 0.46)
Ceranoglu T1 = 0 year N = 144 Scale (7-item) Academic performance (GPA) predicted GD while controlling for age, family and
[39] T2 = 1 years 52.8% male; 47.2% Family attachment peer environment and GD at T1. GD at T1 and hours
female Peer delinquency played per week did not significantly predict attention
Mage = 12.7 (SD = 1.96, problems at T2. Attention problems (but not GD)
range = N/A) predicted lower GPA at T2 (β = − .23).
Forrest et al. Australia N/A Adults Game Addiction Scale Maladaptive gaming cognitions: Those who developed GD at T4 had higher baseline
[40] T1 = 0 months T1: N = 465 (21-item) perfectionism, cognitive salience, perfectionism (d = 1.20), cognitive salience (d = 0.74)
T2 = 3 months T2: N = 374 regret, behavioral salience and regret (d = 0.69) scores than those who remained
T3 = 6 months T3: N = 329 Video game play non-problem gamers. Behavioral salience was
T4 = 9 months T4: N = 290 non-significant.
84% males; 16% females Higher baseline perfectionism (β = .26), cognitive
Mage = 26.8 (SD = 7.59, salience (β = .90) and regret (β = .88) scores and a
range N/A) change in cognitive salience (β = 1.34), regret
(β = 1.32), behavioral salience (β = .46), and weekly
play time (β = .09) predicted GD. Cognitive change
accounted for an additional 28% of the variance in
GD scores beyond gender, age and frequency of
gaming. The entire model explained 73.9% of
variance in GD at T4.
Those who were GD at T1 and became non-GD at T4
had lower baseline perfectionism scores (d = 0.62)
that those who remained with GD.
Gentile et al. Singapore 2007–2009 Children, adolescents Pathological Gaming Impulsivity Of the youth with GD at T1 84% had GD at T3.
[41] T1 = 2007 T0: N = 3034 Scale (10-item) Social competence Risk factors for GD over a two-year period include:
T2 = 2008 T1: N = 2998 Emotion regulation lower social competence (p = .004), greater
T3 = 2009 T2: N = 2605 Depression impulsivity (p = .001), less empathy (p = .006),
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points

T3: N = 2532 Empathy poorer emotion regulation (p = .001), going to LAN

72.7% male; 27.3% Parent-child relationship computer centers (p = .001), and greater identification
female Attention deficit hyperactivity with game characters (p = .001).
Mage = 11.2 (SD = 2.06, disorder GD at T1 also predicted the following at T3: depression
range N/A) Social phobia (p = .009), anxiety (p = .001), social phobia
Anxiety (p = .001), impulsivity (p = .014), poorer relationship
School performance with parents (p = .001).
Video game play Reciprocal relationships were identified for depression,
anxiety, social phobia, and school performance in the
longitudinal growth curve model (p < .001).
Gentile et al. Singapore 2007–2009 Children, adolescents Pathological Gaming Video games in bedroom Bedroom video games and time spent playing video
[42] T1 = 2007 T0: N = 3034 Scale (10-item) Normative beliefs about aggression games at T1 predicted GD at T2 and T3. Overall,
T2 = 2008 T1: N = 2998 Physical aggression bedroom media impact on GD is fully explained by
T3 = 2009 T2: N = 2605 School performance time spent playing (indirect pathway to GD at T2
T3: N = 2532 Sleep time [β = .031]; to GD at T3 [β = .015]).
72.7% male; 27.3% Video game play
female
Mage = 11.2 (SD = 2.06,
range N/A)
Guillot et al. USA 2011–2014 Emerging adults Video Game Addiction Anhedonia Anhedonia at T1 predicted a greater likelihood of GD at
[43] T1 = 0 months N = 503 (1-item) Compulsive Internet use T2 (OR = 1.33, 95% CI [1.11–1.60], p = .003).
T2 = 9–18 months 47.7% male; 53.3% Internet addiction Significant covariates in this model include male
female gender (OR = 3.1, p = .008) and high-school
Mage = N/A (range 19–24) graduation (OR = 3.9, p = .008).
Haagsma et al. Netherlands 2010–2011 Adolescents, emerging Game Addiction Scale Theory of Planned Behavior (TPB; The most important factor of TPB at T1 that predicted
[44] T1 = 0 month adults (7-item short form) attitude, subjective norm, GD at T2 was perceived behavioral control
T2 = 6 months N = 288 perceived behavioral control, (β = − .18), which was negatively related to GD.
72.6% male; 28.4% descriptive norm, social pressure, TPB predictors predicted intention to play video games
female intention) (β = .12), which was then associated with GD
Mage = 15.7 (SD = 1.70, Video game play indirectly through more playing time (β = .37).
range 12–22) Attitude, perceived behavioral control and intention to
play too much at T1 explained 19% of the variances
of GD at T2 after controlling for playing time at
follow-up.
Henchoz et al. Switzerland September 2010–April Emerging adults Game Addiction Scale Level of sport and exercise Cross-lagged correlations: sport negatively predicted
[45] 2013 T1: N = 5223 (7-item short form) GD (β = − .039, p = .024); GD negatively predicted
T1 = 0 months T2: N = 4933 sport (β = − .052, p = .001). There were reciprocal
T2 = 15 months 100% male relationships between GD and level of sport and
Mage = 19.95 (SD = 1.20, exercise after adjusting for confounders.
range N/A) Those who never played sports were 3.47 times more
likely to report GD compared to those who played
spots almost every day. Playing sports few times a
Curr Addict Rep
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points
Curr Addict Rep

day compared to almost every day increased the odds

of GD by 2.18.
Jeong et al. South N/A Adolescents Young Internet Parents’ excessive interference Poor communication with parents (β = − .156), parent’s
[46] Korea T1 = 0 years N = 986 Addiction Scale Communication with parents excessive interference (β = .317), and less teacher
T4 = 4 years 49.3% male; 50.7% (adapted for gaming) Friends’ support and teacher support support (β = − .187) at T1 predicted academic stress
female (20-item) Academic stress at T2. In turn, academic stress at T2 predicted poorer
Mage = 13.3 (SD = N/A, Self-control self-control (β = − .702) and more daily hours gaming
range N/A) Video game play hours (β = − .199) at T3, and GD at T4 (β = .206).
Finally, worse self-control (β = − .408) and more
daily gaming hours (β = .239) at T3 predicted GD at
T4.
Self-control and daily gaming partially mediated the
relationship between academic stress and GD.
Kaptsis et al. Australia June/September 2015 Adults Internet Gaming Positive and negative affect Abstinence for those with GD compared to those
[47] (online) T0 = Thursday N = 24 Disorder Scale-Short Depression, anxiety and stress without GD resulted in withdrawal symptoms
T1 = Friday 12:00 83.3% male; 16.7% Form 9 (9-item) Craving and withdrawal including negative affect, depression, anxiety, and
T2 = Saturday 12:00 female Video game play stress. The decline in withdrawal symptoms was most
T3 = Sunday 12:00 Mage = 24.6 (SD = 5.8, pronounced in the first 24 h of abstinence.
T4 = Monday 12:00 range 18–36)
King et al. [48] Australia January 2010–August Adults T1: Video gaming Depression, anxiety, stress Among all participants (including those reporting and
2011 T1: N = 360 status: Do you have a Video game play not reporting problems gaming), there was a reported
T1 = 0 months T3: N = 117 problem gaming? decrease in GD symptoms over an 18-month period.
T2 = 6 months 92% male; (Y/N) For the GD group, GD scores went from 48.7 to
T3 = 18 months 8% female T2/T3: Problematic 44.0 at 6 months, to 41.0 at 18 months,
Mage = 24 (SD = 5.5, Video Game Playing
range N/A) Test (20-item)
King et al. [49] Australia June 2015–July 2015 Adults Internet Gaming 13 open-ended questions were asked After the period of gaming abstinence for those with
(online) T0 = Thursday N = 24 Disorder Scale-Short about craving related to video GD, they noted that they “stay addicted” because of
T1 = Friday 12:00 83.3% male; 16.7% Form 9 (9-item) game play internal (boredom, stress, unsettled, excitement) and
T2 = Saturday 12:00 female The fifth survey included two items external cues (devices, apps, events, updates, emails),
T3 = Sunday 12:00 Mage = 24.6 (SD = 5.8, eliciting subjective appraisal beliefs and assumptions (value, prioritizing,
T4 = Monday 12:00 range 18–36) (positive of negative) of preoccupation), negative affect related to abstinence
T5 = 7 days abstinence. (boredom, lack of purpose, social isolation), poor
T6 = 28 days The two final surveys included two coping strategies, and procrastination. Abstinence in
questions about the effect of the some led to engagement in other activities that were
study perceived as beneficial (e.g., time with partner, new
hobby).
Koning et al. Netherlands 2015–2016 Adolescents Internet Gaming Social media disorder More frequent parent–adolescent communication about
[50] T1 = 2015 T1: N = 544 Disorder Scale Internet-specific rules the Internet at T1 predicted more GD at T2
T2 = 2016 T2: N = 352 (9-item) Reactive restrictions by parents: (β = 0.26). This was particularly pronounced among
48.9% male; 51.1% Frequency of communication about boys (β = 0.64).
female gaming
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points

Mage = 13.9 (SD = 0.74, Quality of communication about GD symptoms at T1 predicted more reactive rules
range 11–15) gaming (β = 0.22) and lower quality communication with
parents (β = − 0.13).
In boys, GD symptoms at T1 predicted more frequent
communication about the Internet at T2 (β = 0.16)
In girls, GD symptoms at T1 predicted more reactive
results (β = .30) and lower quality of communication
about internet at T2 (β = − .24).
Konkoly Canada 2006–2011 Adults Behavioral Addiction Help seeking and from whom Generally see greater GD among males compared to
Thege et al. T1 = 2006/7 N = 4121 Measure (adapted for Substance abuse/dependence females at all time points and a moderately strong
[51] T2 = 2007/8 45.3% male; 54.7% gaming; 21-item) Problem gambling decrease in symptom severity over time
T3 = 2008/9 female (F(4) = 26.94, p < .001, n2 = 0.29) for both males and
T4 = 2010/11 Mage = 46.1 (SD = 14.1, females.
T5 = 2011/12 range N/A) Help seeking was very low in the case of excessive
video gaming, with rates under 16% of those with
problems at all time points. Rates of help-seeking
from professions were at 8% among those with GD.
Krossbakken Norway 2012–2014 Adolescents Game Addiction Scale Depression and anxiety Physical aggression was an antecedent to GD (p < .05).
et al. [52] T1 = 2012 T1: N = 2059 (7-item short form) Loneliness Anxiety was a consequence of GD (p < .05).
T2 = 2013 T2: N = 1334 Physical and Verbal aggression Depression and loneliness were both reciprocally
T3 = 2014 T3: N = 1277 Alcohol use disorder associated with GD (p < .05).
38.3% male; 61.7% Investigation of the three typologies (addicted, problem,
female engaged) of gamers identified physical aggression as
Mage = 17.5 (SD = N/A, an antecedent to all typologies. High alcohol
range N/A) consumption was an antecedent to addicted gamers,
and low alcohol consumption was found to be an
antecedent to problem gamers.
Anxiety was a consequence of addicted gamers. Verbal
aggression was a consequence of engaged and
problem gaming problems after 1 year. This was not
consistent over the three waves.
The estimated stability of video game addiction was
35%. For all typologies of gamers, except addicted
gamers, remaining in the same category over 2 years
had a higher probability than changing typology. 53%
of addicted gamers became problem gamers; 0% went
from addicted to engaged; 2% went from engaged to
addicted.
Lee et al. [53] South N/A Emerging adults Young’s Internet Depression Individuals with IGD (n = 18) had higher scores on the
Korea T1 = 0 years N = 36 Addiction Test & Anxiety BDI, BAI, and BIS evaluation tests compared to
T2 = 1.8 years 100% male Internet Gaming Alcohol use controls (n = 18) during both initial and follow-up
(SD = 6.7 months) Impulsivity assessments.
Curr Addict Rep
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points
Curr Addict Rep

Mage = 23.8 (SD = 2.0, Disorder Interview There were no differences in gray matter volume in
range N/A) (IGDI) those with GD across time and between the GD group
and non-GD group.
Those with GD had significantly weaker functional
connectivity between the ACC and frontal regions
(superior/middle frontal gyri) than controls.
Correlation analyses within the GD group showed a
positive correlation between the changes in the dorsal
putamen-middle occipital gyrus functional
connectivity and time spent gaming each day (r = .56,
p = .023).
Lemmens et al. Netherlands 2008–2009 Adolescents Game Addiction Scale Physical aggression Aggression at T1 did not significantly predict GD at T2
[54] T1 = December 2008 T1: N = 1024 (7-item short form) Video game play (p = .22).
T2 = June 2009 T2: N = 540 Violent game play Greater GD symptoms regardless of violent content in
70% male; 30% female games predicted an increase in physical aggression
Mage = 13.9 (SD = 1.4, (β = .10, p < .05) even when controlling for T1
range 11–17) aggression.
This was the case among boys (higher GD predicted
physical aggression (β = .13, p = .02). In girls, GD
led to decrease in physical aggression (β = − .24,
p < .01)
Lemmens et al. Netherlands 2008–2009 Adolescents Game Addiction Scale Loneliness Social competence (β = − .15), self-esteem (β = − .10)
[55] T1 = December 2008 T1: N = 1024 (7-item short form) Life satisfaction and loneliness (β = .12) predicted GD 6 months later.
T2 = June 2009 T2: N = 540 Social competence Loneliness was also a consequence of GD (β = .12),
70% male; 30% female Self-esteem but self-esteem and social competence were
Mage = 13.9 (SD = 1.4, Video game play non-significant.
range 11–17)
Li et al. [56] China N/A Children, adolescents Korean Internet Observations of parent-child Coerciveness (i.e., control dimension of parental
T1 = 0 months N = 241 family dyads Addiction Scale for interactions behavior) at T1 was a temporal risk factor for
T2 = 12 months 57% male; 43% female Adolescents adapted Violent video game play symptoms of GD at T2 (β = 1.62; 3% variance
Mage = 12.09 (SD = 1.41, for Gaming Disorder explained).
range 8–15) (15-item)
Liau et al. [57] Singapore 2007–2009 Children, adolescents Pathological Gaming Personal strengths Personal strengths (emotional awareness [β = − .20],
T1 = 2007 T1: N = 2998 Scale (10-item) Emotional awareness emotion regulation [β = − .25], goal setting
T2 = 2008 T2: N = 2605 Emotion regulation [β = − .22], social competence [β = − .29]) and
T3 = 2009 T3: N = 2532 Goal setting familial factors (parent-child connectedness
72.7% male; 27.3% Empathy [β = − .19], and warm family environment
female Competence [β = − .28]) were protective factors for later GD.
Mage = 11.2 (SD = 2.06, Parent-child connectedness Empathy was non-significant.
range N/A) Parental involvement in media Increases in levels of emotional regulation (β = − .20)
Family environment and family environment warmth (β = − .25) were
Depression related to decreases in GD.
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points

Higher initial levels and increases in GD were related to

higher levels of later depressive symptoms,
controlling for earlier levels of depressive symptoms
(β = .39–.46).
Liew et al. [58] Australia May 2016–July 2016 Emerging adults Internet Gaming Proto-Self-Presence (PSP) PSP at T1 predicted GD symptoms at T2 (β = .76). The
T1 = 0 months T1: N = 61 Disorder Scale-Short Physical Activity full model accounted for 37% of variance in GD with
T2 = 1 month T2: N = 56 Form 9 (9-item) PSP accounting for 36% of the variance. For each
T2 = 2 months T3: N = 43 point increase of PSP, GD behaviors increased by
73.8% male; 26.2% 0.76 (β = 0.76, p < .001)
female Physical activity buffered the relationship between PSP
Mage = 22.53 (SD = 3.04, and GD (β = − .10). The full model accounted for
range 18–29) 41% of the variance of GD. There was a ceiling level
of 144.92 min of physical activity.
Liu et al. [27] China N/A Children Pathological Gaming Autistic traits Autistic traits at T1 predicted GD at T4 (β = 0.09; in the
T1 = 0 months T1: N = 520 Scale (11-item) Emotion regulation mediation model)
T2 = 6 months T2: N = 468 School connectedness After controlling for age, sex, and sensation seeking,
T3 = 12 months T3: N = 440 results showed that autistic traits (T1) were related to
T4 = 18 months T4: N = 420 decreased emotion regulation at T2 (β = − .20) which
52.4% male; 47.6% in turn was related to lower school connectedness at
female T3 (β = .18) which was related to increased GD at T4
Mage = 9.74 (SD = 0.45, (β = − .17). Overall, there is a partial mediation
range N/A) between autistic traits and GD.
Liu et al. [59] China 2011–2014 Adolescents, emerging Chinese Internet Depression There was a reciprocal interrelationship between
T1 = 2011 adults Addiction scale depression and GD symptoms; however, depression
T2 = 2012 N = 563 modified for GD was a stronger predictor of GD. This pattern was
T3 = 2013 78.0% male; 22.0% (26-item) stable across all 4 years. Detailed results are as
T4 = 2014 female follows:
Mage = 18.31 (SD = 0.89, T1 depression predicted T2 GD (β = .118);
range 16–21) T2 depression predicted T3 GD (β = .126);
T3 depression predicted T4 GD (β = .127).
T1 GD predicted T2 depression (β = .07);
T2 GD predicted T3 depression (β = .066);
T3 GD predicted T4 depression (β = .07).
Lukavska [60] Czech 2012–2015 Adolescents, adults Core Addiction Scale Time perspective A reduced level of future positive time perspective at T1
Republic T1 = 2012 N = 48 (7-items) Video game play predicted GD symptoms immediately and 3 years
T2 = 2015 90.8% male; 9.2% female later (β = − .28). Negative time perspective was
T1: Mage = 24.1 non-significant.
(SD = 6.8, range
12–67)
T2: Mage = 22.4
(SD = 4.7, range
13–37)
Curr Addict Rep
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points
Curr Addict Rep

Marmet et al. Switzerland August Emerging adults Game Addiction Scale Attention deficit hyperactivity Individuals with both GD and ADHD at T1 had worse
[61] 2010–March 2018 T1: N = 5987 (7-item short form) disorder (ADHD) outcomes at T2 (higher depression, lower mental
T1 = 2010–11 T2: N = 5067 Alcohol use disorder health, lower life satisfaction, poorer performance at
T2 = 2016–2018 100% male Cannabis use disorder work/school).
Mage = 20 (SD = N/A, Tobacco use disorder ADHD at age 20 increased the risk for GD at age 25
range N/A) Depression (OR = 1.72).GD was greatest at T2 in those with GD
Mental health and ADHD at T1 (32.3%), followed by GD only
Life satisfaction (20.4%) and ADHD only at T1 (8%).
Poor performance at work/school ADHD at T1 increased the odds of GD at T2
(AOR = 1.81); GD at age 20 also increased the risk
for ADHD at T2 (AOR 1.47).GD at T1 was
associated with new onset of ADHD at T2–9.1%
compared to 5.7% (OR = 1.63).ADHD was not more
persistent at T2 among people with GD and ADHD at
T1 compared to those with ADHD only at T1.
The combination of ADHD and GD at T2 was most
frequent (10.8%) among those who already had
ADHD and GD in T1, but this combination
persistence was not very high (10.8%).
Only the inattention subscale depicted a bidirectional
longitudinal relationship with GD scores: from
inattention at age 20 to GD at age 25 (β = .09); and
from GD at age 20 to inattention at age 25 (β = .044).
Molde et al. Norway 2013–2015 Adolescents, adults Game Addiction Scale Problem gambling GD scores at T1 predicted gambling problems at T2
[62] T1 = 2013 N = 4601 (7-item short form) (β = .15), while the reverse relationship was
T2 = 2015 47.2% male; 52.8% non-significant.
female
Mage = 48 (SD = 15.1,
range 16–74)
Mößle & Germany 2008–2009 Children, adolescents Video Game Addiction Media ownership Playing video games in reaction to real world failures
Rehbein T1 = November 2005 N = 739 Scale (CSAS) Academic self-concept predicted GD 1 year later (β = .18). Low academic
[63] T2 = May 2006 50.6% male; 49.4% (11-item) ADHD symptoms self-concept also predicted GD 1 year later (β = .14).
T3 = May 2007 female Overall wellbeing Hyperactivity and depression were non-significant
T4 = May 2008 Mage = 11.5 (SD = N/A, Depression predictors of GD.
T5 = May 2009 range N/A) Peer problems Peer problems also predicted greater GD 1 year later
Family violence (β = .08). Little parental care and family violence was
Parental care non-significant.
Video game genre There was an indirect effect of low parental care on GD
via a lower academic self-concept (β = .33) and
playing games in reaction to real world failures
(β = .13). The model explained 35% of the variance
of GD at T5.
Netherlands Adolescents Attention problems
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points

Peeters et al. February T1: N = 544 Internet Gaming Social vulnerability Social vulnerability (β = .297) and attention problems
[64] 2015–February T2: N = 354 Disorder Scale Life satisfaction (β = .298) at T1 predicted GD symptoms at T2 while
2016 48.9% male; 51.1% (9-item) controlling for gender. Attention problems had the
T1 = 0 months female strongest association with GD in those who were also
T2 = 12 months Mage = 13.9 (SD = 0.74, socially vulnerable (β = .681) and less satisfied with
range 11–15) their life (β = .485).
Peeters et al. Netherlands 2016–2017 Adolescents Internet Gaming Social competence Male problematic gamers scored significantly higher on
[65] T1 = 0 years N = 1928 Disorder Scale Life satisfaction impulsivity and hyperactivity, and lower on social
T2 = 1 year 57% male; 43% female (9-item) Attention, hyperactivity, impulsivity competence compared to both recreational gamers
Mage = 13.3 (SD = 0.91, Video game play and engaged gamers at T1 (all p < .05). At T2,
range N/A) impulsivity scores were higher and social competence
scores were lower as gaming became more
problematic (all p < .05).
Latent class analyses revealed three classes for boys
(recreational, engaged, and problematic) and two
classes for girls (recreational and engaged). Only
three of the 26 problematic gamers at T1 were
problem gamers at T2. However, 38.5% of problem
gamers dropped out of the study.
Rehbein & Germany 2005–2010 Children, adolescents Video Game Addiction Parental devotion Being male (β = .34), single parent families (β = .10),
Baier [66] T1 = 0 years N = 406 Scale (CSAS) Parental situation lower school wellbeing (β = − .10) and weaker social
T2 = 5 years 45.3% male; 54.7% (11-item) Media accessibility in bedroom integration in the classroom (β = − .10) at T1
female Parental supervision of media predicted GD at T2. This model explained 21% of
T1: Mage = 9.7 Self-concept of school performance variance in GD.
(SD = 0.63, range N/A) Academic performance
T2: Mage = 15 (SD = 0.58, Social integration in class
range N/A) Social exclusion
Attachment to school
School related wellbeing
Video game play
Rothmund Germany N/A Adolescents Pathological Gaming Video game play The stability of GD was 27% over a 1-year period
et al. [67] T1 = 0 years N = 488 Scale (11-item) School performance (among 2.8% of the sample). There was a greater
T2 = 1 year 47% male; 46% female Escapism motives noted stability for GD despite poor school
Mage = 14.13 (SD = 1.07, performance (r = .49) and escapist motives (r = .46)
range 12–17)
Scharkow et al. Germany 2011–2013 Adolescents, adults Game Addition Scale Social capital and support The stability of GD was 26% over a 2-year period. High
[68] T1 = 2011 T1: N = 4500 (7-item short form) Life satisfaction GD scores seemed to fluctuate over time, whereas
T2 = 2012 T2: N = 2190 Life success non-problem gaming scores did not. There was no
T3 = 2013 T3: N = 902 Video game play evidence for the relationship between GD and social
56% male; 44% female capital/support over time.
Mage = 37.7 ± (SD = N/A,
range N/A)
USA N/A Emerging adults Expected college engagement
Curr Addict Rep
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points
Curr Addict Rep

Schmitt & T1 = Fall (week before N = 477 Pathological Gaming Academic performance (GPA) GD symptoms predicted a lower first year college GPA
Livingston class) 100% male Scale (10-item) Drug and alcohol violations on when controlling for high-school GPA (β = − 0.09).
[69] T2 = Spring (end of Mage = 18 (SD = N/A, campus GD was also a significant negative predictor of
semester) range N/A) Video game play drug/alcohol violations when controlling for GPA
(β = − .39).
Schoenmacker Netherlands 2003/6–2007/10 Adolescents, emerging Pathological Gaming Attention-deficit hyperactivity More severe ADHD-inattentive symptoms at T1 led to
et al. [70] T1 = 0 years adults Questionnaire based disorder more severe GD at T2 even after controlling for sex.
T2 = 4.4 years N = 362 on the Game Conduct problems Conduct problems were not identified as a significant
(SD = 0.71) 80.7% male; 19.3% Addiction Scale Alcohol use mediator of this relationship.
female (24-item) Nicotine dependence
Mage = 15.93 (SD = 2.46, Other drug habits
range 12–24)
Stavropoulos Australia June 2016–September Emerging adults Internet Gaming Presence Presence at T1 failed to predict the GD-slope across
et al. [71] 2016 T1: N = 61 Disorder Scale-Short time-points. However, presence was significantly
T1 = June 2016 T2: N = 56 Form 9 (9-item) associated with the GD intercept (β = .369) meaning
T3 = September 2016 T3: N = 43 that presence scores were significantly associated
77% male; 23% female with initial GD scores, but was not predictive of
Mage = 23.02 (SD = 3.42, change over time.
range 18–29)
Su et al. [72] China 2011–2012 Adolescents Internet Gaming Peer victimization Peer victimization at T1 predicted deviant peer
T1 = 0 years T1: N = 386 Addiction Scale Deviant peer affiliation affiliation at T2 (β = .22), which in turn predicted
T2 = 1 year T2: N = 351 (16-item) Normative beliefs about aggression normative beliefs regarding aggression at T3
T3 = 2 years T3: N = 323 (β = .21), and ultimately, greater GD symptoms at T3
47% male; 53% female (β = .27).
Mage = 14.83 (SD = 0.49,
range 13–16)
Su et al. [73] China 2012–2013 Children, adolescents Young Internet Parental monitoring The cross-lagged model revealed that parental
T1 = October 2012 T1: N = 1830 Addiction Scale Parent-child relationship monitoring at T1 predicted lower GD symptoms at T2
T2 = April 2013 T2: N = 1680 (adapted for gaming) (β = − .069).
T3 = October 2013 T3: N = 1490 (8-item) Greater GD symptoms at T2 predicted lower parental
54.6% male; 45.4% monitoring at T3 (β = − 0.048).
female GD at T1 predicted poorer father-child relationship at T2
Mage = 12.03 (SD = 1.59, (β = − .101). This in turn increased parental
range 10–15) monitoring at T3 (β = 0.075). GD at T1 also
predicted poorer mother-child relationship at T2 (B
-.096).
The father–child relationship had a reciprocal, indirect
effect on the relationship between parental monitoring
and Internet gaming disorder (β = 0.005), while the
mother–child relationship did not.
Vadlin et al. Sweden 2012–2015 Adolescents Gaming Addiction Problem gambling GD was relatively stable over time (γ = 0.739, p ≤ .001),
[74] T1 = 2012 N = 1576 Identification Test Gambling activities with 46.1% still reporting GD symptoms after
T2 = 2015 42% male; 58% female (15-item) Video game play
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points

Mage = N/A (SD = N/A, 3 years. Of those without GD at T1 (88.6%), 11.4%

range 13–15) developed GD at T2.
GD at the first wave predicted problem gambling
3 years later (OR = 1.89) and explained 3.7% of the
variance when accounting for covariates.
van den Netherlands 2015–2017 Adolescents Internet Gaming Social Media use GD symptoms at T1 predicted lower life satisfaction at
Eijnden T1 = 2015 N = 538 Disorder Scale Social Media Disorder symptoms T2 (β = − 0.17) but was non-significant at T3.
et al. [75] T2 = 2016 48.9% male; 51.1% (9-item) Perceived social competence GD symptoms at T1 predicted lower perceived social
T3 = 2017 female Life Satisfaction competence at both T2 (β = − .22) and T3 (β = − .29).
Mage = 12.9 (SD = 0.73, Academic performance Both effects were only significant for boys.
range 12–15) Video game play
van Rooij et al. Netherlands 2008–2009 Adolescents Compulsive Internet Use Self-esteem Of the 3% reporting GD at T1, 50% remained GD at T2.
[76] T1 = 2008 T1: N = 1572 Scale (14-item) Loneliness Those with GD had a significantly higher depressed
T2 = 2009 T2: N = 1476 Depression mood at T2 (p = .04), loneliness at T1 (p = .001) and
81% male; Anxiety T2 (p = .06), and negative self-esteem at T2
19% female Video game play (p = .001). At T2, those with GD were more
Mage = N/A (SD = N/A, depressed than the heavy but non-disordered gamers.
range 13–16)
Wartberg et al. Germany 2016–2017 Adolescents Internet Gaming Mental health The stability of GD over a 1-year period was 14.4%.
[77] T1 = 2016 T1: N = 1095 Disorder Scale-Short Anti-social behavior Self-esteem (β = .07–.08) and higher levels of
T2 = 2017 T2: N = 985 family dyads Form 9 (9-item) Anger control problems hyperactivity/inattention (β = .14–.13) at T1
50.7% male; 49.3% Anxiety and depression predicted GD at T2. This model explained 23% of the
female Self-esteem variance for females and 27% for males.
Mage = 12.99 (SD = 0.82, Attention deficit hyperactivity GD at T1 was a predictor of emotional distress at T2
range 12–14) disorder (β = 0.06–0.08), explaining 17% of the variance for
Parental depression females and 18% for males.
Parental anxiety
Weinstein USA October Adults Internet Gaming Social, mental, physical health Psychological needs satisfaction promoted health
et al. [78] 2015–March 2016 N = 2316 Disorder Scale-Short Psychological need satisfaction (β = .07) and served as a protective factor against GD
T1 = 2015 38.2% male; 61.8% Form 9 (9-item) Physical and social activity (β = − .08).
T2 = 2016 female Video game play GD at T1 predicted less satisfaction of basic
Mage = 49.21 (SD = 1.32, psychological needs at T2 (β = − .06).
range 19–91) D was indirectly related to health by its impact on basic
psychological needs (β = − .004). There was also a
significant mediation by autonomy, competence and
relatedness (all β = − .003)
GD criteria were moderately stable and GD symptoms
and basic psychological need satisfaction had a
reciprocal relationship over time (β = .56).
Wichstrom Norway 2013–2015 Children Internet Gaming Attention-deficit hyperactivity Low interest in previous hobbies, gaming to
et al. [79] T1 = 2013 T1: N = 995 Disorder Interview disorder (ADHD) escape/relieve negative mood, and lying to family
T2 = 2015 T2: N = 740 (IGDI) Oppositional defiant disorder (ODD) members at T1 predicted GD at T2.
Conduct disorder (CD)
Curr Addict Rep
Table 2 (continued)

Authors Country Duration and time Sample characteristics GD measure Other measures Main findings
points
Curr Addict Rep

48.6% male; 51.4% Anxiety disorders In the multivariate model, limited social and emotion
female Depression regulation skills (β = − 0.01) at T1 predicted more
T1: Mage = 8.0 Temperament GD symptoms at T2.
(SD = N/A, range N/A) Intelligence Correlations of symptoms with GD include: ADHD:
Executive functions r = .08; ODD: r = .15; anxiety disorders: r = 0.13;
Self-concept & self-esteem depression: r = 0.12; and CD r = 0.07 (ns).
Social skills
Victimization
Emotion regulation
Family climate
Parental functioning
Physical activity
Yu et al. [80] China 2011–2013 Adolescents Pathological Gaming Teacher autonomy support (7th Teacher autonomy support at T1 increased basic
T1 = Fall of grade 7 N = 356 Scale (11-item) grade) psychological needs satisfaction at T2 (β = 0.25),
T2 = Fall of grade 8 41% male; Psychological needs satisfaction (8th which in turn increased school engagement at T3
T3 = Fall of grade 9 59% female grade) (β = 0.35), which ultimately decreased GD at T3
Mage = 14.83 (SD = 0.49, School engagement (9th grade) (β = − .23; indirect β = − .020).
range 13–16) Furthermore, teacher autonomy support at T1 directly
increased T3 school engagement (β = 0.21), which in
turn decreased GD at T3 (indirect β = − .048).
Zhang et al. China N/A Emerging adults DSM-5 criteria for Purpose in life The prevalence of probable GD at T1 and T2 was 14.8%
[81] T1 = 0 years T1: N = 469 Internet Gaming Social support and 9.9%, respectively.
T2 = 1 year T2: N = 283 Disorder (9-item; 0–1 Results indicate that T1 purpose in life (β = − .173), but
39.6% male; 60.4% response) not social support predicted fewer GD symptoms at
female T2 (p < .001). In addition, social support and purpose
Mage = 20.47 (SD = 1.15, in life predicted one another across time (β = .099;
range 18–27) .134, SS and PIL, respectively).
Zhu et al. [82] China 2012–2013 Adolescents Young Internet Parent-adolescent relationship School connectedness and deviant peer affiliation at T2
T1 = October 2012 N = 833 Addiction Scale School connectedness fully mediated the association between the
T2 = April 2013 54.9% male; 45.1% (adapted for gaming) Deviant peer affiliation parent–adolescent relationship at T1 and GD at T3 (all
T3 = October 2013 female (8-item) indirect paths β = − .03).
Mage = N/A (SD = N/A, There was a complete mediation whereby low-quality
range 11–14) parental relationships at T1 predicted low school
connectedness at T2 (β = − .27), which in turn
predicted deviant peer affiliation (β = − .12) at T3,
and GD at T3 (β = .19; overall sequential path
β = − .01).

Note. N/A not available, GD gaming disorder

 Curr Addict Rep

measurements (at baseline and 1 year later; n = 18), 2-year a control variable, with some studies assessing type or genre
time range with three measurements (at baseline, 1 year and of video game played. A count overview for the most fre-
2 years later; n = 7), and 6-month time range with two mea- quently measured variables is presented in Fig. 2.
surements (at baseline and 6 months later; n = 4). The remain-
ing 28 studies had diverse time ranging from 3.5 days to Central Findings
5 years, with anywhere between two and five measurement
points. Temporal Stability of Gaming Disorder Evidence for the tem-
poral stability of GD varied significantly between studies de-
Measures of Gaming Disorder pending on the age of participants and follow-up periods.
Among adolescents, the temporal stability of GD was reported
Twenty-four different measurement tools were included to as being 20% [65], 27% [67], and 50% [76] over a 1-year
measure GD across the 57 included articles. A description of period; 35% [52] and 84% [41] over a 2-year period; and
each of the measurement tools is provided in Table 3. The 46% over a 3-year period [74]. In adults, GD had greater
most frequently used measure for GD was the Game levels of fluctuation and higher rates of spontaneous remission
Addiction Scale (GAS) [83], followed by the Pathological over time, indicating that the temporal stability of GD may be
Gaming Scale (PGS) [84] and the Internet Gaming Disorder different for adolescents and adults. Of note, Scharkow and
Scale-Short Form 9 (IGDS-SF-9) [86]. Six studies utilized the colleagues [68] found that 26% of adults with GD at baseline
DSM-5 criteria for Internet gaming disorder (labeled as either still met the diagnostic criteria of GD 2 years later.
the DSM-5 criteria for Internet gaming disorder [2], the Furthermore, King, Delfabbro, and Griffiths [48] report that
Internet Gaming Disorder Scale [87], or Gaming Addiction over a period of 18 months, all adult participants meeting the
Identification Test [88]) and four studies utilizing a semi- diagnostic criteria for GD experienced a decrease in symp-
structured clinical interview to assess symptoms of Internet toms over time. Similar findings are reported by Konkoly
gaming disorder based on the DSM-5 criteria. Thege and colleagues [51], whereby a moderately strong de-
Nine studies utilized scales based on measures of Internet crease in GD symptom severity was reported over a 5-year
addiction that were adapted and refined to assess GD. This period, with no participant meeting the diagnostic threshold
includes the Video Game Addiction Scale [89, 90]; Internet for GD consistently across all five measurements.
Addiction Scale (IAS) [91]; Chinese Internet Addiction Scale
[92]; Korean Internet Addiction Scale [93]; Compulsive Antecedents to Gaming Disorder Thirty-three longitudinal
Internet Use Scale [94]; and Problematic Internet Use Scale studies have looked at the role of various biological, cognitive,
[95]. Furthermore, two studies utilized measures developed to psychological, and social factors as risk or protective factors
assess behavioral addictions which were adapted to reflect for the development of GD. Only one study by Coyne and
video gaming including the Behavioral Addiction Measure colleagues [35] investigated biological factors as a risk for GD
[96] and the Core Addiction Scale [97]. The remaining studies among adolescents, and they reported that a lack of physio-
utilized measures or items that were developed specifically for logical stimulation in response to a cognitively stimulating
the study in question. These include the Internet Gaming and family problem-solving task predicted GD symptoms
Addiction Scale [72]; Smartphone Game Addiction Screen 1 year later in the entire sample, and increased physiological
[33]; Pathological Gaming Scale [39]; and a single Video stress to the family problem-solving task predicted greater GD
Game Addiction question [43]. symptoms among females.
Regarding psychological factors, four studies reported that
Other Measured Variables emotion dysregulation at baseline predicted greater symptoms
of GD 1 year later in adolescents [35], and 2 years later in
Studies were compared based on the variables they measured children [41, 57, 79]. In three studies, attention problems were
as predictors and outcomes or as being reciprocally associated related to greater GD symptoms among adolescents 1 year later
with GD. Four studies focused primarily on biological and [39, 77], especially for those who reported greater difficulties
psychological factors; 19 studies measured psychological fac- establishing and retaining close friendships [64]. Moreover,
tors, six studies measured social factors, 26 studies measured inattentive symptoms predicted GD severity more than 4 years
both psychological and social factors, and two studies includ- later among adolescents and young adults [70]. Mixed results
ed biological, psychological, and social factors. All studies were reported regarding the predictive role of hyperactivity [63,
accounted for demographic variables including age and gen- 65, 70, 77] and empathy [41, 57] in children and adolescents.
der, while others measured socio-economic status, ethnicity, Other psychological risk factors for GD among children and
level of education, relationship status, parental status, and adolescents included greater impulsivity [41, 65], poor self-
community of residence. Furthermore, most studies measured esteem [35, 55, 77], higher levels of physical aggression [52],
frequency of video game play either as a predictor of GD or as and autistic traits (with emotion dysregulation and poor school
Table 3 Measurement tools used to assess problem or disordered gaming

Name of measure Number of articles Available versions Scale type; interview questions Content of measure
utilizing the
measure
Curr Addict Rep

Game Addiction Scale 12 [28, 29, 40, 44, 21-item self-report; 5-point Likert scale Assesses both online and offline gaming over the past 6 months; items assessing seven
(GAS) [83] 45, 52, 54, 55, 7-item self-report; criteria of GD (i.e., salience, tolerance, relapse, withdrawal, mood modification,
61, 62, 68, 70] extended conflict, and problems).
(24-item)
self-report
Pathological Gaming Scale 10 [27, 32, 34, 35, 10-item self-report; Trichotomous response (yes, no, Assesses severity of GD based on an adaptation of the pathological gambling criteria
(PGS) [84] 41, 42, 57, 67, 11-item self-report or sometimes) of the DSM-IV-TR [85]; 11-item scale includes an additional item for skipping
69, 80] household responsibilities to play video games.
Internet Gaming Disorder 8 [14, 30, 47, 49, 9-item self-report 5-point Likert scale Assesses GD based on the nine core criteria of Internet gaming disorder according to
Scale-Short Form 9 58, 71, 77, 78] the DSM-5 over the past 12 months.
(IGDS-SF-9) [86]
DSM-5 criteria for Internet 6 [36–38, 53, 79, 9-item self-report; Scale is dichotomous response (yes or no); Assesses GD based on the nine core criteria of Internet gaming disorder according to
gaming disorder [2] 81] semi-structured interview included primary questions with the DSM-5 over the past 12 months.
clinical interview probes and optional follow-up questions
Internet Gaming Disorder 4 [50, 64, 65, 75] 9-item self-report Dichotomous response (yes or no) Assesses GD based on the nine core criteria of Internet gaming disorder according to
Scale [87] the DSM-5 over the past 12 months.
Gaming Addiction 1 [74] 15-item self-report 5-point Likert scale Assesses symptoms of GD among adolescents over the past 12 months based on core
Identification Test (GAIT) criteria of GD.
[88]
Video Game Addiction Scale 2 [63, 66] 11-item self-report 4-point Likert scale Internet addiction scale adapted to assess GD covering dimensions
(CSAS) [89, 90] including preoccupation/salience, conflict, loss of control, and
withdrawal symptoms.
Internet Addiction Scale (IAS) 4 [46, 48, 73, 82] 8-item self-report; 3-point Likert scale for the 8-item scale; Internet addiction scale adapted to assess GD covering the core criteria for pathological
or Problematic Video Game 20-item self-report 5-point Likert scale for the 20-item scale gambling based on the DSM-IV-TR [85] including preoccupation, loss of control,
Playing Test [91] repeated unsuccessful attempts to reduce behavior, and harmful consequences.
Chinese Internet Addiction 1 [59] 26-item self-report 4-point Likert scale Internet addiction scale adapted to assess GD including five symptom dimensions
Scale [92] including compulsive use, withdrawal, tolerance, problems with interpersonal
relationships, and problems with time management.
Korean Internet Addiction 1 [56] 15-item self-report 5-point Likert scale Internet addiction scale adapted to assess GD including four symptom dimensions
Scale [93] including weakened control over gaming behaviors, prioritization of gaming over
other activities/interests, maintaining or increasing gaming despite adverse
consequences, and major impairments in areas of daily functioning.
Compulsive Internet Use Scale 1 [76] 14-item self-report 5-point Likert scale Internet addiction scale adapted to assess GD among adolescents including core
[94] dimensions typical of behavioral addictions (i.e., withdrawal symptoms, loss of
control, salience, conflict and mood modification).
Problematic Internet Use Scale 1 [31] 14-item self-report 5-point Likert scale Internet addiction scale adapted to assess GD based on five core
[95] behavioral symptoms including tolerance, withdrawal, impulsivity, preoccupation,
and craving.
Behavioral Addiction Measure 1 [51] 21-item self-report Dichotomous response (yes or no) Behavioral addiction scale adapted to reflect video gaming covering three symptom
[96] domains including psychosocial problems caused by the behavior, impaired
control, and other addiction-related characteristics (e.g., craving, preoccupation).
Core Addiction Scale [97] 1 [60] 7-item self-report 5-point Likert scale Behavioral addiction scale adapted to reflect video gaming assessing dimensions
including behavioral salience, interpersonal conflict, conflict with other activities,
withdrawal symptoms, and relapse and reinstatement.
1 [72] 16-item self-report 3-point Likert scale
 Curr Addict Rep

to get enough sleep, gaming interfering with other activities, feeling agitated when
Assesses core symptoms of addiction applied to smartphone gaming including failing
Assesses symptoms of Internet gaming addiction based on the degree of engagement

Assesses symptoms of GD based on core symptoms including thinking about games

Single question to assess GD, asking participants “if they felt they were addicted to
connection partially mediating this relationship) [27]. As for

so often that it interferes with other activities, gaming behaviors interfering with
not playing games, and making unsuccessful attempt to reduce time playing.
cognitive factors, greater perfectionism, cognitive salience,
and regret scores predicted the maintenance of GD symptoms
over a 9-month period among adults [40]. Furthermore, beliefs
that the future will not be positive predicted symptoms of GD
3 years later [60], and reduced perceptions of behavioral con-
trol predicted symptoms of GD 6 months later in adolescents
and young adults [44].

online or offline video games in the past 30 days”.

Motivations and game-related factors were also investigated
with Internet games over the past 6 months.

as potential risk factors for GD. Playing in response to real-

world failures [63], escapism [79], and achievement motives
[31] predicted symptoms of GD both 1 year and 2 years later,
respectively. Furthermore, young adults who highly identified
with their in-game avatar were more likely to develop symptoms
interpersonal relationships.

of GD 3 months later, while greater engagement in physical

activity weakened the strength of this association [58]. Similar
Content of measure

findings regarding identification with the in-game avatar have

been reported among children over a 2-year period [41].
While various studies have investigated the predictive role
of depression on GD, mixed findings have been reported.
Whereas Burleigh and colleagues [30] found that depressive
symptoms predicted symptoms of GD 3 months later among
young adults, and that this relationship was moderated by a
greater attachment to the in-game avatar, Mößle and Rehbein
[63] failed to report a significant relationship between depres-
sive symptoms and GD among children and adolescents.
Number of articles Available versions Scale type; interview questions

Moreover, Dang and colleagues [36] report that while emotion-

al intelligence predicted GD 1 year later in emerging adults, this
Dichotomous response

Dichotomous response

Dichotomous response

relationship was fully mediated by depressive symptoms and

coping flexibility. In relation to depression, Zhang and col-
leagues [81] found that a greater purpose in life was related to
(yes or no)

(yes or no)

(yes or no)

decreased GD symptomatology 1 year later. Guillot and col-

leagues [43] reported that higher levels of trait anhedonia (i.e.,
difficulty experiencing pleasure) predicted symptoms of GD
18 months later among at-risk emerging adults.
4-item self-report

7-item self-report

1-item self-report

As for social factors, most studies focused on children

and adolescents, with a single-study finding that lower
levels of perceived success in school predicted symptoms
of GD 2 years later in a sample of adolescents and adults
[68]. With regard to familial factors, greater parent-child
connection and closeness [34, 57], a warm family environ-
ment [57], and having a two-parent family [66] were iden-
utilizing the

tified as protective of GD among children and adolescents.

measure

1 [33]

1 [39]

1 [43]

Furthermore, whereas parental coerciveness predicted

symptoms of GD 1 year later [56], parental restrictions of
video gaming behavior had no effect on symptoms of GD
Smartphone Game Addiction

Video Game Addiction [43]

Pathological Gaming Scale
Internet Gaming Addiction

[34]. Gentile and colleagues [42] reported that having video

Table 3 (continued)

game devices in the bedroom of children predicted symp-

Name of measure

toms of GD 2 years later, this relationship being explained

by a greater amount of time playing video games. Lastly, a
Screen [33]
Scale [72]

more balanced family cohesion buffered the relationship

[39]

between anxious symptoms and GD over a 3-month period

among emerging adults [71].
Curr Addict Rep

Fig. 2 Measures included in Measures in Longitudinal Studies of Gaming Disorder

longitudinal studies of gaming 18
disorder. Note. ADHD: attention-
deficit hyperactivity disorder; 16
AUD: alcohol use disorder; SUD:
substance use disorder 14

12

10

Studies also indicated significant interactions between famil- and colleagues [47] reported that although these individuals
ial, peer, and academic factors in predicting GD. In a serial experienced negative affect and psychological distress when
mediation model, poor parent-adolescent relationships were abstaining from gaming, these symptoms largely declined
predictive of reduced feelings of school connectedness 6 months within the first 24 h of abstinence. Elaborating on this study,
later, leading to greater affiliation with deviant peers and symp- King and colleagues [49] conducted a qualitative longitudinal
toms of GD at 12 months [82]. Moreover, Jeong and colleagues follow-up 7 and 28 days following the cessation of gaming
[46] reported that positive parental communication and teacher and identified various factors that maintained GD among these
support predicted reductions in academic stress, which was participants. These factors included internal and external cues,
associated with higher levels of self-control and fewer symp- beliefs and assumptions regarding gaming, abstinence-related
toms of GD. Other academic risk factors for GD included a negative affect, and poor coping strategies.
lower academic self-concept [63], poor well-being in school Both cognitive and psychological consequences have been
[66], poor social integration in the classroom [66], and control- reported as outcomes of GD. Negative psychological conse-
ling teaching styles [80]. Additionally, peer problems [63] and quences included anxiety [41, 52], depression [28, 29, 41, 57,
poorer social competences [41, 55, 57] were predictive of GD. 76], loneliness [55, 76], emotional distress [77], and poor self-
In a serial mediation model, Su and colleagues [72] report that esteem [76]. With regard to problems with smartphone gam-
peer victimization predicted symptoms of GD 2 years later, ing, Chen and colleagues [33] reported that among problem
which was explained by greater affiliation with delinquent gamers, thought suppression resulted in an increase in intru-
peers and an increase in normative beliefs regarding aggression. sive thoughts and impulsive playing, which led to more per-
ceived difficulties in quitting, reductions in self-efficacy, and
Outcomes of Gaming Disorder Twenty-four longitudinal stud- an increased amount of time playing games.
ies have focused on the consequences or outcomes associated As for social constructs, life satisfaction [74, 75], percep-
with GD. In studies investigating neurological changes in activ- tions of success [74], and social competence [75] have been
ity and connectivity among young adults with and without GD, reported to decrease 1 year later for those with GD. Among
potential biological changes in the mesolimbic region were not- children and adolescents, decreases in school performance
ed in regular gamers that developed GD over a 1-year period [41] and a poor relationship with parents [32, 41] have been
[37], individuals with GD no longer meeting the diagnostic reported in those with greater symptoms of GD. Decreases in
criteria 1 year later [38], and among individuals with GD that academic performance have also been noted among adoles-
spent the most time gaming over a 22-month period [53]. cents [29] and young adults [69]. Two studies identified asso-
In a study investigating the craving and withdrawal symp- ciations with aggressive and conduct problems, whereby GD
toms of adults meeting the diagnostic criteria for GD, Kaptsis symptoms predicted increases in physical aggression among
 Curr Addict Rep

adolescent boys [54] and increases in serious, aggressive, and Discussion

covert conduct problems in both boys and girls [29]. No as-
sociations were reported in adolescents and young adults with The purpose of this review was twofold: to identify and syn-
GD with regard to heavy episodic drinking [28, 29]. Further, thesize the empirical evidence for the temporal stability, risk
GD was found to be a negative predictor of drug and alcohol factors, and consequences of GD across the lifespan, and to
violations by university students on campus [69]. However, evaluate the four existing theoretical models for the develop-
two studies reported that symptoms of GD were predictive of ment of GD based on existing evidence. Results from the in-
gambling problems 2 and 3 years later among adolescents and cluded studies indicate that the temporal stability of GD ranged
adults [62, 74]. from 20 to 84% depending on the duration of the study and
age-group investigated. Moreover, results indicate that com-
Reciprocal Relationships Nine longitudinal studies identified monly reported psychological risk factors for GD among youth
reciprocal relationships between certain variables and GD. include emotional dysregulation [35, 41, 57, 79], attention
Concerning psychological factors, depression and loneli- problems [43, 70, 77], and negative self-esteem [35, 55, 77],
ness were reciprocally associated with GD among adoles- with some evidence for under-stimulation/boredom [35], ag-
cents over a 6-month [55], 2-year [41, 52], and 3-year pe- gression [52], autistic traits [27], anhedonia [43], and depres-
riod [59]. Furthermore, Marmet and colleagues [61] report- sion [30, 36]. Moreover, commonly reported social risk factors
ed that inattentive symptoms had bidirectional relationships include a negative parent-child relationship [34, 57], poor so-
with symptoms of GD over a 5-year period. Of note, those cial competencies [41, 55, 57], and loneliness [49, 50, 62]. The
with symptoms of both inattention and GD at age 20 report- above-noted psychological risk factors are consistent with the
ed greater depressive symptoms, lower life satisfaction, and internal factors considered in Paulus and colleagues’ [13] inte-
poor work/school performance at age 25 [61]. grated model and the predisposing variables of the I-PACE
Psychological needs satisfaction, defined as the need for model [25, 26]. Moreover, Paulus and colleagues [13] also
autonomy, competence, and relatedness, was also reported consider the role of some social or external factors such as
as having a reciprocal relationship with GD over a 1-year negative parent-child interactions in the development of GD,
period among adults [78]. with loneliness being an important motivator and psychosocial
Reciprocal relationships were identified between various risk factor for GD outlined in the I-PACE model [25].
parental factors and GD. Koning and colleagues [50] reported Motivations for engagement in gaming that are temporally
that the more frequently an adolescent communicated with associated with GD include playing in reaction to real-world
their parent about the Internet, the greater GD symptoms they failures [63], escapism [79], and achievement motives [31].
endorsed 1 year later, which was particularly pronounced Additionally, a strong identification with an in-game avatar
among boys. Further, symptoms of GD predicted more reac- was predictive of GD across developmental periods [30, 41,
tive rules on the part of parents and a lower quality of parental 58]. Based on the included theoretical models, Benarous and
communication about the Internet among girls, and more fre- colleagues [22] broadly acknowledge the role of school failures
quent communication about the Internet among boys. As for or social rejection as maintaining factors for GD; Paulus and
parental monitoring, Su and colleagues [73] reported that the colleagues [13] mention game-related factors or rewards as be-
father-child relationship, but not mother-child relationship, ing a risk for GD; and Brand and colleagues [25] make reference
had a reciprocal indirect effect on the relationship between to specific motives and emotional coping. To the authors’
GD symptoms and parental monitoring. Results suggested knowledge, only Lee, Lee, and Choo [23] explicitly mention
that while symptoms of GD predicted a poor-quality father- the role of various specific motives (e.g., social recognition,
child relationship which in turn led to increases in parental competition) in being associated with different vulnerabilities
monitoring, high parental monitoring resulted in a better- for the development of GD. Although not all of these motives
quality father-child relationship and a decrease in GD symp- are supported by longitudinal evidence (e.g., safe social engage-
toms [73]. ment, belongingness, competition), some (i.e., escapism and
Lastly, mixed findings were reported with regard to the achievement) are supported with additional research being nec-
bidirectional relationships between physical activity and GD. essary to identify whether different types of problem gamers
Henchoz and colleagues [45] reported reciprocal relationships have different motives reinforcing their gaming behaviors.
between GD and engagement in sports and exercise, whereby Theoretical models for the development of GD suggest that
symptoms of GD predicted less engagement in sports and underlying vulnerabilities may be responsible for the develop-
exercise, and greater engagement in sports and exercise pre- ment of this disorder over time. This includes the role of emo-
dicted reduced symptoms of GD 15 months later. However, tion dysregulation, antisocial/aggressive tendencies, and im-
Weinstein and colleagues [78] did not identify significant re- pulsive traits [13, 22, 23, 25, 26]. Developing their typology
lationships between levels of physical activity and symptoms of GD based on the pathways model [24], Lee and colleagues
of GD over a 6-month period. [23] highlight that impulsive/aggressive problem gamers play
Curr Addict Rep

GAMING DISORDER
games to release their aggressive impulses, seek sensations,
and alleviate boredom while emotionally vulnerable problem
gamers play video games to escape or modify negative or
unpleasant mood states. These two underlying vulnerabilities
are consistent with longitudinal findings, where diffuse un-
pleasant or negative affective states (e.g., depression, low

- Perceived difficulties stopping

self-esteem), poor impulse control, and emotion dysregulation

Precipitating Risk Factors

Outcomes & Risk Factors
Outcomes/Consequences
- Intrusive thoughts regarding

- Increased impulsive gaming

- Low perceived life success

- Increased regret regarding

- Identification with avatar
are predictive of GD [35, 77].

- Inability to function on
- Psychological needs

other activities due to

- Gambling problems
Few longitudinal studies have investigated the role of cog-

gaming rumination

time spent gaming

- Perfectionism
nitions in predicting and maintaining GD. Cognitive factors

Adulthood

frustration
that have been investigated longitudinally include perfection-

gaming
ism, rumination, regret [40], negative beliefs about the future
[60], and reduced perceptions of behavioral control [44]. Of

GAMING DISORDER
the four included developmental models, only the I-PACE
model refers to specific cognitions and their relationship with
GD, with diminished behavioral control being an important
feature in the development of GD [25, 26]. Moreover, al-
though Paulus et al. [13] include lower self-control/self-regu-
lation in their model, only one longitudinal study has investi-
gated the role of self-control in predicting GD [46]. Additional

- Abstinence-related negative

Outcomes & Risk Factors

- Low emotional intelligence

Precipitating Risk Factors
nucleus activity after gaming
Outcomes/Consequences

- Negative beliefs regarding

Emerging adulthood

- Greater bilateral lentiform

- Reduced levels of sports/

- Decreased purpose in life

- Identification with avatar
- Increased procrastination
research would be necessary to replicate the finding that self-

- Achievement motives
- Attention difficulties
psychological distress

and coping flexibility

affect (e.g., boredom)
- Abstinence-related

- Escapism motives
control is a risk factor for GD in addition to generalizing these

physical activity
- Depression

- Anhedonia
- Loneliness
results beyond a single sample of South Korean adolescents.

the future
- Anxiety
In their model, Benarous and colleagues [22] outlined an
internalized and an externalized pathway to the development of
GD. Empirical evidence from the present review appears to sup-

GAMING DISORDER
port the presence of these pathways. Of note, evidence is avail-
able for precipitating risk factors within the internalized pathway
such as difficulties with peers [63], familial difficulties [34, 57],
poor social integration [66], and depression [30]; risk factors for
the externalized pathway including aggression [52], and inatten-
tion [39, 77]; and maintaining factors for both pathways includ-
Outcomes & Risk Factors

Precipitating Risk Factors

ing parental coercion [56], emotion dysregulation [35, 41, 57, communication with parents

- Reduced behavioral control

Outcomes/Consequences

- Poor parental relationship

- Delinquent peer affiliation

- Anxiety & social phobia

- Identification with avatar

- Low social competence
- Lower life satisfaction
- Relational aggression

- Emotion dysregulation
- Academic difficulties

79], and escapism [79]. However, numerous factors remain to

- Low RSA, High GSR
- Attention difficulties
- Physical aggression

- Social vulnerability

- Reduced quality of
- Conduct problems

- Peer victimization
- Low self-esteem

be investigated longitudinally including attachment style, adverse

Adolescence

- Depression

- Impulsivity
- Loneliness

life events, isolation, learning difficulties, and substance use.

- Male

Limitations of the GD Literature

GAMING DISORDER

Various limitations should be noted across the longitudinal

studies. First, the measures used to assess GD were primarily
self-reported. Although widely used, self-report measures are
susceptible to bias, subject to errors in recollection, and are
vulnerable to other sources of inaccurate responding. Second,
Precipitating Risk Factors

Fig. 3 Comprehensive conceptual model for the development of GD„

- Emotion dysregulation

- Problems with parents

- Academic difficulties
- Problems with peers

across the lifespan. Across the lifespan, various biological,

- Lack of empathy

psychological, and social factors converge to influence the development

- Autistic traits
Childhood

- Impulsivity

of GD. In each developmental period, various risk factors increase the

- Male

likelihood that an individual will develop GD, with certain risk factors
also being outcomes of GD earlier in life. This model suggests that
different risk factors are responsible for the relapse and reappearance, or
the spontaneous remission of GD. Note. RSA: respiratory sinus
arrhythmia; GSR: galvanic skin response
 Curr Addict Rep

regarding the longitudinal designs, it is questionable whether may encourage these efforts. Furthermore, there is a need for
the findings of each study extend beyond the period of data longitudinal research over periods of time exceeding 5 years
collection. With most studies being of 1 year in duration, to establish robust research findings across developmental pe-
further research is necessary to establish more distal predictors riods. Additionally, there is a need for further research inves-
and outcomes of GD. Third, with regard to the statistical tigating the role of variables that have only been cross-
methods, some studies did not control for baseline GD scores sectionally associated with GD (e.g., introversion, sensation
(e.g., Charlie et al. [32]), limiting the reliability of their find- seeking, generalized anxiety disorder, boredom). The devel-
ings. Fourth, as studies were included from different countries, opment of a comprehensive conceptual model for GD high-
cultural differences may limit the generalizability of certain lights additional directions for future research. Importantly, all
findings. Further, some studies included samples of individ- risk factors and outcomes included in the model are candidates
uals ranging from adolescence to late adulthood. These broad for replication and falsification. Additionally, there appears to
samples limit the potential to discern clear developmental tra- be a relative paucity of research investigating GD in adulthood
jectories and potential differences between subgroups. Lastly, compared to childhood and adolescence. Furthermore, al-
some studies had high rates of attrition across measurement though a diversity of psychological and social variables are
periods (e.g., Forrest et al. [40]) and others included small included in the model, there is a lack of longitudinal evidence
sample sizes (e.g., Dong et al. [38]). As such, results from elucidating the neurobiological changes that may occur in
these studies should be interpreted with caution given the individuals with GD. In addition, existential factors are largely
potential for biased attrition and reduced power to detect sta- unexplored and are missing from the conceptual model with
tistically significant differences. the exception of one study investigating the predictive role of
purpose in life [81]. The role of meaning, responsibility, and
Conceptual Model for the Development of Gaming belongingness may be relevant to the development of GD and
Disorder should be thoroughly considered.
In conclusion, the present review has synthesized longitu-
Although conceptual models for the development of GD inte- dinal evidence for the development of GD across the lifespan.
grate neurobiological, psychological, and social factors, a gen- Overall, GD is a complex disorder, influenced by a multitude
eral limitation of these models is that they do not specify the of intrapersonal and interpersonal risk and resiliency factors,
developmental processes and age-based trajectories of these while being associated with a host of negative outcomes.
various factors across the lifespan. Moreover, a significant lim- Given these findings, future research should critically differ-
itation of these models is that they predominantly integrate entiate between the antecedents and consequences of GD,
variables based on cross-sectional research. This can be prob- while conceptualizing this disorder from a developmentally
lematic as temporal associations cannot be established based informed perspective.
on these findings, with no distinguishing between factors that
are predictors or outcomes of GD. To address some of the Authors’ Contributions All authors contributed to coming up with the
idea of the review article. Jérémie Richard performed the literature search
abovementioned limitations and in goal of summarizing longi-
and drafted an initial version of the review article. Caroline Temcheff and
tudinal evidence, an integrated conceptual model for the devel- Jeffrey Derevensky provided critical review and suggestions for the final
opment of GD is proposed (Fig. 3). This model is unique in version of the article. All authors approved the final version of the man-
that it considers antecedents, consequences, and their overlap, uscript to be published.
in the developmental trajectory of GD across the lifespan.
Although this model is comprehensive, it is limited in that it Availability of Data and Material Not applicable.
does not account for the unique variance of each risk factor or
consequence over and above problem stability. Moreover, as Compliance with Ethical Standards
the included variables were not measured across different stud-
ies, it is possible that the significance of certain findings would Conflict of Interest The authors declare they have no conflict of interest.
disappear if the variance of each variable was accounted for
within the same model. Despite these limitations, this model is Ethics Approval Not applicable.
useful in that it summarizes all of the relevant variables than
have been prospectively associated with GD. Consent to Participate Not applicable.

Consent to Publication Not applicable.

Implications for Future Research
Human and Animal Rights and Informed Consent This article does not
Based on the findings of the present review, there is a need for contain any human or animal subjects.
greater consistency in the measures used to assess GD, with
the prospect that the inclusion of GD within the ICD-11 [1] Code Availability Not applicable.
Curr Addict Rep

References 18. Griffiths MD, Kuss DJ, Lopez-Fernandez O, Pontes HM.

Problematic gaming exists and is an example of disordered gaming.
J Behav Addict. 2017;6(3):296–301. https://doi.org/10.1556/2006.
1. World Health Organization. International statistical classification of
6.2017.037.
diseases and related health problems (11th Revision). 2019. https://
icd.who.int/browse11/l-m/en 19. Dosman CF, Andrews D, Goulden KJ. Evidence-based milestone
ages as a framework for developmental surveillance. Paediatr Child
2. American Psychiatric Association. Diagnostic and statistical man-
Health. 2012;17(10):561–8. https://doi.org/10.1093/pch/17.10.561.
ual of mental disorders (5th ed.). Washington: American
20. Loeber R, Farrington D, Stouthamer-Loeber M, van Kammen W.
Psychiatric Association; 2013.
Multiple risk factors for multi-problem boys: co-occurrence of de-
3. Grant JE, Potenza MN, Weinstein A, Gorelick DA. Introduction to
linquency, substance use, attention deficit, conduct problems, phys-
behavioral addictions. Am J Drug Alcohol Abuse. 2010;36(5):233–
ical aggression, covert behavior, depressed mood, and shy/
41. https://doi.org/10.3109/00952990.2010.491884.
withdrawn behavior. In: Jessor R, editor. New perspectives on ad-
4. Karim R, Chaudhri P. Behavioral addictions: an overview. J olescent risk behavior. New York: Cambridge University Press;
Psychoactive Drugs. 2012;44(1):5–17. https://doi.org/10.1080/ 1998. p. 90–149.
02791072.2012.662859.
21. Steinberg L, Morris AS. Adolescent development. Annu Rev
5. Mihara S, Higuchi S. Cross-sectional and longitudinal epidemio- Psychol. 2001;52(1):83–110. https://doi.org/10.1146/annurev.
logical studies of Internet gaming disorder: a systematic review of psych.52.1.83.
the literature. Psychiatry Clin Neurosci. 2017;71(7):425–44. https:// 22. Benarous X, Morales P, Mayer H, Iancu C, Edel Y, Cohen D.
doi.org/10.1111/pcn.12532. Internet gaming disorder in adolescents with psychiatric disorder:
6. Kuss DJ, Griffiths MD. Internet gaming addiction: a systematic two case reports using a developmental framework. Front
review of empirical research. Int J Ment Heal Addict. 2012;10(2): Psychiatry. 2019;10(336). doi:https://doi.org/10.3389/fpsyt.2019.
278–96. https://doi.org/10.1007/s11469-011-9318-5. 00336.
7. Feng W, Ramo DE, Chan SR, Bourgeois JA. Internet gaming dis- 23. Lee S-Y, Lee HK, Choo H. Typology of Internet gaming disorder
order: trends in prevalence 1998-2016. Addict Behav. 2017;75:17– and its clinical implications. Psychiatry Clin Neurosci. 2017;71(7):
24. https://doi.org/10.1016/j.addbeh.2017.06.010. 479–91. https://doi.org/10.1111/pcn.12457.
8. Reed GM, First MB, Kogan CS, Hyman SE, Gureje O, Gaebel W, 24. Blaszczynski A, Nower L. A pathways model of problem and path-
et al. Innovations and changes in the ICD-11 classification of men- ological gambling. Addiction. 2002;97(5):487–99.
tal, behavioural and neurodevelopmental disorders. World 25. Brand M, Wegmann E, Stark R, Muller A, Wolfling K, Robbins
Psychiatry. 2019;18(1):3–19. https://doi.org/10.1002/wps.20611. TW, et al. The Interaction of Person-Affect-Cognition-Execution
9. Kuss DJ, Griffiths MD. Online gaming addiction in children and (I-PACE) model for addictive behaviors: update, generalization to
adolescents: a review of empirical research. J Behav Addict. addictive behaviors beyond internet-use disorders, and specification
2012;1(1):3–22. https://doi.org/10.1556/jba.1.2012.1.1. of the process character of addictive behaviors. Neurosci Biobehav
10. Saunders JB, Hao W, Long J, King DL, Mann K, Fauth-Bühler M, Rev. 2019;104:1–10. https://doi.org/10.1016/j.neubiorev.2019.06.
et al. Gaming disorder: its delineation as an important condition for 032.
diagnosis, management, and prevention. J Behav Addict. 26. Brand M, Young KS, Laier C, Wolfling K, Potenza MN. (2016).
2017;6(3):271–9. https://doi.org/10.1556/2006.6.2017.039. Integrating psychological and neurobiological considerations re-
11. Cheng C, Cheung MWL, Wang H-y. Multinational comparison of garding the development and maintenance of specific Internet-use
internet gaming disorder and psychosocial problems versus well- disorders: an Interaction of Person-Affect-Cognition-Execution (I-
being: meta-analysis of 20 countries. Comput Hum Behav. PACE) model. Neurosci Biobehav Rev. 2016;71:252–66. https://
2018;88:153–67. https://doi.org/10.1016/j.chb.2018.06.033. doi.org/10.1016/j.neubiorev.2016.08.033.
12. González-Bueso V, Santamaría J, Fernández D, Merino L, Montero 27. Liu S, Yu C, Conner BT, Wang S, Lai W, Zhang W. Autistic traits
E, Ribas J. Association between internet gaming disorder or patho- and internet gaming addiction in Chinese children: the mediating
logical video-game use and comorbid psychopathology: a compre- effect of emotion regulation and school connectedness. Res Dev
hensive review. Int J Environ Res Public Health. 2018;15(4):668. Disabil. 2017;68:122–30. https://doi.org/10.1016/j.ridd.2017.07.
https://doi.org/10.3390/ijerph15040668. 011.
13. Paulus FW, Ohmann S, von Gontard A, Popow C. Internet gaming 28. Baggio S, Dupuis M, Studer J, Spilka S, Daeppen J-B, Simon O,
disorder in children and adolescents: a systematic review. Dev Med et al. Reframing video gaming and internet use addiction: empirical
Child Neurol. 2018;60(7):645–59. https://doi.org/10.1111/dmcn. cross-national comparison of heavy use over time and addiction
13754. scales among young users. Addiction. 2016;111(3):513–22.
14. Adams BLM, Stavropoulos V, Burleigh TL, Liew LWL, Beard CL, https://doi.org/10.1111/add.13192.
Griffiths MD. Internet gaming disorder behaviors in emergent 29. Brunborg GS, Mentzoni RA, Froyl LR. Is video gaming, or video
adulthood: a pilot study examining the interplay between anxiety game addiction, associated with depression, academic achievement,
and family cohesion. Int J Ment Heal Addict. 2019;17(4):828–44. heavy episodic drinking, or conduct problems? J Behav Addict.
https://doi.org/10.1007/s11469-018-9873-0. 2014;3(1):27–32.
15. Griffiths M. A ‘components’ model of addiction within a 30. Burleigh TL, Stavropoulos V, Liew LW, Adams BL, Griffiths MD.
biopsychosocial framework. J Subst Use. 2005;10(4):191–7. Depression, internet gaming disorder, and the moderating effect of
https://doi.org/10.1080/14659890500114359. the gamer-avatar relationship: an exploratory longitudinal study. Int
16. Krossbakken E, Pallesen S, Molde H, Mentzoni RA, Finseras TR. J Ment Heal Addict. 2018;16(1):102–24.
Not good enough? Further comments to the wording, meaning, and 31. Chang S-M, Lin SSJ. Online gaming motive profiles in late adoles-
the conceptualization of Internet Gaming Disorder. J Behav Addict. cence and the related longitudinal development of stress, depres-
2017;6(2):114–7. https://doi.org/10.1556/2006.6.2017.013. sion, and problematic internet use. Comput Educ. 2019;135:123–
17. Sugaya N, Shirasaka T, Takahashi K, Kanda H. Bio-psychosocial 37. https://doi.org/10.1016/j.compedu.2019.02.003.
factors of children and adolescents with internet gaming disorder: a 32. Charlie CWD, HyeKyung C, Khoo A. Role of parental relation-
systematic review. Biopsychosoc Med. 2019;13(1):3. https://doi. ships in pathological gaming. Procedia Soc Behav Sci. 2011;30:
org/10.1186/s13030-019-0144-5. 1230–6. https://doi.org/10.1016/j.sbspro.2011.10.238.
 Curr Addict Rep

33. Chen C, Zhang KZK, Gong X, Lee MKO. Failure to decrease the 50. Koning IM, Peeters M, Finkenauer C, van den Eijnden R.
addictive usage of information technologies: a theoretical model Bidirectional effects of Internet-specific parenting practices and
and empirical examination of smartphone game users. Comput compulsive social media and Internet game use. J Behav Addict.
Hum Behav. 2019;92:256–65. https://doi.org/10.1016/j.chb.2018. 2018;7(3):624–32.
11.025. 51. Konkoly Thege B, Woodin EM, Hodgins DC, Williams RJ. Natural
34. Choo H, Sim T, Liau AKF, Gentile DA, Khoo A. Parental influ- course of behavioral addictions: a 5-year longitudinal study. BMC
ences on pathological symptoms of video-gaming among children Psychiatry. 2015;15:4.
and adolescents: a prospective study. J Child Fam Stud. 2015;24(5): 52. Krossbakken E, Pallesen S, Mentzoni RA, King DL, Molde H,
1429–41. https://doi.org/10.1007/s10826-014-9949-9. Finserås TR, et al. A cross-lagged study of developmental trajecto-
35. Coyne SM, Dyer WJ, Densley R, Money NM, Day RD, Harper JM. ries of video game engagement, addiction, and mental health. Front
Physiological indicators of pathologic video game use in adoles- Psychol. 2018;9. https://doi.org/10.3389/fpsyg.2018.02239.
cence. J Adolesc Health. 2015;56(3):307–13. https://doi.org/10. 53. Lee D, Namkoong K, Lee J, Jung YC. Dorsal striatal functional
1016/j.jadohealth.2014.10.271. connectivity changes in Internet gaming disorder: a longitudinal
36. Dang DL, Zhang MX, Leong KKH, Anise MSW. The predictive magnetic resonance imaging study. Addict Biol. 2019. https://doi.
value of emotional intelligence for internet gaming disorder: a 1- org/10.1111/adb.12868.
year longitudinal study. Int J Environ Res Public Health. 54. Lemmens JS, Valkenburg PM, Peter J. The effects of pathological
2019;16(15). doi:https://doi.org/10.3390/ijerph16152762. gaming on aggressive behavior. J Youth Adolesc. 2011;40(1):38–
37. Dong G, Wang M, Liu X, Liang Q, Du X, Potenza MN. Cue- 47.
elicited craving-related lentiform activation during gaming depriva- 55. Lemmens JS, Valkenburg PM, Peter J. Psychosocial causes and
tion is associated with the emergence of Internet gaming disorder. consequences of pathological gaming. Comput Hum Behav.
Addict Biol. 2019;25. https://doi.org/10.1111/adb.12713. 2011;27(1):144–52. https://doi.org/10.1016/j.chb.2010.07.015.
38. Dong GH, Wang M, Zhang J, Du X, Potenza MN. Functional 56. Li AY-L, Lo BC-Y, Cheng C. It is the family context that matters:
neural changes and altered cortical-subcortical connectivity associ- concurrent and predictive effects of aspects of parent-child interac-
ated with recovery from Internet gaming disorder. J Behav Addict. tion on video gaming-related problems. Cyberpsychol Behav Soc
2019;8(4):692–702. https://doi.org/10.1556/2006.8.2019.75. Netw. 2018;21(6):374–80. https://doi.org/10.1089/cyber.2017.
39. Ferguson CJ, Ceranoglu TA. Attention problems and pathological 0566.
gaming: resolving the ‘chicken and egg’ in a prospective analysis. 57. Liau AK, Choo H, Li D, Gentile DA, Sim T, Khoo A. Pathological
Psychiatry Q. 2014;85(1):103–10. video-gaming among youth: a prospective study examining dynam-
40. Forrest CJ, King DL, Delfabbro PH. Maladaptive cognitions predict ic protective factors. Addict Res Theory. 2015;23(4):301–8. https://
changes in problematic gaming in highly-engaged adults: a 12- doi.org/10.3109/16066359.2014.987759.
month longitudinal study. Addict Behav. 2017;65:125–30. https:// 58. Liew LWL, Stavropoulos V, Adams BLM, Burleigh TL, Griffiths
doi.org/10.1016/j.addbeh.2016.10.013. MD. Internet gaming disorder: the interplay between physical ac-
41. Gentile DA, Choo H, Liau A, Sim T, Li D, Fung D, et al. tivity and user–avatar relationship. Behav Inform Technol.
Pathological video game use among youths: a two-year longitudi- 2018;37(6):558–74. https://doi.org/10.1080/0144929x.2018.
nal study. Pediatrics. 2011;127(2):e319–29. 1464599.
42. Gentile DA, Berch ON, Choo H, Khoo A, Walsh DA. Bedroom 59. Liu L, Yao Y-W, Li C-SR, Zhang J-T, Xia C-C, Lan J, et al. The
media: one risk factor for development. Dev Psychol. 2017;53(12): comorbidity between internet gaming disorder and depression: in-
2340–55. terrelationship and neural mechanisms. Front Psych. 2018;9.
43. Guillot CR, Bello MS, Tsai JY, Huh J, Leventhal AM, Sussman S. https://doi.org/10.3389/fpsyt.2018.00154.
Longitudinal associations between anhedonia and internet-related 60. Lukavska K. The immediate and long-term effects of time perspec-
addictive behaviors in emerging adults. Comput Hum Behav. tive on internet gaming disorder. J Behav Addict. 2018;7(1):44–51.
2016;62:475–9. 61. Marmet S, Studer J, Grazioli VS, Gmel G. Bidirectional associa-
44. Haagsma MC, King DL, Pieterse ME, Peters O. Assessing prob- tions between self-reported gaming disorder and adult attention
lematic video gaming using the theory of planned behavior: a lon- deficit hyperactivity disorder: evidence from a sample of young
gitudinal study of Dutch young people. Int J Ment Heal Addict. Swiss men. Front Psych. 2018;9. https://doi.org/10.3389/fpsyt.
2013;11(2):172–85. https://doi.org/10.1007/s11469-012-9407-0. 2018.00649.
45. Henchoz Y, Studer J, Deline S, N’Goran AA, Baggio S, Gmel G. 62. Molde H, Holmøy B, Merkesdal AG, Torsheim T, Mentzoni RA,
Video gaming disorder and sport and exercise in emerging adult- Hanns D, et al. Are video games a gateway to gambling? A longi-
hood: a longitudinal study. Behav Med. 2016;42(2):105–11. https:// tudinal study based on a representative Norwegian sample. J Gambl
doi.org/10.1080/08964289.2014.965127. Stud. 2019;35(2):545–57. https://doi.org/10.1007/s10899-018-
46. Jeong EJ, Ferguson CJ, Lee SJ. Pathological gaming in young 9781-z.
adolescents: a longitudinal study focused on academic stress and 63. Mößle TR, F. Predictors of problematic video game usage in child-
self-control in South Korea. J Youth Adolesc. 2019;48(12):2333– hood and adolescence. Sucht. 2013;59(3):153–64. https://doi.org/
42. 10.1024/0939-5911.a000247.
47. Kaptsis D, King DL, Delfabbro PH, Gradisar M. Trajectories of 64. Peeters M, Koning I, Van Den Eijnden R. Predicting internet gam-
abstinence-induced Internet gaming withdrawal symptoms: a pro- ing disorder symptoms in young adolescents: a one-year follow-up
spective pilot study. Addict Behav Rep. 2016;4:24–30. study. Comput Hum Behav. 2018;80:255–61. https://doi.org/10.
48. King DL, Delfabbro PH, Griffiths MD. Trajectories of problem 1016/j.chb.2017.11.008.
video gaming among adult regular gamers: an 18-month longitudi- 65. Peeters M, Koning I, Lemmens J, Eijnden RVD. Normative, pas-
nal study. Cyberpsychol Behav Soc Netw. 2013;16(1):72–6. sionate, or problematic? Identification of adolescent gamer sub-
https://doi.org/10.1089/cyber.2012.0062. types over time. J Behav Addict. 2019;8(3):574–85. https://doi.
49. King DL, Kaptsis D, Delfabbro PH, Gradisar M. Craving for inter- org/10.1556/2006.8.2019.55.
net games? Withdrawal symptoms from an 84-h abstinence from 66. Rehbein F, Baier D. Family-, media-, and school-related risk factors
massively multiplayer online gaming. Comput Hum Behav. of video game addiction: a 5-year longitudinal study. J Media
2016;62:488–94. https://doi.org/10.1016/j.chb.2016.04.020. Psychol. 2013;25(3):118–28.
Curr Addict Rep

67. Rothmund T, Klimmt C, Gollwitzer M. Low temporal stability of Comput Hum Behav. 2015;50:159–68. https://doi.org/10.1016/j.
excessive video game use in German adolescents. J Media Psychol. chb.2015.03.079.
2018;30(2):53–65. https://doi.org/10.1027/1864-1105/a000177. 83. Lemmens JS, Valkenburg PM, Peter J. Development and validation
68. Scharkow M, Festl R, Quandt T. Longitudinal patterns of problem- of a game addiction scale for adolescents. Media Psychol.
atic computer game use among adolescents and adults-a 2-year 2009;12(1):77–95. https://doi.org/10.1080/15213260802669458.
panel study. Addiction. 2014;109(11):1910–7. https://doi.org/10. 84. Gentile D. Pathological video-game use among youth ages 8 to 18:
1111/add.12662. a national study. Psychol Sci. 2009;20(5):594–602. https://doi.org/
69. Schmitt ZL, Livingston MG. Video game addiction and college 10.1111/j.1467-9280.2009.02340.x.
performance among males: results from a 1 year longitudinal study. 85. American Psychiatric Association. Diagnostic and statistical man-
Cyberpsychol Behav Soc Netw. 2015;18(1):25–9. https://doi.org/ ual of mental disorders (4th ed. text revision). Washington:
10.1089/cyber.2014.0403. American Psychiatric Association; 2000.
70. Schoenmacker GH, Groenman AP, Sokolova E, Oosterlaan J, 86. Pontes HM, Griffiths MD. Measuring DSM-5 Internet gaming dis-
Rommelse N, Roeyers H, et al. Role of conduct problems in the order: development and validation of a short psychometric scale.
relation between attention-deficit hyperactivity disorder, substance Comput Hum Behav. 2015;45:137–43. https://doi.org/10.1016/j.
use, and gaming. Eur Neuropsychopharmacol. 2020;30:102–13. chb.2014.12.006.
https://doi.org/10.1016/j.euroneuro.2018.06.003. 87. Lemmens JS, Valkenburg PM, Gentile DA. The internet gaming
71. Stavropoulos V, Burleigh TL, Beard CL, Gomez R, Griffiths MD. disorder scale. Psychol Assess. 2015;27(2):567–82. https://doi.org/
Being there: a preliminary study examining the role of presence in 10.1037/pas0000062.
internet gaming disorder. Int J Ment Heal Addict. 2019;17:880–90. 88. Vadlin S, Åslund C, Nilsson KW. Development and content valid-
https://doi.org/10.1007/s11469-018-9891-y. ity of a screening instrument for gaming addiction in adolescents:
72. Su P, Yu C, Zhang W, Liu S, Xu Y, Zhen S. Predicting Chinese the Gaming Addiction Identification Test (GAIT). Scand J Psychol.
adolescent internet gaming addiction from peer context and norma- 2015;56(4):458–66. https://doi.org/10.1111/sjop.12196.
tive beliefs about aggression: a 2-year longitudinal study. Front 89. Mößle T. [“fat, stupid, addicted, violent?” Problematic media usage
Psychol. 2018;9. https://doi.org/10.3389/fpsyg.2018.01143. behavior and its consequences in childhood. Results of the Berlin
73. Su B, Yu C, Zhang W, Su Q, Zhu J, Jiang Y. Father–child longi- longitudinal study media.]. Baden Baden: Nomos; 2012.
tudinal relationship: parental monitoring and internet gaming disor- 90. Rehbein F, Kleimann M, Mößle T. Prevalence and risk factors of
der in Chinese adolescents. Front Psychol. 2018;9. https://doi.org/ video game dependency in adolescence: results of a German na-
10.3389/fpsyg.2018.00095. tionwide survey. Cyberpsychol Behav Soc Netw. 2010;13(3):269–
74. Vadlin S, Aslund C, Nilsson KW. A longitudinal study of the 77. https://doi.org/10.1089/cyber.2009.0227.
individual- and group-level problematic gaming and associations
91. Young KS. Internet addiction: the emergence of a new clinical
with problem gambling among Swedish adolescents. Brain
disorder. CyberPsychol Behav. 1998;1(3):237–44. https://doi.org/
Behav. 2018;8(4):e00949.
10.1089/cpb.1998.1.237.
75. van den Eijnden R, Koning I, Doornwaard S, van Gurp F, Ter Bogt
92. Chen C-Y, Huang M-F, Yen J-Y, Chen C-S, Liu G-C, Yen C-F,
T. The impact of heavy and disordered use of games and social
et al. Brain correlates of response inhibition in Internet gaming
media on adolescents' psychological, social, and school function-
disorder. Psychiatry Clin Neurosci. 2015;69(4):201–9. https://doi.
ing. J Behav Addict. 2018;7(3):697–706.
org/10.1111/pcn.12224.34.
76. Van Rooij AJ, Schoenmakers TM, Vermulst AA, Van Den Eijnden
93. Mak KK, Nam JK, Kim D, Aum N, Choi JS, Cheng C, et al. Cross-
RJJM, Van De Mheen D. Online video game addiction: identifica-
cultural adaptation and psychometric properties of the Korean Scale
tion of addicted adolescent gamers. Addiction. 2011;106(1):205–
for Internet Addiction (K-Scale) in Japanese high school students.
12. https://doi.org/10.1111/j.1360-0443.2010.03104.x.
Psychiatry Res. 2017;249:343–8. https://doi.org/10.1016/j.
77. Wartberg L, Kriston L, Zieglmeier M, Lincoln T, Kammerl R. A
psychres.2017.01.044.
longitudinal study on psychosocial causes and consequences of
94. Meerkerk GJ, Van Den Eijnden RJ, Vermulst AA, Garretsen HF.
Internet gaming disorder in adolescence. Psychol Med.
The Compulsive Internet Use Scale (CIUS): some psychometric
2019;49(2):287–94. https://doi.org/10.1017/s003329171800082x.
properties. CyberPsychol Behav. 2009;12(1):1–6. https://doi.org/
78. Weinstein N, Przybylski AK, Murayama K. A prospective study of
10.1089/cpb.2008.0181.
the motivational and health dynamics of Internet Gaming Disorder.
PeerJ. 2017;5:e3838. 95. Liao PC, Chen SK, Lin SSJ. Latent profiles of stress and their
79. Wichstrom L, Stenseng F, Belsky J, von Soest T, Hygen BW. relationships with depression and problematic Internet use among
Symptoms of Internet gaming disorder in youth: predictors and college freshmen. Scand J Psychol. 2018;59(6):621–30. https://doi.
comorbidity. J Abnorm Child Psychol. 2019;47(1):71–83. org/10.1111/sjop.12489.
80. Yu C, Li X, Zhang W. Predicting adolescent problematic online 96. Williams RJ, Volberg RA. The classification accuracy of four prob-
game use from teacher autonomy support, basic psychological lem gambling assessment instruments in population research. Int
needs satisfaction, and school engagement: a 2-year longitudinal Gambl Stud. 2014;14(1):15–28. https://doi.org/10.1080/
study. Cyberpsychol Behav Soc Netw. 2015;18(4):228–33. 14459795.2013.839731.
https://doi.org/10.1089/cyber.2014.0385. 97. Charlton JP, Danforth IDW. Distinguishing addiction and high en-
81. Zhang MX, Wang X, Yu SM, Wu AMS. Purpose in life, social gagement in the context of online game playing. Comput Hum
support, and internet gaming disorder among Chinese university Behav. 2007;23(3):1531–48. https://doi.org/10.1016/j.chb.2005.
students: a 1-year follow-up study. Addict Behav. 2019;99: 07.002.
106070. https://doi.org/10.1016/j.addbeh.2019.106070.
82. Zhu J, Zhang W, Yu C, Bao Z. Early adolescent Internet game Publisher’s Note Springer Nature remains neutral with regard to jurisdic-
addiction in context: how parents, school, and peers impact youth. tional claims in published maps and institutional affiliations.