Skip to main content

Empirical evidence for robust personality-gaming disorder associations from a large-scale international investigation applying the APA and WHO frameworks

PLOS ONE, 2021

Christian Montag, Christopher Kannen, Bruno Schivinski, Halley M. Pontes

PLOS ONE, volume 16, issue 12, article e0261380, 22 December 2021

  • Open access
  • 50,925 gamers
  • 150 countries
  • Online survey
  • Big Five personality
  • WHO vs APA frameworks
  • Open data (OSF)
Download the full text (PDF)Published version

Free full text: the open access article (CC BY 4.0), re-typeset as a clean PDF with the same content as the published version. Please cite the published version.

Keywords
  • gaming disorder
  • internet gaming disorder
  • disordered gaming
  • personality
  • Big Five
  • conscientiousness
  • neuroticism
  • extraversion
  • agreeableness
  • openness
  • time spent gaming
  • individual differences
  • addictive behaviours
  • ICD-11
  • DSM-5
  • video games
  • I-PACE model
  • Big Five Inventory (BFI-44)
  • Gaming Disorder Test (GDT)
  • Internet Gaming Disorder Scale–Short-Form (IGDS9-SF)
  • large-scale online survey
  • cross-cultural study
  • Bayesian t-tests
  • multiple linear regression

Gamers low in conscientiousness and high in neuroticism are the most likely to show symptoms of gaming disorder, and this holds whether disorder is defined by the WHO or the APA. In 50,925 gamers from 150 countries, personality was linked much more strongly to disordered gaming than to the number of hours spent gaming.

Key findings

  1. Low conscientiousness and high neuroticism go with gaming disorder. Among 50,925 gamers from 150 countries, gaming disorder symptoms correlated with low conscientiousness (r = −.37 under the WHO framework, −.33 under the APA framework) and high neuroticism (r = .27 and .33) (Montag et al., 2021).
  2. The links hold across both diagnostic frameworks. Controlling for age, gender and weekly gaming time, conscientiousness (β = −.28 WHO, −.20 APA) and neuroticism (β = .14 WHO, .21 APA) remained the strongest personality predictors; the models explained about 22% of the variance in symptoms (Montag et al., 2021).
  3. Personality says little about how long people play. Correlations between the Big Five traits and weekly time spent gaming were small (−.12 to .07), so personality matters for disordered gaming rather than for gaming time (Montag et al., 2021).
  4. Disordered gamers differ on every trait. Gamers meeting the WHO criteria (2.2% of the sample) or the APA criteria (5.5%) scored lower on conscientiousness, agreeableness, extraversion and openness and higher on neuroticism than other gamers (all p < .001) (Montag et al., 2021).

At a glance

Design
Cross-sectional online survey on the platform do-i-play-too-much.com, part of the ‘Smart Gaming’ campaign with the Electronic Sports League (ESL); May to December 2019 (before the COVID-19 pandemic)
Participants
50,925 gamers from 150 countries (90.3% men; mean age 24.9, SD 7.4), from 68,132 who took part
Measures
Big Five Inventory (BFI-44); Gaming Disorder Test (GDT, WHO framework); Internet Gaming Disorder Scale–Short-Form (IGDS9-SF, APA framework); weekly time spent gaming
Analysis
Welch’s and Bayesian t-tests by gender, correlations with Holm correction and multiple linear regression, in R 4.0.4 and IBM SPSS
Data quality
Exclusions for age, consent, English proficiency, a sham game item (‘Semeron’), not gaming in the past year and implausible gaming hours; long-string screening for careless responding
Ethics
Nottingham Trent University ethics committee (PONTES 2018/95)
Data
Aggregated data on the Open Science Framework (osf.io/ntyhr)

Figures 1 and 2. Personality of disordered and non-disordered gamers

Non-disordered gamersDisordered gamersWHO framework (Gaming Disorder Test)non-disordered n = 49,789; disordered n = 1,13612345Openness, non-disordered: 3.633.63Openness, disordered: 3.443.44OpennessConscientiousness, non-disordered: 3.223.22Conscientiousness, disordered: 2.562.56Conscien-tiousnessExtraversion, non-disordered: 2.762.76Extraversion, disordered: 2.472.47ExtraversionAgreeableness, non-disordered: 3.533.53Agreeableness, disordered: 3.233.23AgreeablenessNeuroticism, non-disordered: 2.922.92Neuroticism, disordered: 3.523.52NeuroticismAPA framework (IGDS9-SF)non-disordered n = 48,138; disordered n = 2,78712345Openness, non-disordered: 3.643.64Openness, disordered: 3.473.47OpennessConscientiousness, non-disordered: 3.233.23Conscientiousness, disordered: 2.732.73Conscien-tiousnessExtraversion, non-disordered: 2.772.77Extraversion, disordered: 2.472.47ExtraversionAgreeableness, non-disordered: 3.543.54Agreeableness, disordered: 3.223.22AgreeablenessNeuroticism, non-disordered: 2.902.90Neuroticism, disordered: 3.513.51NeuroticismMean trait score (BFI-44, 1–5)
Redrawn from Figs 1 and 2 of the article (values as labelled on the published charts). Under both frameworks, disordered gamers score markedly lower on conscientiousness and higher on neuroticism. All contrasts significant at p < .001.

How strongly each trait correlates with gaming disorder symptoms

Gaming disorder (WHO)Conscientiousness (low)−.37Neuroticism (high).27Agreeableness (low)−.19Extraversion (low)−.16Openness (low)−.11Weekly time spent gamingConscientiousness−.12Neuroticism.07
Absolute correlations with Gaming Disorder Test scores (WHO framework; Table 2 of the article, N = 50,925). Neuroticism is positive; the other traits are negative.

Tables

Table 1. Descriptive statistics and gender-based analysis of relevant variables

Variable All Men Women t (df) p BF10 d
Gaming Disorder 9.05 (3.26) 9.05 (3.25) 9.01 (3.35) −0.84 (5960.1) .40 −3.73 −.01
Internet Gaming Disorder 18.25 (6.25) 18.23 (6.22) 18.43 (6.52) 2.03 (5931.2) .04 −2.03 .03
Weekly Gaming Time (hours) 24.43 (16.94) 24.66 (16.89) 22.21 (17.23) −9.52 (5988.7) <.001 41.19 −.14
Openness 3.63 (0.58) 3.62 (0.58) 3.71 (0.60) 10.58 (5972.7) <.001 51.74 .16
Conscientiousness 3.21 (0.66) 3.21 (0.66) 3.15 (0.71) −6.13 (5880.1) <.001 14.70 −.09
Extraversion 2.75 (0.78) 2.77 (0.78) 2.63 (0.79) −11.84 (6009.3) <.001 65.88 −.17
Agreeableness 3.52 (0.62) 3.53 (0.61) 3.47 (0.67) −6.07 (5837.2) <.001 14.32 −.09
Neuroticism 2.94 (0.83) 2.88 (0.82) 3.44 (0.82) 45.45 (6011.8) <.001 > 100 .68

Means with standard deviations in brackets. t = Welch’s t-test statistic. Gaming disorder measured with the Gaming Disorder Test (GDT); Internet gaming disorder with the Internet Gaming Disorder Scale–Short-Form (IGDS9-SF). BF10 = Bayes factor; d = Cohen’s d. Overall N = 50,925; men n = 45,996; women n = 4,929. Bayes factors as printed.

Table 2. Associations between personality traits, disordered gaming, time spent gaming, age, and gender

Openness Conscientiousness Extraversion Agreeableness Neuroticism
Gaming Disorder −.11 −.37 −.16 −.19 .27
Internet Gaming Disorder −.12 −.33 −.19 −.21 .33
Weekly Time Spent Gaming −.10 −.12 −.10 −.09 .07
Age .11 .17 −.01 .09 −.04
Gender .05 −.03 −.05 −.03 .20

All correlations remained significant at p < .001 after Holm correction. Gaming disorder measured with the GDT; Internet gaming disorder with the IGDS9-SF.

Table 3. Multiple linear regression of the relationship between disordered gaming and key individual differences predictors

Predictor B SE β t p 95% CI lower 95% CI upper
Outcome: Gaming Disorder
(Intercept) 11.831 .160 – 74.034 <.001 11.517 12.144
Age −.005 .002 −.011 −2.674 .008 −.008 −.001
Gender .332 .044 .030 7.535 <.001 .246 .419
Weekly time spent gaming .046 .001 .240 60.440 <.001 .045 .048
Openness −.051 .023 −.009 −2.142 .032 −.096 −.004
Conscientiousness −1.347 .021 −.275 −63.041 <.001 −1.389 −1.305
Extraversion −.062 .018 −.015 −3.395 .001 −.098 −.026
Agreeableness −.276 .022 −.052 −12.383 <.001 −.319 −.232
Neuroticism .527 .018 .135 29.502 <.001 .492 .562
Variance explained by model: Adjusted R2 = .223 (22.3%)
Statistical significance of model: F(8, 50916) = 1826.904, p < .001
Outcome: Internet Gaming Disorder
(Intercept) 20.136 .306 – 65.761 <.001 19.536 20.736
Age −.002 .003 −.003 −.673 .501 −.009 .004
Gender .661 .084 .031 7.826 <.001 .496 .827
Weekly time spent gaming .089 .001 .242 60.923 <.001 .086 .092
Openness −.149 .045 −.014 −3.349 <.001 −.237 −.062
Conscientiousness −1.895 .041 −.201 −46.284 <.001 −1.975 −1.815
Extraversion −.231 .035 −.029 −6.596 <.001 −.300 −.163
Agreeableness −.587 .043 −.058 −13.762 <.001 −.671 −.504
Neuroticism 1.607 .034 .214 46.948 <.001 1.540 1.674
Variance explained by model: Adjusted R2 = .226 (22.6%)
Statistical significance of model: F(8, 50916) = 1861.415, p < .001

B: unstandardised coefficient; SE: standard error; β: standardised coefficient. Gaming disorder measured with the GDT; Internet gaming disorder with the IGDS9-SF.

The question

Personality is thought to shape vulnerability to disordered gaming, but most studies used small, single-country samples and only the APA framework, and none compared how personality relates to disordered gaming versus time spent gaming. In a large, culturally diverse sample we asked whether the Big Five traits, especially low conscientiousness and high neuroticism, relate to disordered gaming under both the WHO and APA frameworks, and whether they relate differently to gaming time.

Implications

Practical implications

Short screenings of conscientiousness and neuroticism could help identify gamers vulnerable to disordered gaming and target prevention. Because personality relates to disordered gaming much more than to hours played, high gaming engagement should not be confused with disorder.

Theoretical implications

The study confirms low conscientiousness and high neuroticism as robust correlates of disordered gaming across both diagnostic frameworks and many countries, supporting the role of personality in models such as I-PACE, and shows that disordered gaming and time spent gaming are distinct.

Limitations and future research

  • The cross-sectional design cannot show whether personality causes disordered gaming or results from it.
  • All measures are self-reports, and classifying disordered gaming from questionnaires needs clinical interviews to yield a diagnosis.
  • The English-language survey and the self-help platform may have biased the sample towards heavier or more vulnerable gamers, and few women took part.

Questions this paper answers

Which personality traits are linked to gaming disorder?

Low conscientiousness and high neuroticism, in a study of 50,925 gamers from 150 countries; lower agreeableness, extraversion and openness showed weaker links (Montag et al., 2021).

Is personality related to how much people play video games?

Only weakly. Correlations between the Big Five traits and weekly gaming time ranged from −.12 to .07, much smaller than their correlations with disordered gaming (Montag et al., 2021).

How common was gaming disorder in this sample?

About 2.2% of gamers met the WHO criteria for gaming disorder and 5.5% met the APA criteria for Internet gaming disorder (Montag et al., 2021).

Do men and women differ in disordered gaming?

In this sample, men and women did not differ meaningfully in disordered gaming scores, although men played more hours and women scored higher on neuroticism (Montag et al., 2021).

Key concepts

Big Five personality traits
Openness, conscientiousness, extraversion, agreeableness and neuroticism, the most widely used model of personality.
Conscientiousness
A tendency to be organised, dependable, self-disciplined and goal-directed.
Neuroticism
A tendency to experience negative emotions such as anxiety, worry and moodiness.
Gaming disorder (WHO)
An ICD-11 diagnosis marked by impaired control over gaming, increasing priority given to gaming and continued gaming despite negative consequences, with significant impairment.
Internet gaming disorder (APA)
A DSM-5 condition for further study, defined by at least five of nine criteria within 12 months.
I-PACE model
The Interaction of Person-Affect-Cognition-Execution model of how addictive behaviours develop.

Open science and materials

Study materials

Scoring
Disordered gaming (WHO): all four GDT items answered 4 = ‘often’ or 5 = ‘very often’. Disordered gaming (APA): five of the nine IGDS9-SF items answered 4 or 5. BFI-44 items rated from 1 = ‘disagree strongly’ to 5 = ‘agree strongly’.
Country-level results
Participant numbers per country and personality–disordered gaming correlations for each country with at least 1% of the sample are in S1 Data of the article.

How to cite

Montag, C., Kannen, C., Schivinski, B., & Pontes, H. M. (2021). Empirical evidence for robust personality-gaming disorder associations from a large-scale international investigation applying the APA and WHO frameworks. PLOS ONE, 16(12), Article e0261380. https://doi.org/10.1371/journal.pone.0261380

Open access under a Creative Commons CC BY 4.0 licence.

Related publications

All publications