Introduction
Gaming has become increasingly popular across the world during the COVID-19 pandemic (King et al., Reference King, Delfabbro, Billieux and Potenza2020), especially among adolescents (Donati et al., Reference Donati, Guido, De Meo, Spalice, Sanson, Beccari and Primi2021). Governments and public health authorities, in an effort to curb the spread of COVID-19, utilised various measures to increase social distancing such as the cancellation of most social activities available to adolescents (Ko and Yen Reference Ko and Yen2020). Gaming emerged as an acceptable activity during the pandemic for adolescents and was promoted by the World Health Organization (WHO) in their ‘#PlayApartTogether’ campaign to encourage social distancing among young people (Donati et al., Reference Donati, Guido, De Meo, Spalice, Sanson, Beccari and Primi2021). While the gaming industry is now reporting a decline in engagement in online gaming services as people return to more real-world pursuits (Gross Reference Gross2022), overall levels of gaming are significantly higher than pre-pandemic levels (Howley Reference Howley2022).
Gaming can be a positive activity to help adolescents cope with the stressors of social distancing restrictions, such as social isolation (Jones et al., Reference Jones, Scholes, Johnson, Katsikitis and Carras2014).. Gaming can have benefits as a coping strategy in times of heightened psychological stress (Russoniello et al., Reference Russoniello, O’Brien and Parks2009), and casual levels of gameplay have been shown to reduce symptoms of anxiety, depression, stress and low mood over a 1-month period (Fish et al., Reference Fish, Russoniello and O’Brien2014). Despite gaming’s multiple benefits, increased frequency and duration of gaming can increase the risk of developing gaming disorder for a minority of individuals (Mihara and Higuchi Reference Mihara and Higuchi2017). An increase in gaming disorder rates among adolescents were observed in many studies over the course of the COVID-19 pandemic, with psychological distress and social isolation due to the pandemic cited as driving factors (Han et al., Reference Han, Cho, Sung and Park2022). Worldwide, there is an estimated 3.05% prevalence rate of gaming disorder within the adolescent population, with significantly higher prevalence in the male population with a ratio of 2.5:1 (Stevens et al., Reference Stevens, Dorstyn, Delfabbro and King2021).
It is important to delineate the differences between non-problematic gaming and gaming disorder or an addiction to gaming. Gaming exists on a spectrum from non-problematic occasional gaming, through to problematic gaming and gaming disorder for a minority of individuals (Griffiths et al., Reference Griffiths, Kuss, Lopez-Fernandez and Pontes2017). Problematic gaming is an example of disordered gaming where there are significant disruptions to psychosocial functioning due to gaming (Griffiths et al., Reference Griffiths, Kuss, Lopez-Fernandez and Pontes2017). The International Classification of Diseases 11th Revision (ICD-11) have described gaming disorder as a ‘pattern of persistent or recurrent gaming behaviour (‘digital gaming’ or ‘video gaming’), which may be online (i.e. over the internet) or offline’ with the following characteristics demonstrated over a 12-month period:
-
‘Impaired control over gaming (e.g. onset, frequency, intensity, duration, termination, context)
-
Increasing priority given to gaming to the extent that gaming takes precedence over other life interests and daily activities
-
Continuation or escalation of gaming despite the occurrence of negative consequences’ (World Health Organization 2018).
There was significant debate amongst scholars about the inclusion of gaming disorder in the ICD-11 (Griffiths et al., Reference Griffiths, Kuss, Lopez-Fernandez and Pontes2017), as well as critique of the terminology used in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) for Internet Gaming Disorder as an emerging disorder (Kuss et al., Reference Kuss, Griffiths and Pontes2017), where the terms internet addiction, gaming addiction and gaming disorder are used interchangeably (American Psychiatric Association, 2013). For clarity, this review will use the term ‘gaming disorder” to also describe gaming addiction and problematic gaming. Since the introduction of gaming disorder in the ICD-11, there still remains some ongoing debate amongst researchers as to the validity and reliability of the proposed clinical criteria for gaming disorder (Brand et al., Reference Brand, Rumpf, King, Potenza and Wegmann2020). In support of the validity of gaming disorder, there are a growing number of neuroimaging studies supporting gaming disorder as a distinct entity from a neurobiological perspective (Kuss et al., Reference Kuss, Pontes and Griffiths2018, Vaccaro and Potenza Reference Vaccaro and Potenza2019).
There have been a number of systematic reviews exploring the neurobiological findings associated with gaming disorder in adults. Magnetic resonance imaging (MRI) studies have shown impaired cognitive control and a deficiency in ventral striatal reward systems (Kuss et al., Reference Kuss, Pontes and Griffiths2018). Alterations in the striatum, amygdala and orbitofrontal cortex have been associated with higher levels of video game craving in gaming-disordered adults (Choi et al., Reference Choi, Shin, Ryu, Jung, Hyun, Kim and Park2021). Electroencephalogram (EEG) findings in adult gaming disorder are similar to cocaine and alcohol addiction, with increased P300 latency and decreased P300 amplitudes suggesting poor attention capacity and allocation of attention (Kuss et al., Reference Kuss, Pontes and Griffiths2018). While some systematic reviews have focused on younger age groups (Sugaya et al., Reference Sugaya, Shirasaka, Takahashi and Kanda2019), these reviews mixed adult and adolescent samples in their findings. To our knowledge, there are no systematic reviews focused purely on the neurobiological findings of gaming disorder in an adolescent (under 18 years old) age group. This cohort are especially vulnerable given the developing nature of the adolescent brain and increasing rates of gaming. In addition, increased knowledge in this area would be of importance to child and adolescent mental health professionals treating under-18-year-old patients attending their services, especially given the recent inclusion of gaming disorder in ICD-11 and the potential for referrals of this nature to child and adolescent services. Therefore, this systematic review explored the neuroimaging literature in adolescents (under 18 years old) with gaming disorder.
Methods
Using PRISMA 2020 guidelines, a systematic review on studies investigating neuroimaging findings in gaming disorder in an adolescent (under 18 years old) population. The methods for this review were adapted from similar systematic reviews conducted in gaming disorder (Kuss and Griffiths Reference Kuss and Griffiths2012, Yao et al., Reference Yao, Liu, Ma, Shi, Zhou, Zhang and Potenza2017, Kuss et al., Reference Kuss, Pontes and Griffiths2018, Sugaya et al., Reference Sugaya, Shirasaka, Takahashi and Kanda2019, Burleigh et al., Reference Burleigh, Griffiths, Sumich, Wang and Kuss2020). All included studies met the following criteria: 1) adolescents with a sample mean age of under 18 years old, 2) studies focused on gaming disorder, 3) neuroimaging techniques used within the study, 4) primary studies (cross-sectional and cohort studies) and 5) papers in the English language. Exclusion criteria were 1) type of study (case reports, case series), 2) studies with a sample mean age of over 18 years old, 3) studies that mislabelled generalised internet addiction as gaming disorder, 4) inadequate or absent comparison groups and 5) non-English-language studies. Electronic databases used to search for primary studies were PubMed, PsycINFO, Web of Science and CINAHL Plus. Comprehensive search terms focused on population and intervention were entered into an advanced search engine including (‘Addiction’ OR ‘Pathology’ OR ‘Disorder’ OR ‘Compulsive’ OR ‘Problem’) AND (‘Gaming’ OR ‘Internet’) AND (‘imaging’ OR ‘neurobiological’ OR ‘neuroscience’ OR neuropsychological’). Search filters were applied to include child, adolescent and English-based papers while excluding literature reviews, systematic reviews and meta-analyses.
A total of 3288 papers were identified on initial searches. After removing duplicates (692 papers), application of exclusion criteria and screening of papers based on title, abstract, keywords and publication type, 24 studies were included for analysis in this review (see Table 1). A PRISMA flow diagram (see Fig. 1) is included as per the PRISMA 2020 statement guidelines (Page et al., Reference Page, McKenzie, Bossuyt, Boutron, Hoffmann, Mulrow, Shamseer, Tetzlaff, Akl and Brennan2021). Studies were assessed for their risk of bias with JBI’s Critical Appraisal Tools (Ma et al., Reference Ma, Wang, Yang, Huang, Weng and Zeng2020). This systematic review was registered with PROSPERO and the ID number is CRD42021264905. To avoid confusion, the term gaming disorder will be used interchangeably with gaming addiction and problematic gaming.
Table 1. Studies included in systematic review of adolescent gaming disorder (GD)
Figure 1. Prisma flow diagram of study selection.
Results
Electroencephalography (EEG)
Two studies conducted using task-based interventions and resting-state EEG compared gaming-disordered samples and the non-disordered adolescent controls. EEG can be used to detect event-related potentials (ERP), small voltages generated by the brain in response to different sensory, motor, or cognitive actions made by participants (Sur and Sinha Reference Sur and Sinha2009). During task-based interventions, adolescents with gaming disorder demonstrated less positive P3 amplitudes compared to controls in response inhibition tasks (Li et al., Reference Li, Wang, Yang, Dai, Zheng, Sun and Liu2020). Less positive P3 amplitudes have been linked with increased task difficulty. Given that P3 amplitudes become less positive with increased difficulty within the task, adolescents with gaming disorder may have more difficulty inhibiting their responses compared with controls (Li et al., Reference Li, Wang, Yang, Dai, Zheng, Sun and Liu2020). This finding, along with higher scores demonstrated on impulsivity scales, suggest that gaming disordered adolescents demonstrate abnormalities in inhibitory control systems (Li et al., Reference Li, Wang, Yang, Dai, Zheng, Sun and Liu2020). Adolescents with gaming disorder in reward tasks showed a less positive feedback-related negativity (FRN) amplitude compared with controls after receiving a gain on a gambling task (Li et al., Reference Li, Wang, Yang, Dai, Zheng, Sun and Liu2020). A blunted FRN response can be an indicator of reduced reward sensitivity (Li et al., Reference Li, Wang, Yang, Dai, Zheng, Sun and Liu2020). There was no difference between FRN responses on the gambling task after a loss, indicating while gaming-disordered adolescents may share similar loss avoidance, they may have increased reward-seeking behaviours (Li et al., Reference Li, Wang, Yang, Dai, Zheng, Sun and Liu2020).
EEG can be used to compare spontaneous electrical activity of the brains during task-related events (Light et al., Reference Light, Williams, Minow, Sprock, Rissling, Sharp, Swerdlow and Braff2010). One study examined qualitative EEG (QEEG) in adolescents with gaming disorder and comorbid ADHD, compared to adolescents with ADHD, and adolescent healthy controls (Park et al., Reference Park, Hong, Han, Min, Lee, Kee and Kim2017b). Adolescents with comorbid ADHD and gaming disorder showed increased theta coherence in the electrodes in the temporal region, suggesting an increased stimulation of the working memory circuit in gaming disorder. Increased intra-hemispheric coherence was noted in the right parieto-occipital region in the gaming disorder group compared to controls, implying a consistent gaming stimulus activates visuospatial working memory circuits (Park et al., Reference Park, Hong, Han, Min, Lee, Kee and Kim2017b). Increased beta cortical activity observed in this study has been compared to similar increases in beta cortical activity arising from methylphenidate usage in ADHD, implying that gaming may result in increased attentional capacity and a subconscious means of enhancing attention (Park et al., Reference Park, Hong, Han, Min, Lee, Kee and Kim2017b)
Functional Magnetic Resonance Imaging (fMRI)
Eighteen functional MRI (fMRI) studies compared resting-state brain activity between gaming-disordered samples and the non-disordered adolescent controls. fMRI uses changes in blood oxygen levels in the brain to map different brain regions, traditionally when performing a specific task such as tapping your finger (Lv et al., Reference Lv, Wang, Tong, Williams, Zaharchuk, Zeineh, Goldstein-Piekarski, Ball, Liao and Wintermark2018). Resting-state fMRI, which is fMRI conducted without the subject performing a task, uses spontaneous changes in blood oxygen levels to map the intrinsic brain activity (Lv et al., Reference Lv, Wang, Tong, Williams, Zaharchuk, Zeineh, Goldstein-Piekarski, Ball, Liao and Wintermark2018).. Taken together, adolescents with gaming disorder demonstrate brain changes comparable with substance addiction, differences in functional connectivity, emotional regulation, inhibition and self-identity compared to controls. Compared with healthy controls, one study demonstrated increased cerebral blood flow in the left inferior temporal lobe, left parahippocampal gyrus, amygdala, right medial frontal lobe/anterior cingulate cortex, bilateral insula, right-middle temporal gyrus, right precentral gyrus, left supplementary motor area, left cingulate gyrus and right inferior parietal lobe of adolescents with gaming disorder (Feng et al., Reference Feng, Chen, Sun, Zhou, Sun, Ding, Zhang, Zhuang, Xu and Du2013). These areas correlate with traditional substance addiction pathways, such as the involvement of the insula in urges to use substances, prefrontal cortex involvement in drug seeking behaviour and craving, anterior cingulate cortex activation in video game craving, and parietal cortex activity as an underlying cause of inhibition failure (Feng et al., Reference Feng, Chen, Sun, Zhou, Sun, Ding, Zhang, Zhuang, Xu and Du2013).
At rest, three studies demonstrated increased functional connectivity in the left superior frontal cortex, left middle frontal cortex and right inferior temporal cortex (Du et al., Reference Du, Yang, Gao, Qi, Du, Zhang, Li and Zhang2017b), including the left frontal eye field to the dorsal anterior cingulate and right interior insula in adolescents with gaming disorder (Han et al., Reference Han, Kim, Bae, Renshaw and Anderson2015). This increased functional connectivity within the dorsolateral prefrontal cortex (DLPFC), temporal lobe and left temporoparietal junction (TPJ) (Han et al., Reference Han, Kim, Bae, Renshaw and Anderson2015) may correlate with improved performance in visuospatial and visual attention processing (Du et al., Reference Du, Yang, Gao, Qi, Du, Zhang, Li and Zhang2017b). These findings may represent the risk of over-connectivity in response to increased amounts of game play and may impact executive control networks, in turn impacting mental health (Han et al., Reference Han, Kim, Bae, Renshaw and Anderson2015). This can be observed when measuring the connection efficiency between the dorsal striatum and left DLPFC between regular gamers and adolescents with gaming disorder (Chen et al., Reference Chen, Li, Zhang, Zhou, Wang, Tian and Xiang2021). In adolescents with gaming disorder, increased dorsal striatum activity suppressed the left DLPFC, as opposed to the activity of the dorsal striatum stimulating the left DLPFC (Chen et al., Reference Chen, Li, Zhang, Zhou, Wang, Tian and Xiang2021).
Compared with controls, two studies demonstrated that adolescents with gaming disorder were characterised by reduced functional connectivity between different brain regions and the dorsal striatum. Adolescents with gaming disorder demonstrated decreased functional connectivity between the right-middle frontal gyrus and the left cingulate gyrus, to the caudate (Han et al., Reference Han, Bae, Hong, Kim, Son and Renshaw2021), and decreased functional connectivity between the dorsal putamen and left insula-operculum (Hong et al., Reference Hong, Harrison, Dandash, Choi, Kim, Kim, Shim, Kim, Kim and Yi2015). Time spent gaming online was found to positively predict increased functional connectivity between the putamen and bilateral postcentral cortices, with control subjects showing significantly lower connectivity in the same brain regions (Hong et al., Reference Hong, Harrison, Dandash, Choi, Kim, Kim, Shim, Kim, Kim and Yi2015). Higher scores on the Young Internet Addiction Test (YIAT), used as a measure of gaming disorder in this study, correlated with greater functional connectivity between the dorsal putamen and left parahippocampal cortex in gaming-disordered adolescents (Hong et al., Reference Hong, Harrison, Dandash, Choi, Kim, Kim, Shim, Kim, Kim and Yi2015). Increased functional connectivity in gaming-disordered adolescents was observed between the posterior superior temporal sulcus (pSTS) and the posterior cingulate cortex (PCC) and anterior insular cortex (AIC) compared to controls (Lee et al., Reference Lee, Lee, Namkoong and Jung2020). These brain regions were used as proxy measures for the social brain network and social processing respectively, implying that gaming disorder may inhibit social brain functioning in adolescents (Lee et al., Reference Lee, Lee, Namkoong and Jung2020). Dysfunction in the social brain network, such as deficits in social cognition and mentalisation, is related to executive dysfunction and cognitive problems in adolescents (Lee et al., Reference Lee, Lee, Namkoong and Jung2020).
Changes in amplitude of low-frequency fluctuations (ALFF), a parameter that reflects the power of regional spontaneous neuronal activity (Zang et al., Reference Zang, He, Zhu, Cao, Sui, Liang, Tian, Jiang and Wang2007), were measured alongside functional connectivity in gaming disorder in adolescents pre- and post-receiving cognitive behavioural therapy (CBT) (Han et al., Reference Han, Wang, Jiang, Bao, Sun, Ding, Cao, Wu, Du and Zhou2018). The pre-treatment adolescent group showed significantly increased ALFF values in the bilateral putamen, right medial orbitofrontal cortex (OFC), bilateral supplementary motor area (SMA), left postcentral gyrus and left anterior cingulate (ACC) compared to healthy controls (Han et al., Reference Han, Wang, Jiang, Bao, Sun, Ding, Cao, Wu, Du and Zhou2018). Post-treatment with CBT, the ALFF values in the treatment group significantly decreased in the left superior OFC and the left putamen, and functional connectivity between these brain regions significantly increased post-CBT (Han et al., Reference Han, Wang, Jiang, Bao, Sun, Ding, Cao, Wu, Du and Zhou2018). Increased OFC activity is mirrored in another study examining functional ALFF between gaming-disordered adolescents and student pro-gamers, in addition to increased activity in the left subcallosal gyrus, left orbital and left inferior frontal gyrus. Both pro-gamers and gaming-disordered adolescents demonstrated increased parietal lobe activity and increased activity within the attention network of the brain, however pro-gamers improved from a behavioural and emotional perspective during the study, whereas the gaming-disordered adolescents showed no improvement (Kwak et al., Reference Kwak, Hwang, Kim and Han2020).
Two studies explored gaming disorder from an emotional regulation perspective. In a fMRI study examining swear word processing in adolescent gaming-disordered subjects, stronger activation was demonstrated in the right superior temporal gyrus compared to controls when using negative words (Chun et al., Reference Chun, Choi, Cho, Lee and Kim2015). There was less activation in the left inferior frontal gyrus and caudate nucleus compared to controls - areas related to language, emotional processing and automatic processing of swear words (Chun et al., Reference Chun, Choi, Cho, Lee and Kim2015). Control subjects elicited higher activation of the dorsal anterior cingulate gyrus (dACC) and right orbitofrontal cortex (rOFC) in response to swear words (Chun et al., Reference Chun, Choi, Cho, Lee and Kim2015). During a Stroop task, gaming-disordered adolescents showed weaker dACC activation and stronger insular activations compared to controls (Lee et al., Reference Lee, Lee, Chun, Cho, D-j and Jung2015). Combined with the weaker dACC involvement in the dorsal attention network, stronger activation of the right insula could imply that gaming-disordered adolescents were more distracted by the emotional interference of angry faces during the Stroop task (Lee et al., Reference Lee, Lee, Chun, Cho, D-j and Jung2015).
Compared to the weakened involvement and activation of the dACC in emotional regulation studies, the right ACC was found to be hyperactive in studies investigating impulse inhibition in adolescent gaming disorder (Ding et al., Reference Ding, Sun, Sun, Chen, Zhou, Zhuang, Li, Zhang, Xu and Du2014). Activation of right sided brain structures, such as the right ACC, is consistent with other studies related to gaming disorder. However, predominately left sided structures (left superior medial frontal gyrus, left inferior parietal lobe, left precentral gyrus, left precuneus, left cuneus) were found to be significantly hyperactive in impulse inhibition tasks in adolescent gaming disorder (Ding et al., Reference Ding, Sun, Sun, Chen, Zhou, Zhuang, Li, Zhang, Xu and Du2014). Interhemispheric connections between the right and left prefrontal lobes and orbitofrontal cortex (superior frontal gyrus, inferior frontal gyrus, superior frontal gyrus and middle frontal gyrus) were reduced compared to controls in adolescent gaming disorder (Wang et al., Reference Wang, Yin, Sun, Zhou, Chen, Ding, Wang, Li, Xu and Du2015). These findings suggest that the right hemisphere may be fully engaged, requiring the left hemisphere to play a role in response inhibition (Ding et al., Reference Ding, Sun, Sun, Chen, Zhou, Zhuang, Li, Zhang, Xu and Du2014). In addition, higher regional global efficiency in the fronto-sensorimotor, frontal-temporal, frontal-limbic and temporal region correlated with increased impulsivity in gaming disorder (Park et al., Reference Park, Chun, Cho, Jung, Choi and Kim2017a).
Lower activation of other frontal structures, such as the right DLPFC, demonstrated negative correlation with impulsivity scores during risky decision-making in adolescent gaming disorder (Qi et al., Reference Qi, Du, Yang, Du, Gao, Zhang, Qin, Li and Zhang2015). After a risky decision-making task, adolescents with gaming disorder showed an increased activation of prefrontal cortex structures (left inferior frontal cortex, bilateral ventromedial PFC) when experiencing negative feedback, suggesting increased resources were needed to evaluate risk values post-receiving negative feedback on a risky decision (Qi et al., Reference Qi, Yang, Dai, Gao, Du, Zhang, Du, Li and Zhang2016). This decreased activation of the right DLPFC has been observed in other substance misuse disorders and may reflect impaired executive control, leading to increased inhibition (Qi et al., Reference Qi, Du, Yang, Du, Gao, Zhang, Qin, Li and Zhang2015). While lower activation may be observed in the right DLPFC in risky decision-making, increased right sided connections between the right DLPFC to right TPJ, right auditory cortex to right motor cortex, right auditory cortex to SMA and right auditory cortex to dorsal anterior cingulate are associated with increased gameplay in adolescent gaming disorder (Han et al., Reference Han, Kim, Bae, Renshaw and Anderson2015).
The effects of relationships and identity in adolescent gaming disorder were the focus of two fMRI studies. While examining the interaction between family relationships and functional connectivity, adolescents with gaming disorder demonstrated decreased brain connectivity from the cingulate to the striatum (Hwang et al., Reference Hwang, Hong, Kim and Han2020). Positive correlations between functional connectivity values and Family Environmental Scale-R scores were identified between the left cingulate and the left and right lentiform nucleus, while functional connectivity values from the left cingulate to the left & right lentiform nucleus were negatively correlated with YIAS scores, a measure of internet addiction (Hwang et al., Reference Hwang, Hong, Kim and Han2020). In terms of self-identity, significant differences in brain region activation have been seen between healthy controls and adolescent gaming disorder when comparing thoughts about themselves versus thoughts about their game avatar (Choi et al., Reference Choi, Taylor, Hong, Kim, Kim, McIntyre and Yi2018). Adolescents with gaming disorder utilised the medial prefrontal cortex (MPC) and ACC when thinking about their game avatar, with even stronger activation of these regions compared to self-thoughts (Choi et al., Reference Choi, Taylor, Hong, Kim, Kim, McIntyre and Yi2018). These brain regions are linked with thoughts about the self and activated in healthy controls when thinking about aspects of their self, demonstrating that adolescents with gaming disorder identify more with their gaming avatar than their own self (Choi et al., Reference Choi, Taylor, Hong, Kim, Kim, McIntyre and Yi2018).
Structural Magnetic Resonance Imaging (MRI)
Three structural MRI studies were included in this review. Structural MRI measures differences in grey and white matter structures to provide information on microstructural changes in adolescents with gaming disorder. Using voxel-based morphometry (VBM) and tract-based spatial statistics (TBSS) to investigate microstructural changes, adolescent subjects showed significant grey matter atrophy in the right orbitofrontal cortex, bilateral insula and right SMA (Weng et al., Reference Weng, Qian, Fu, Lin, Han, Niu and Wang2013). Grey matter volumes of the right orbitofrontal cortex, bilateral insula and fractional anisotropy (FA) values of the right external capsule were positively correlated with increased YIAT scores (Weng et al., Reference Weng, Qian, Fu, Lin, Han, Niu and Wang2013).
Higher impulsivity scores were correlated with decreased grey matter volumes in similar brain regions (bilateral insula, orbitofrontal cortex), as well as the right dorsomedial prefrontal cortex (dmPFC), right amygdala and left fusiform gyrus (Du et al., Reference Du, Qi, Yang, Du, Gao, Zhang, Qin, Li and Zhang2016). In relation to white matter changes and impulsivity, gaming-disordered adolescents showed higher correlations between non-planning impulsiveness (NI) score and FA values of right corticospinal tract and right occipital white matter region (Du et al., Reference Du, Liu, Yang, Qi, Gao, Zhang, Zhu, Du, Dai and Li2017a). Other forms of impulsivity (attentional impulsiveness, motor impulsiveness) did not show any correlation between groups (Du et al., Reference Du, Liu, Yang, Qi, Gao, Zhang, Zhu, Du, Dai and Li2017a). These findings demonstrated dysfunction of these brain areas involved in behaviour inhibition, attention and emotion regulation might contribute to impulse control problems in gaming-disordered adolescents.
Proton Magnetic Resonance Spectroscopy (MRS)
Proton MRS is an imaging technique used to assess regional brain chemistry (Novotny et al., Reference Novotny, Ashwal and Shevell1998) and has been utilised in one study in adolescent gaming disorder. Adolescents with gaming disorder and attention-deficit hyperactivity disorder (ADHD) were examined using proton MRS and measured against both healthy controls and adolescents with ADHD (Bae et al., Reference Bae, Han, Kim, Shi and Renshaw2016). Levels of N-acetyl-aspartate (NAA) in both ADHD groups were lower compared to controls, with levels of glutamate and glutamine higher in the ADHD-only group compared to other two groups in the study (Bae et al., Reference Bae, Han, Kim, Shi and Renshaw2016). Decreased levels of NAA in both ADHD groups may suggest a common pathway of hypoactivity in the frontal lobe, with decreased levels of glutamate and glutamine in the gaming disorder and comorbid ADHD group (compared to ADHD-only group) possibly mediated by increased dopamine levels from engagement in video gaming (Bae et al., Reference Bae, Han, Kim, Shi and Renshaw2016).
Discussion
Examining the results from the four imaging modalities outlined, adolescents with gaming disorder demonstrate brain changes comparable with substance addiction, differences in emotional regulation, inhibition and self-identity, as well as some positive cognitive adaptations due to consistent gameplay. Adolescent gaming disorder demonstrated common findings in several brain regions associated with traditional substance addiction pathways. The insula, a brain region crucial in the conscious urge and the emotional connection towards substances in addiction (Naqvi and Bechara Reference Naqvi and Bechara2009), demonstrated reduced grey matter volume (Weng et al., Reference Weng, Qian, Fu, Lin, Han, Niu and Wang2013) and was associated with increased internet addiction scores (Weng et al., Reference Weng, Qian, Fu, Lin, Han, Niu and Wang2013) and emotional sensitivity (Lee et al., Reference Lee, Lee, Chun, Cho, D-j and Jung2015). Modifications in dorsal putamen connections have been hypothesised as a biological mechanism for moving from voluntary to habitual use in substance related disorders (Hong et al., Reference Hong, Harrison, Dandash, Choi, Kim, Kim, Shim, Kim, Kim and Yi2015). Reduced reward sensitivity and increased reward-seeking behaviour are cornerstones of dopamine dysfunction in substance misuse (Volkow et al., Reference Volkow, Wang, Fowler, Tomasi, Telang and Baler2010), with blunted FRN responses in task-based EEG studies demonstrating this finding in adolescent gaming disorder (Li et al., Reference Li, Wang, Yang, Dai, Zheng, Sun and Liu2020). Given the increased dopamine from increased engagement in video gaming (Bae et al., Reference Bae, Han, Kim, Shi and Renshaw2016), the common biochemical pathways with ADHD-only adolescents demonstrated during this review (Bae et al., Reference Bae, Han, Kim, Shi and Renshaw2016) and ADHD being 2–3 times more likely than the general population to develop a substance misuse disorder (Schellekens et al., Reference Schellekens, van den Brink, Kiefer and Goudriaan2020), this provides further neurological similarity between substance misuse disorders and adolescent gaming disorder. Finally, effective treatments for substance misuse, such as CBT, was noted to improve connectivity in the orbitofrontal cortex in adolescent gaming disorder (Han et al., Reference Han, Wang, Jiang, Bao, Sun, Ding, Cao, Wu, Du and Zhou2018). It is noteworthy that this is a region associated with compulsion and drive to take substances (Volkow and Fowler Reference Volkow and Fowler2000).
Adolescence is a period where there is a greater vulnerability to disruption of normal emotional development and a predisposition to strong emotional states (Dahl Reference Dahl2004). Previous studies have shown that adolescents with gaming disorder can present with dysfunctional emotional regulation strategies (Yen et al., Reference Yen, Yeh, Wang, Liu, Chen and Ko2018). In this review, adolescents with gaming disorder demonstrated higher activations of the dACC and OFC, two areas that play roles in cognitive control and emotional regulation, respectively, suggesting that gaming-disordered adolescents have less emotional sensitivity, less cognitive control (Chun et al., Reference Chun, Choi, Cho, Lee and Kim2015), and are more distracted by emotional interference compared to controls (Lee et al., Reference Lee, Lee, Chun, Cho, D-j and Jung2015). These findings suggest that adolescents, given their heightened vulnerability, may be susceptible to the emotional regulation sequalae of gaming disorder.
Adolescents demonstrate higher levels of risk taking and poor impulse control compared to adults, largely due to the slower development of brain regions involved in inhibition, such as the prefrontal cortex and anterior cingulate cortex (Blakemore and Robbins Reference Blakemore and Robbins2012). Adolescents with gaming disorder demonstrated increased activity in these regions and other frontal brain regions, even compared to their non-addicted peers, when attempting inhibition tasks (Şalvarli and Griffiths Reference Şalvarli and Griffiths2022). This higher activation is suggested to represent a failure to recruit pathways that could be crucial for cognitive control and response inhibition (Park et al., Reference Park, Chun, Cho, Jung, Choi and Kim2017a), such as when evaluating risks after negative feedback (Qi et al., Reference Qi, Yang, Dai, Gao, Du, Zhang, Du, Li and Zhang2016). With higher effort being placed in inhibition tasks, other brain structures not normally involved in inhibition are recruited (Ding et al., Reference Ding, Sun, Sun, Chen, Zhou, Zhuang, Li, Zhang, Xu and Du2014) and brain regions normally involved in inhibition can be suppressed (Chen et al., Reference Chen, Li, Zhang, Zhou, Wang, Tian and Xiang2021).
Self-identity development is an important stage of adolescence (Beyers and Çok Reference Beyers and Çok2008) and previous studies have examined whether or not increased internet usage is due to increasing internet addiction or self-exploration of adolescent identity and understanding of themselves (Israelashvili et al., Reference Israelashvili, Kim and Bukobza2012). Similar exploration and testing of alterative ideas and beliefs can be seen in adolescent gaming disorder (Young Reference Young2009). The MPC and ACC are important for adolescent self-evaluation (Pfeifer and Berkman Reference Pfeifer and Berkman2018) and these regions are active during gaming-disordered adolescent thoughts related to their gaming avatar, implying changes in self-identity (Choi et al., Reference Choi, Taylor, Hong, Kim, Kim, McIntyre and Yi2018). The family microsystem plays an important role in identity formation (Beyers and Çok Reference Beyers and Çok2008) with gaming-disordered adolescents showing disrupted family relationships, which was associated with the severity of the disorder and disconnectivity within the brain’s reward circuit (Hwang et al., Reference Hwang, Hong, Kim and Han2020). Alongside self-identity, adolescents with gaming disorder can demonstrate lower social competence (Torres-Rodríguez et al., Reference Torres-Rodríguez, Griffiths, Carbonell and Oberst2018), with aberrations in social brain functioning demonstrated in this review (Lee et al., Reference Lee, Lee, Namkoong and Jung2020).
Multiple studies demonstrated brain changes associated with positive adaptations to increased engagement in gaming. Increased functional connectivity in brain regions associated with visuospatial memory (Park et al., Reference Park, Hong, Han, Min, Lee, Kee and Kim2017b) and visuospatial attention processing (Du et al., Reference Du, Yang, Gao, Qi, Du, Zhang, Li and Zhang2017b) can have a positive effect on performance in gaming. Higher visuospatial processing has been demonstrated in studies recruiting naïve gamers to play shooter-type video games, with an associated improved accuracy of attention allocation (Granic et al., Reference Granic, Lobel and Engels2014). Increased connectivity between auditory circuits and brain regions (right DLPFC, right motor cortex, dorsal anterior cingulate) in adolescent gaming disorder demonstrate the adaptation of the salience network to process and integrate information more effectively from multiple senses during gameplay (Han et al., Reference Han, Kim, Bae, Renshaw and Anderson2015). Regular gamers of shooter-type games are more efficient at allocating their attentional resources during a pattern recognition challenge, demonstrating their ability to filter out irrelevant information more effectively compared to controls (Granic et al., Reference Granic, Lobel and Engels2014).
There are several limitations noted within this systematic review. First, many of the studies included in this review have small sample size and were conducted in Asian populations, likely given an increased focus on possible problematic gaming in Asian populations. This limits the ability to generalise to a worldwide population. The intended age group to review were those aged under 18, however some studies despite a mean age under 18 may have had a standard deviation exceeding 18 years old. Hence a small number of subjects in the review may exceed the 18-year-old age range. Within the papers selected for this review, there were differing definitions of gaming disorder, with this term used interchangeably with internet addiction, excessive gaming, online gaming disorder and problematic gaming. This is a known issue within gaming disorder research (Griffiths Reference Griffiths2018). In addition to differing definitions of gaming disorder, the majority of questionnaires used in this review measure internet addiction primarily and not specifically gaming disorder. These questionnaires were used alongside clinical evidence of gaming disorder to denote gaming disorder in these studies. Given the widespread nature of this method in these studies, this appears to be the accepted practice in researching gaming disorder in this population. Further research should focus on gaming-disorder specific questionnaires in order to decrease bias and confusion between what findings constitute internet addiction and/or gaming disorder. The studies included in the literature review are cross-sectional studies. Longitudinal studies would be required to establish causal relationships between altered brain structures and gaming disorder.
This review highlighted a number of key brain regions that can be affected in the adolescent brain due to gaming disorder. These findings can help clinicians understand adolescent presentations with gaming disorder and recognise important facets of clinical presentation from a neurobiological perspective. Referrals for gaming disorder treatment have been increasing internationally (King et al., Reference King, Achab, Higuchi, Bowden-Jones, Müller, Billieux, Starcevic, Saunders, Tam and Delfabbro2022) so it is important clinicians have a greater understanding of gaming disorder as a whole. Future studies should focus on forming a more robust neurobiological and clinical framework for adolescent gaming disorder.
Financial support
This research received no specific grant from any funding agency, commercial or not-for- profit sectors.
Competing interests
EK, DC, FMcN and NM have no conflicts of interest to disclose.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committee on human experimentation with the Helsinki Declaration of 1975, as revised in 2008. Ethical approval was not required for this study.
