Authors: Fan Mao, Zexin Liu, Qun Fang, Yike Liu
Categories: Systematic Review, Cognitive function, Game therapy, Exergaming, Digital game, ADHD
Source: BMC Psychiatry
Authors: Fan Mao, Zexin Liu, Qun Fang, Yike Liu
Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder characterized by cognitive deficits in areas such as attention, working memory, and executive function. Traditional treatments often address behavioral symptoms but may not fully target cognitive impairments. Game-based interventions (GBIs) have emerged as potential tools to enhance cognitive performance in children and adolescents with ADHD.
This systematic review and meta-analysis evaluated the efficacy of GBIs in improving cognitive outcomes in children and adolescents with ADHD. A comprehensive search of databases, including PubMed, Medline, PsycINFO, Web of Science, and Cochrane Library, identified 20 randomized controlled trials (RCTs) with 1,376 participants. Studies were included based on cognitive performance measures such as attention, metacognition, and executive function. Standardized Mean Differences (SMD) were calculated, and subgroup analyses were conducted based on intervention modality, session length, duration, and setting.
The meta-analysis revealed a moderate overall effect size of GBIs on cognitive performance (SMD = 0.42, 95% CI: 0.143 to 0.697, p < 0.01), with large heterogeneity (I^2^ = 86.6%). Significant improvements were observed in attention (SMD = 0.724), cognitive flexibility (SMD = 0.565), and working memory (SMD = 0.486). However, effects on inhibitory control and metacognition were less pronounced.
GBIs show moderate efficacy in enhancing cognitive functions, particularly attention, cognitive flexibility, and working memory, in children and adolescents with ADHD. However, these findings should be interpreted with caution due to the substantial heterogeneity across studies. Future research could prioritize standardized outcome measures and multicentre RCTs across diverse cultural contexts to strengthen the evidence base.
Not applicable.
Attention-Deficit/Hyperactivity Disorder (ADHD) is a persistent neurodevelopmental condition characterized by a chronic pattern of inattention, hyperactivity, and impulsivity that significantly impairs daily functioning across multiple domains [1]. It is one of the most prevalent childhood psychiatric disorders, affecting an estimated 5% of children and adolescents worldwide [2]. The impact of ADHD extends beyond the core symptoms, influencing a wide range of cognitive processes, including executive functioning, working memory, and attentional control [3, 4]. These cognitive deficits are often more debilitating than the behavioral symptoms, contributing to academic underachievement, social difficulties, and a diminished quality of life [5–7]. The cognitive impairments associated with ADHD can persist into adulthood, exacerbating the risk of comorbid psychiatric conditions such as anxiety, depression, and substance abuse disorders [8, 9]. These deficits are believed to stem from atypical neural development and dysfunctions in brain regions involved in cognitive control, particularly the prefrontal cortex and its associated networks [10, 11]. As such, ADHD is not merely a behavioral issue but a complex cognitive disorder with profound implications for an individual’s educational, occupational, and social outcomes [12].
Traditional interventions for ADHD, including pharmacotherapy and behavioral therapy, have been extensively studied and widely implemented. Pharmacotherapy, particularly the use of stimulant medications such as methylphenidate and amphetamines, has been shown to reduce core ADHD symptoms effectively [13]. However, while these medications can provide significant symptom relief, they often come with a range of potential side effects, such as appetite suppression, sleep disturbances, and increased cardiovascular risk, which can limit their long-term use and acceptance among patients and caregivers [14]. In contrast, behavioral therapy focuses on modifying environmental contingencies and teaching adaptive behavioral skill [15]. Common approaches include parent training, classroom-based behavioral interventions, contingency management, and cognitive behavioral therapy components [16]. Behavioral interventions can improve functional outcomes and reduce disruptive behaviors, especially when combined with medication, and they avoid pharmacological side effects [17]. However, they are resource-intensive, require trained personnel and sustained implementation, and show variable generalization to cognitive domains [18]. Consequently, there is a growing interest in exploring complementary therapeutic approaches that target the cognitive deficits in ADHD, with the aim of improving overall functioning and quality of life.
In recent years, game-based interventions (GBIs) have garnered increasing attention as a potential tool for enhancing cognitive performance in individuals with ADHD [19]. In this review, GBIs are characterized as structured interventions that incorporate explicit game design elements (e.g., goals, rules, feedback/rewards, progression and adaptive difficulty) [20]. Unlike traditional cognitive training, which relies on monotonous tasks, GBIs incorporate game mechanics such as adaptive difficulty, immediate feedback, goal-setting, rewards, and progress tracking to sustain user interest [21]. GBIs are typically digital—delivered via computer, tablet, or console—but may also include non-digital formats such as board-based cognitive games or active exergames combining physical movement with cognitive challenges [22, 23]. Examples include serious games targeting executive functions, commercial video games adapted for attention training, and structured exergaming platforms that train multitasking skills [24, 25]. Despite this diversity, GBIs share the core purpose of improving specific cognitive domains (e.g., attention, working memory, inhibitory control) in an enjoyable, motivating format. Currently, no universally accepted protocol exists for GBIs, and intervention designs vary in duration, setting, and supervision requirements.
A growing body of empirical research has explored the efficacy of GBIs for cognitive enhancement in ADHD. For example, research has demonstrated that a 20-week serious games intervention significantly improved overall EF performance and behavioral outcomes in children with ADHD [23]. Additionally, a 3-week video game training targeting specific cognitive functions resulted in improvements in children’s attention and fine motor skills [26]. GBIs are delivered in various formats, including computer games, exergaming, digital games, and targeted EF cognitive exercises. These diverse interventions address the limitations of pharmacological treatments, offering better accessibility, minimal side effects, and a lower risk of misuse. GBIs also raise important questions about feasibility. They are relatively easy to deliver at home or in schools, require minimal physical resources beyond digital access, and can be personalized through adaptive algorithms [27]. This may reduce barriers to participation, particularly for children who struggle with traditional interventions. However, implementation depends on access to digital devices, stable internet connections, and ongoing monitoring to ensure adherence and prevent excessive screen time [28]. These practical considerations must be weighed alongside efficacy when evaluating their clinical potential.
This systematic review and meta-analysis aim to provide a comprehensive evaluation of the efficacy of GBIs on cognitive performance in children and adolescents with ADHD. The primary objectives of this study are (1) to systematically review the current evidence on the impact of GBIs on cognitive outcomes in ADHD, (2) to quantify the overall effect size of GBIs through meta-analytic techniques, and (3) to identify potential moderators that may influence the effectiveness of these interventions, such as intervention duration, type of game, and participant characteristics. By synthesizing data from multiple studies, this review seeks to clarify the extent to which GBIs can enhance cognitive functions in ADHD and identify key factors that may optimize their implementation in clinical practice.
This systematic review and meta-analysis were conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [29], and the protocol for the systematic review was registered in the International Prospective Register of Systematic Reviews (PROSPERO; CRD42024575382). The aim was to evaluate the efficacy of GBIs on cognitive performance in children and adolescents diagnosed with ADHD. The methodology was designed to systematically collect, analyze, and synthesize data from existing studies to provide a comprehensive assessment of the impact of GBIs on cognitive outcomes, while also identifying potential moderators of intervention effectiveness.
A comprehensive search of electronic databases was conducted to identify relevant studies. The databases searched included PubMed, Medline, PsycINFO, Web of Science, and Cochrane Library. The search was conducted for studies published from January 2000 to October 2024. Keywords and MeSH terms related to “Attention-Deficit/Hyperactivity Disorder,” “GBIs,” “cognitive function,” “executive functions,” and “children or adolescents” were used to maximize the retrieval of relevant studies. The search strategy was developed in consultation with an experienced librarian to ensure thorough coverage of the literature. Additionally, reference lists of all included studies and relevant reviews were manually screened to identify any additional eligible studies. A snowball and citation search were conducted to identify relevant articles further. The subject term and free word search strategies were as (((“attention deficit hyperactivity disorder”[MeSH Terms]) OR (ADHD[Title/Abstract]) OR (“attention-deficit/hyperactivity disorder”[Title/Abstract])) AND ((“child”[MeSH Terms]) OR (child*[Title/Abstract]) OR (adolescent*[Title/Abstract]) OR (“young people”[Title/Abstract]))) AND ((“game-based”[Title/Abstract]) OR (gamified[Title/Abstract]) OR (“video game”[Title/Abstract]) OR (“computer game”[Title/Abstract]) OR (“serious game”[Title/Abstract]) OR (exergame[Title/Abstract]) OR (“active video game”[Title/Abstract])) AND ((“cognitive function”[Title/Abstract]) OR (“executive function”[Title/Abstract]) OR (“working memory”[Title/Abstract]) OR (“cognitive performance”[Title/Abstract]) OR (attention[Title/Abstract]) OR (“cognitive flexibility”[Title/Abstract]) OR (“inhibitory control”[Title/Abstract]) OR (metacognition[Title/Abstract])).
The inclusion criteria for this review were defined using the PICOS framework (Population, Intervention, Comparator, Outcomes, Study design). Population: Studies were included if they involved participants of children and adolescents aged 6–18 years diagnosed with ADHD, as determined by standard diagnostic criteria (e.g., DSM-IV or DSM-5, and ICD-10 or ICD-11). Studies that included participants with comorbid psychiatric conditions were excluded unless ADHD was the primary diagnosis. Intervention: The review focused on studies that implemented GBIs explicitly designed to improve cognitive functions. GBIs include several (1) EF-specific training games—targeted, task-based games designed to train executive functions (e.g., n-back, working memory games); (2) computerized cognitive training incorporating gamified elements but primarily drill-based (e.g., ACTIVATE™, Braingame Brain); (3) digital games—narrative or task-based therapeutic video games developed for clinical use (e.g., BrainFit); and (4) exergames—active video games combining physical activity with cognitive demands (e.g., Nintendo Wii, Xbox Kinect). To ensure clarity and isolate the specific effects of GBIs, we excluded studies in which GBIs were combined with other therapeutic approaches (e.g., pharmacotherapy, behavioral therapy, or strategy-based training). Comparator: Eligible studies included those that compared GBIs with either a control group (e.g., no intervention, placebo, or waitlist) or an alternative treatment (e.g., traditional cognitive training, standard ADHD medication). Outcomes: The primary outcomes of interest were cognitive performance measures, specifically executive functions, attention, and working memory. These outcomes needed to be assessed using standardized and validated cognitive assessment tools. Study Design: Only randomized controlled trials (RCTs) were included in this systematic review and meta-analysis due to their high level of evidence in evaluating the efficacy of interventions [30]. Non-randomized trials, observational studies, case studies, and other non-experimental designs were excluded.
Two independent reviewers (FM and YL) conducted the study selection process in two stages. First, titles and abstracts of all identified articles were screened for relevance based on the eligibility criteria. In the second stage, full-text articles of potentially relevant studies were retrieved and assessed for eligibility. Discrepancies between reviewers were resolved through discussion or by consulting a third reviewer (QF). A detailed record of the selection process, including reasons for exclusion at the full-text screening stage, was maintained.
Data extraction was performed independently by two reviewers (FM and YL) using a standardized data extraction form. Extracted data included study characteristics (e.g., author, year of publication), participant characteristics (e.g., age, sample size, ADHD diagnostic criteria), intervention details (e.g., type of game, setting, duration, frequency, session length), outcome measures (e.g., cognitive assessment tools used), and study design. For each study, effect sizes (e.g., standardized mean differences) were calculated for the primary outcomes of interest. If necessary, authors of the original studies were contacted for additional data or clarification. The extracted data were entered into a comprehensive database for analysis. Considering the different outcomes and measurement units used across the studies, Standardized Mean Differences (SMDs) were calculated based on the pre–post change scores between the intervention and control groups, and each SMD was weighted by its inverse variance.
The meta-analysis was conducted using the Comprehensive Meta Analysis (CMA) software 3.3 (BioStat Inc., Englewood, NJ, USA). The primary analysis focused on the overall effect size of GBIs on cognitive performance in children and adolescents with ADHD. After examining overall cognitive outcomes, we performed subgroup analyses based on the cognitive performance. Considering the different outcomes and measurement units used across the studies, SMDs were calculated based on the pre–post change scores between the intervention and control groups, and each SMD was weighted by its inverse variance. When only post-intervention data were reported, we computed between-group SMDs from post-intervention means and pooled standard deviations. The magnitude of the effect sizes was assessed using SMD, with SMD values of < 0.2, 0.2 ≤ SMD < 0.5, and 0.5 ≤ SMD < 0.8 indicating small, moderate, and large effect sizes, respectively [30]. A random-effects model was used to account for heterogeneity between studies. Heterogeneity was assessed using the I^2^ statistic, with values of 25%, 50%, and 75% representing small, medium, and high heterogeneity, respectively [31]. Publication bias was assessed using funnel plots and Egger’s test [32]. A two-tailed test with p-value less than 0.05 was considered significant publication bias.
Subgroup analyses were conducted by cognitive domain (cognitive flexibility, inhibitory control, working memory, attention, and metacognition) and by intervention characteristics (modality: computerized/EF-specific/digital/exergame; school/home/clinical; ≥ 6 weeks vs < 6 weeks; > 3/week vs ≤ 3/week; session > 40 min vs ≤ 40 min). To investigate the potential moderating effects, meta-regression was performed based on continuous variables, including age and intervention duration
The risk of bias in the included studies was assessed using the PEDro (Physiotherapy Evidence Database) scale, a widely recognized tool for evaluating the methodological quality of clinical trials [33]. The PEDro scale consists of 11 items that measure various aspects of trial quality, including eligibility criteria, random allocation, concealed allocation, baseline comparability, blinding (of subjects, therapists, and assessors), adequate follow-up, intention-to-treat analysis, between-group statistical comparisons, and measures of variability and point estimates [34]. Each study was independently evaluated by two reviewers according to the PEDro criteria. Disagreements between reviewers regarding the PEDro scores were resolved through discussion, and if necessary, a third reviewer was consulted to reach a consensus.
The initial database search yielded a total of 1470 records. After removing 774 duplicates, 696 records were screened based on titles and abstracts. Of these, 644 were excluded for not meeting the inclusion criteria, leaving 52 full-text articles to be assessed for eligibility. Following a thorough review, 32 studies were excluded for reasons such as lack of a control group (5), unmatched age group (3), not English articles (12), non-RCT design (7), absence of relevant cognitive performance outcomes (3), or insufficient data for analysis (2). Ultimately, a total of 20 RCTs were included in the systematic review and meta-analysis. A flow diagram of the study selection process is presented in Fig. 1.Fig. 1PRISMA flow diagram of the selection of studies
The main characteristics of the studies included are summarized in Table 1. The total sample included 1376 children and adolescents with ADHD with ages between 7 and 15 years. Altogether, 17 studies targeted participants with formal ADHD diagnoses, following the formal diagnostic criteria. Accordingly, the Diagnostic and Statistical Manual of Mental Disorders (DSM) (fourth or fifth edition) criteria were commonly utilized as the gold standard by clinicians or psychiatrists [53]. Specifically, five studies employed clinical settings, guided by experimenters or coaches who were familiar with the treatment approach, four studies were undertaken at school, and ten studies were applied at home. The GBIs were divided into four exergaming, computerized games, EF-specific games, and digital games. In general, four studies adopted exergaming training (i.e., Nintendo Wii Sport, Xbox Kinect), five studies selected computerized games as a treatment approach (i.e., ACTIVATE™, Braingame Brain), three studies used digital games training procedures (i.e., BrainFit), and eight studies employed EF-specific games intervention (i.e., working memory and n-back training).Table 1Characteristics of the studies included in this meta-analysisStudyParticipant descriptionADHD diagnosticDesignInstrumentCognitive variablesSettingsInterventionsMean age(years)NIntervention armIntervention nameControl armBenzing et al., 2018 [35]EG: 10.5CG: 10.5EG: 24CG: 22ICD-10RCTCSBT; Flanker taskCF; IC; WMHomeExergaming;15 minXBOX KinectWait-listBenzing et al., 2019 [36]EG: 10.5CG: 10.4EG: 28CG: 23ICD-10RCTCSBT; Simon task; FlankerCF; IC; WMHomeExergaming;30 min/session, 3 times/week, 6 weeksXBOX KinectRunningBikic et al., 2018 [25]EG: 9.8CG: 10.1EG: 35CG: 35DSM-4RCTAST; IED; SST; SWM; BRIEFCF; IC; WM; Attention; MetacognitionHomeComputerized games;40 min/session, 6 times/week, 8 weeksACTIVATE™Treatment as usualChang et al., 2022 [37]EG: 8.4CG: 8.4EG: 16CG: 16DSM-5RCTWCST; Stroop taskCF; ICSchoolExergaming;60 min/session, 3times/week, 12 weeksNintendo Wii SportWait-listJi et al., 2023 [38]EG: 9.0CG: 8.9EG: 16CG: 14K-ADHDDSRCTGNG; FAIRAttention; ICClinicalExergaming;50 min/session, 3 times/week, 4 weeksExerHeartBicycle exercisede Oliveira Rosa et al. 2021 [39]EG: 10.4CG: 10.9EG: 24CG: 21DSM-4RCTGNG; Flanker task; LSWMTCF; IC; WMHomeComputerized games;30 min/session; 4 times/week, 12 weeksACTIVATE™School workDovis et al., 2015 [24]EG: 10.6CG: 10.5EG: 31CG: 30DSM-4RCTTMT; CBTT; SCWTCF; IC; WMHomeComputerized games;30 min/session, 5 times/week, 5 weeksBraingame BrainWait-listEstrada-Plana et al., 2019 [40]EG: 9.5CG: 9.5EG: 13CG: 14N/ARCTGNG; CBTT; TMTCF; IC; WMClinicalComputerized games;50 min/session, 1 times/week, 5 weeksBoard gamesWait-listJohnstone et al., 2017 [41]7–13 yearsEG: 44CG: 41DSM-4RCTGNG; DSTIC; WMHomeEF-specific games;20 min/session, 3times/week, 8 weeksFocus PocusWait-listJones et al., 2020 [42]EG: 10.2CG: 10.1EG: 41CG: 39DSM-4RCTCPT; n-backIC; WMHomeEF-specific games;15 min/session, 5 times/week, 4 weeksN-backKnowledge practiceKhalili Kermani et al., 2016 [43]EG: 9.9CG: 9.8EG: 30CG: 30DSM-4RCTDSTWMHomeEF-specific games;60 min/session; 2times/week, 12 weeksWM Structured gamesWait-listKollins et al., 2020 [44]EG: 9.7CG: 9.6EG: 169CG: 160DSM-5RCTAPIAttentionHomeDigital games;25 min/session, 5 times/week, 4 weeksAkili Interactive LabsActive controlLan et al., 2020 [45]EG: 10.9CG: 10.2EG: 27CG: 28DSM-4RCTCPT; OST; WCSTCF; IC; WMN/AEF-specific games;60 min/session; 1time/week, 12 weeksSocial skill trainingWait-listPrins et al., 2011 [46]EG: 9.6CG: 9.3EG: 27CG: 24DSM-4RCTCBTTWMClinicalEF-specific games;35 min/session; 1time/week, 3 weeksComputerized WM GamesWM trainingRajabi et al., 2020 [47]EG: 10.2CG: 10.0EG: 16CG: 16DSM-5RCTIVAAttentionSchoolEF-specific games;45 min/session, 3times/week, 21 weeksSmartMindWait-listRodrigo-Yanguas et al., 2023 [48]EG: 14.7CG: 14.6EG: 35CG: 35N/ARCTBRIEFIC; CF; WMClinicalDigital game;25 min/session, 1time/week, 12 weeksThe Secret Trail of MoonPhone communicationsTucha et al., 2011 [49]EG: 10.8CG: 11.0EG: 16CG: 16DSM-4RCTVSI; ASTAttention; CFClinicalEF-specific game;45 min/session, 2times/week, 4 weeksAixTentVisual perception trainingvan der Donk et al., 2015 [50]EG: 9.8CG: 10.0EG: 50CG: 50DSM-4RCTDST; ISTIC; WM; MetacognitionSchoolComputerized game;45 min/session, 5times/week, 5 weeksCWMTPaying attention in classvan der Oord et al., 2014 [51]EG: 10.0CG: 9.5EG: 18CG: 22DSM-4RCTBRIEFCF; IC; WM; MetacognitionHomeEF-specific game;40 min/session, 5times/week, 5 weeksBraingame BrianWait-listZhao et al., 2024 [52]EG: 8.5CG: 8.3EG: 40CG: 40DSM-5RCTBRIEFMetacognitionSchoolDigital game;30 min/session, 3times/week, 4 weeksBrainFitWait-listK-SADS-PL, Kiddie-Schedule for Affective Disorders and Schizophrenia, Present and Lifetime version; DSM-4 and −5, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition and Fifth Edition; SNAP-4, Swanson Nolan and Pelham Rating Scale, Fourth Edition; ICD-10, International Classification of Diseases, Tenth Revision; CRS, Conner’s Rating Scale; CSB, color span backwards; GNG, Go-No-Go task; LNST: Letter–number-sequencing task; TMT = Trail Making Test; TOL: Tower of London; WCST: Wisconsin Card Sorting Test; DST, Digit span test; SCWT: Stroop Color and Word Test; STOPIT, Stop-Signal Inhibition Task; AWMA, Automated Working Memory Assessment System; ROCF, Rey Osterrieth Complex Figure; WRAML, Wide Range Assessment of Memory and Learning; EG, experimental group; CG, control group; NR, not reported; N/A, not available; K-ADHDDS: Korean Attention-Deficit/Hyperactivity Disorder Diagnostic Scale; API: Attention Performance Index; BRIEF: Behavior rating inventory of executive function; CWMT: Cogmed Working Memory Training; CF: cognitive flexibility; IC: inhibitory control; WM: working memory
The frequency of the intervention ranged from one to eight times per week, with each session lasting 15 to 60 min. Furthermore, only one studies focused on acute intervention with a duration 15 min, while 9 studies adopted a chronic intervention ranging from 6 to 21 weeks, with a total duration of 480 to 2835 min. The intervention protocols varied considerably, with the number of sessions ranging from 1 to 48 in total and the cumulative duration ranging from 15 to 2835 min across studies. The effects of GBIs on cognitive performance were assessed by 20 studies, while five studies assessed a single feature, five examined 2 features, eight studies evaluated 3 features, one examined 4 features, and last one study examined 5 features. Moreover, ten neurocognitive tasks were frequently used by researchers to assess cognitive performance in children and adolescents with ADHD. Further, the Digit Span Forward and Backward Test [54] was commonly used to measure working memory, and the Go-No Go task [55], Flanker Task [56], and Stroop Task [57] were frequently employed to evaluate inhibitory control. The Trail Making Task [58] and Wisconsin Card Sorting Test [59] were selected to measure cognitive flexibility. The Integrated Visual and Auditory Continuous Performance Test [60] and Frankfurter Aufmerksamkeits-Inventar Test [61] were commonly selected to evaluate attention. Lastly, the Behavior Rating Inventory of Executive Function [62] scale was used to assess metacognition.
The meta-analysis was conducted on 20 RCTs that reported on cognitive performance outcomes. Results are exhibited in Fig. 2. The overall effect size for the impact of GBIs on cognitive performance among children and adolescents with ADHD was significant, with a moderate SMD of 0.42 (95% CI 0.143 to 0.697, p < 0.01), with large heterogeneity (I^2^ = 86.6%, p < 0.001). Notably, four studies highlighted significant large training effects (SMD = 0.724, 95% CI 0.025 to 1.422, p < 0.05) on attention, with large heterogeneity (I^2^ = 88.3%, p < 0.001), and 11 studies concentrating on cognitive flexibility specified significant large effects (SMD = 0.565, 95% CI 0.098 to 1.033, p < 0.05), with large heterogeneity (I^2^ = 83.3%, p < 0.001). Fourteen studies on working memory revealed moderate significant effects of interventions (SMD = 0.486, 95% CI 0.076 to 0.895, p < 0.05), with large heterogeneity (I^2^ = 91.7%, p < 0.001). Lastly, 14 studies on inhibitory control shown small training effects (SMD = 0.178, 95% CI −0.232 to 0.588, p > 0.05) together with large heterogeneity (I^2^ = 84.7%, p < 0.001), and four studies on metacognition displayed small effects (SMD = 0.206, 95% CI −0.546 to 0.958, p > 0.05), with small heterogeneity (I^2^ = 39.8%, p = 0.17). The high heterogeneity observed across most cognitive domains is noteworthy. Several factors may explain this variability, including differences in GBI modalities, intervention length participant characteristics, and outcome measures used across studies. In addition, methodological variations such as sample size, control group selection, and trial quality may have contributed to the observed heterogeneity.Fig. 2Forest plot of effect of game-based interventions on cognitive performance
Table 2 recapitulates the results of subgroup analysis and meta-regression for cognitive performance. In the meta-regressions, age was negatively associated SMD (β = −0.142, 95%CI −0.27 to −0.02, p = 0.02). In the subgroup analyses, the effect of GBIs on cognitive performance was significantly moderated by intervention modality (p = 0.001), intervention setting (p < 0.05), and session length (p < 0.05), but not by intervention duration (p = 0.68) and frequency (p = 0.39). The EF-specific games (SMD = 0.943, 95% CI 0.60 to 1.29) produced significant large intervention effects for cognitive performance when compared to computerized games (SMD = 0.168, 95% CI −0.16 to 0.50), digital games (SMD = −0.278, 95% CI −0.86 to 0.31), and exergaming (SMD = 0.374, 95% CI 0.18 to 0.57). As for the intervention setting, school intervention produced significant moderate-to-large effects (SMD = 0.682, 95% CI 0.16 to 1.21) than clinical setting (SMD = −0.061, 95% CI: −0.49 to 0.37) and home setting (SMD = 0.449, 95% CI 0.18 to 0.71) on cognitive performance in children and adolescents with ADHD. With regard to the session length, > 40 min/session produced significant moderate-to-large effects (SMD = 0.709, 95% CI 0.38 to 1.04) than ≤40 min/session (SMD = 0.372, 95% CI 0.13 to 0.62).Table 2Moderator analysis of game-based interventions on cognitive performanceCategorical VariablesLevelNo. of studiesSMD95% CII^2^%Test of heterogeneityQ d.f.*p-*ValueModalityComputerized gamesDigital gamesEF-specific gamesExergaming17516100.168−0.2780.9430.374−0.16 to 0.50−0.86 to 0.310.60 to 1.290.18 to 0.5789.890.579.1016.64 30.001SettingClinicalHomeSchoolN/A112773−0.0610.4490.6821.009−0.49 to 0.370.18 to 0.710.16 to 1.210.68 to 1.3487.183.590.438.87.93 30.047Duration, week≥6 < 623250.4570.3690.15 to 0.760.07 to 0.6791.077.80.16 10.686Frequency, times/week > 3≤ 321270.3110.495−0.01 to 0.630.21 to 0.7885.487.70.72 10.395Session length, min > 40≤ 4017310.7090.3720.38 to 1.040.13 to 0.6290.874.012.81 10.002Continuous moderatorAge, yearsDuration, minsLevel7–1515–2835No. of studies4845β−0.1420.000195% CI−0.27 to −0.02−0.0002 to 0.0004I^2^%84.985.0Q d.f.5.14 10.60 1p-Value0.0230.440EF, executive function; N/A, not available
Table 3 depict the quality assessment for the included studies. Since all the studies were RCTs, the overall quality was high, with an average score of 7.7. All incorporated studies had clear recruitment criteria and maintained a high retention rate during the intervention. The intention-to-treat analysis further demonstrated that the participants data were analyzed according to their original assignment. However, only a few studies successfully blind participants and therapists, due to the challenges associated with executing double-blind procedures in non-pharmacological studies.Table 3Methodological quality assessment for included studiesStudyYearEligibilitycriteriaRandomallocationConcealedallocationSimilarbaselineBlindingsubjectsBlindingtherapistsBlindingassessors85%retentionIntentionto treatBetween-group comparisonsPoint andvariabilitymeasuresTotalscoreBenzing et al.2018YesYesYesYesNoNoNoYesYesYesYes8Benzing et al.2019YesYesYesYesNoNoNoYesYesYesYes8Bikic et al.2018YesYesYesYesYesNoNoYesYesYesYes9Chang et al.2022YesYesNoYesNoNoNoYesYesYesYes7Ji et al.2023YesYesNoYesNoNoNoYesYesYesYes7de Oliveira Rosa et al.2021YesYesNoYesYesNoNoYesYesYesYes8Dovis et al.2015YesYesYesYesYesNoNoYesYesYesYes9Estrada-Plana et al.2019YesYesNoYesNoNoNoYesYesYesYes7Johnstone et al.2017YesYesNoYesNoNoNoYesYesYesYes7Jones et al.2020YesYesYesYesNoNoNoYesYesYesYes8Kermani et al.2016YesYesNoYesNoNoNoNoYesYesYes6Kollin et al.2020YesYesYesYesYesNoNoYesYesYesYes9Lan et al.2020YesYesNoYesYesYesNoYesYesYesYes9Prins et al.2011YesYesNoYesNoNoNoYesYesYesYes7Rajabi et al.2020YesYesNoYesNoNoNoYesYesYesYes7Rodrigo-Yanguas et al.2023YesYesNoYesNoNoNoYesYesYesYes7Tucha et al.2011YesYesNoYesNoNoNoYesYesYesYes7van der Donk et al.2015YesYesNoYesYesNoYesYesYesYesYes9van der Oord et al.2014YesYesNoYesNoNoNoYesYesYesYes7Zhao et al.2024YesYesYesYesNoNoNoYesYesYesYes8Mean score7.7Methodological quality was assessed using the PEDro scale (0–10 points), which evaluates random allocation, concealed allocation, baseline comparability, blinding of subjects, therapists, and assessors, adequate follow-up, intention-to-treat analysis, between-group comparisons, and reporting of point estimates and variability. Total scores represent the sum of all “Yes” responses, with higher scores indicating higher methodological quality
Publication bias was evaluated using funnel plot asymmetry and Egger’s test [32]. The funnel plot (Fig. 3) showed slight asymmetry, suggesting the possibility of publication bias. However, Egger’s test did not indicate a significant bias (t = 1.30, df = 46, p = 0.20), implying that the observed asymmetry may be due to other factors, such as differences in study size or methodological heterogeneity.Fig. 3Funnel plot for visual inspection of publication bias
The results of this systematic review and meta-analysis provide comprehensive insights into the efficacy of GBIs on cognitive performance in children and adolescents with ADHD. The findings support the hypothesis that GBIs can improve key cognitive functions, including attention, working memory, and cognitive flexibility, which are commonly impaired in individuals with ADHD.
The present meta-analysis demonstrates that GBIs exert domain-specific effects, which may be explained by their inherent features. The relatively large improvement in attention (SMD = 0.72, p = 0.04) can be linked to the dynamic feedback, time pressure, and adaptive difficulty commonly embedded in games, which directly stimulate sustained and selective attention [63]. Similarly, the significant effects on cognitive flexibility (SMD = 0.56, p = 0.02) may reflect the frequent need in GBIs to switch between rules, tasks, or strategies, thereby promoting flexible thinking and task-switching abilities [64]. the moderate effect of GBIs on working memory (SMD = 0.48, p = 0.02) may be linked to the repetitive, structured, and goal-oriented activities in these interventions. Games often require players to hold task-relevant information (e.g., sequences, spatial locations, or rules) in mind while simultaneously manipulating new information, which aligns with the cognitive processes underlying working memory [65]. In contrast to the significant improvements observed in attention, working memory, and cognitive flexibility, the effects of GBIs on inhibitory control (SMD = 0.17, p = 0.39) and metacognition (SMD = 0.20, p = 0.59) were non-significant. One possible explanation is that inhibitory processes may be less susceptible to transfer effects from game-like environments, especially when the tasks emphasize engagement and reward rather than impulse suppression. Similarly, the findings for metacognition could be attributed to the limited number of available studies directly targeting self-monitoring, strategy use, or error awareness—core components of metacognitive functioning [66]. Collectively, these results suggest that the unique characteristics of GBIs—such as adaptivity, interactivity, motivational reinforcement, and strategic demands—are key to their cognitive benefits. This highlights the importance of tailoring intervention design to target specific domains.
When compared to traditional ADHD interventions, such as pharmacotherapy and behavioral therapy, GBIs present several advantages, particularly in terms of targeting cognitive deficits directly [67]. While stimulant medications like methylphenidate and amphetamines are effective in reducing core symptoms of ADHD, such as hyperactivity and impulsivity, their impact on cognitive functions is less pronounced (SMD = 0.26) [13, 68]. Furthermore, pharmacological treatments often come with side effects that can limit their long-term use, including appetite suppression, sleep disturbances, and increased cardiovascular risk [69].
On the other hand, behavioral therapies have consistently demonstrated effectiveness in managing ADHD-related behaviors and are widely recommended as part of standard care [70, 71]. They offer high levels of engagement by actively involving children and their caregivers in structured routines and goal-setting, which can enhance treatment adherence and motivation. Moreover, behavioral therapies can be tailored to individual needs and implemented across various settings, supporting long-term behavioral change and functional improvements. In contrast, GBIs provide a more engaging and interactive way to specifically target and improve cognitive functions, which may help complement the behavioral improvements achieved through traditional therapies. However, it is important to acknowledge that GBIs also have certain limitations, including heterogeneous intervention designs, variability in training content, and mixed evidence regarding their long-term transfer effects [44, 46]. These constraints highlight the need to use GBIs as complementary rather than stand-alone treatments within a broader, individualized care framework. Given the multifaceted nature of ADHD, a combination of pharmacotherapy, behavioral interventions, and cognitive training through GBIs may provide a comprehensive treatment plan [72].
The modality of game used in the intervention plays a crucial role in determining its efficacy. Our analysis showed that exergaming, EF-specific games, and computerized games all produced significant cognitive gains, but the most substantial improvements were observed in studies using EF-specific games. These games, designed explicitly to target executive functions like working memory and inhibitory control, may offer more focused and intensive than more generalized game formats [22].
The intervention setting emerged as one of the most critical moderating factors. Among the three primary settings—school, home, and clinical—the school setting demonstrated the most substantial effect on cognitive outcomes. GBIs conducted in school environments showed greater efficacy in improving executive functions, attention, and working memory, with effect sizes significantly larger than those seen in home or clinical settings (SMD = 0.682, 95% CI 0.16 to 1.21). This result aligns with findings from Bustamante et al. [73], who highlighted that the structured, routine nature of school environments fosters consistent participation and adherence, critical for maximizing cognitive gains. School settings may provide a conducive environment due to the combination of social support, peer interaction, and access to technology under the supervision of educators or facilitators, leading to enhanced motivation and engagement [74]. In contrast, GBIs implemented in home settings showed only moderate effects (SMD = 0.449, 95% CI 0.18 to 0.71). While home-based interventions offer greater flexibility and convenience, their effectiveness may be limited by several factors. First, the lack of structured schedules and external supervision can result in lower adherence and inconsistent engagement, which are crucial for cognitive improvement [75]. Moreover, children with ADHD may struggle to maintain focus and self-regulate during home-based training, especially without immediate feedback or reinforcement from instructors. However, home-based GBIs can still be beneficial when supported by parental involvement, gamified reward systems, and digital platforms that provide adaptive feedback and progress monitoring [76]. Clinical settings, by contrast, exhibited the weakest and non-significant effects on cognitive outcomes (SMD = −0.061, 95% CI −0.49 to 0.37). One possible explanation is that the clinical context might feel more formal or therapeutic rather than playful, potentially reducing intrinsic motivation and engagement, which are key mechanisms of GBIs [77]. Another consideration is that clinical populations often include children with more severe ADHD symptoms, who may require longer intervention periods or multimodal approaches to achieve measurable cognitive improvements [78]. These factors collectively may contribute to the lower observed effectiveness of clinical-based GBIs.
Session length also played a significant role in moderating the effectiveness of GBIs. Sessions lasting longer than 40 minutes produced better outcomes than those of shorter duration. Extended sessions allow for more in-depth engagement with the cognitive tasks embedded in the games, providing opportunities for skill refinement and feedback [79]. However, sessions that are too lengthy may lead to fatigue, especially for younger children with ADHD, which can diminish the effectiveness of the intervention [80]. In addition to session length, we also examined the overall length of the intervention protocols. The number of sessions varied considerably across studies, ranging from 1 to 48 sessions in total, with cumulative durations ranging from 15 to 2835 minutes. However, meta-regression analyses indicated that neither the total number of sessions nor the total duration significantly moderated the effects on cognitive outcomes. This suggests that the effectiveness of GBIs may depend more on the quality and content of individual sessions than on the sheer amount of training time. Therefore, finding an optimal balance between sufficient cognitive stimulation within each session and avoiding cognitive overload across the intervention as a whole is essential for maximizing the benefits of GBIs.
Finally, age was another important moderator influencing GBI outcomes. Our meta-regression analysis indicated that younger participants experienced greater cognitive improvements compared to older adolescents. This finding suggests that earlier intervention may be more beneficial, as younger brains exhibit higher neuroplasticity, allowing for more effective cognitive training [4]. Additionally, younger children may be more receptive to the game-based format, viewing it as a natural extension of play rather than a structured intervention [28]. This is consistent with the broader literature on neurodevelopmental disorders, where earlier interventions often yield better long-term outcomes [12]. Moreover, younger participants may display lower baseline symptom severity or more malleable behavioral profiles, which could enhance their responsiveness to intervention compared to older adolescents who often present with more entrenched symptoms and comorbidities [81]. Future studies could examine the role of baseline symptom severity and developmental stage to clarify the mechanisms underlying age-related differences in GBI effectiveness.
Another important but underexplored moderator is cultural context. Most included studies were conducted in European countries or Asian regions, where cultural practices, educational systems, and healthcare infrastructures may shape both intervention implementation and participant engagement. For example, school-based GBIs may be more effective in cultures with structured educational settings and strong teacher involvement. Conversely, in other cultural environments with limited resources or different attitudes toward gaming, outcomes may differ significantly. Future multicentre research across diverse cultural settings is needed to validate the generalizability of our findings.
The findings of this review have important implications for clinical practice. GBIs offer a promising, scalable, and engaging approach for children and adolescents with ADHD. Given their accessibility and appeal, GBIs could be integrated into school-based or home-based treatment programs, providing an adjunct to more traditional therapeutic approaches. However, clinicians should consider the individual needs of each child, including their age, baseline cognitive functioning, and preferences, when selecting an appropriate intervention.
Future research should continue to investigate the mechanisms underlying the effectiveness of GBIs, particularly in relation to neuroplasticity and the specific brain regions targeted by different types of games. Functional neuroimaging studies, such as Functional Magnetic Resonance Imaging (fMRI) and Electroencephalogram (EEG), could provide valuable insights into how GBIs influence neural functioning and whether these changes correspond to cognitive improvements [82]. Additionally, research should explore the potential for integrating GBIs with other therapeutic modalities to create comprehensive treatment plans that address both the cognitive and behavioral dimensions of ADHD.
This systematic review and meta-analysis possess several strengths that enhance the reliability and relevance of the findings. First, by exclusively including RCTs, which are considered the gold standard in intervention research, the study ensures a high level of internal validity, thereby minimizing bias and enhancing the robustness of the causal inferences drawn [83]. Another strength is the full consideration of various moderators such as intervention modality, session length, age, duration, frequency and intervention setting. This meta-analysis identifies key factors that influence the effectiveness of GBIs. However, some limitations must be acknowledged. First, while the meta-analysis included a substantial number of studies, there was significant heterogeneity across the included studies, particularly in terms of intervention design, outcome measures, and sample characteristics. The use of different cognitive assessment tools across studies may have contributed to variability in the results, making direct comparisons challenging. Future research should aim to standardize cognitive outcome measures to facilitate more accurate comparisons across studies. The second limitation is the insufficient number of studies on the cognitive outcomes of attention and metacognition, which need to be further studied. Lastly, only one of the included studies involved acute exercise intervention, which may have influenced the cognitive outcomes. Future research should emphasize the impact of acute GBIs on cognitive performance in children and adolescents with ADHD. In addition, most of the included studies did not directly examine neural or neurophysiological changes. This lack of direct evidence limits our ability to support neuroplasticity as the underlying mechanism, and future research should incorporate neuroimaging or electrophysiological measures to validate this assumption.
This systematic review and meta-analysis provide evidence that GBIs can improve cognitive performance in children and adolescents with ADHD, particularly in domains such as attention and executive functioning. However, the observed effects varied considerably across studies, depending on factors such as the specific modality used, the intervention environment, and the duration of training. Given this variability, GBIs should be considered as a complementary approach to conventional treatments. To strengthen the evidence base, future research should focus on developing standardized protocols and conducting more homogeneous, large-scale multicentre trials to clarify which intervention features produce more robust and generalizable cognitive benefits.