Data Modeling

1. Introduction

There is something quietly urgent about children's mental health that tends to get lost amid more visible public health priorities. A child who struggles silently with anxiety or depression rarely produces the kind of dramatic, attention-grabbing statistic that mobilizes funding or policy — and yet the consequences, left unaddressed, tend to compound over a lifetime, shaping social relationships, academic trajectories, and, eventually, a country's broader productivity and economic stability. Mentally healthy children, by most accounts, go on to build steadier families, do better in school, and participate more fully in their communities. The reverse is also true, if less often said aloud: family strain, difficult living environments, and unsupportive educational settings can leave psychiatric footprints that persist well into adulthood, and a great many of these cases are never formally recognized, let alone treated.

Bangladesh illustrates this gap starkly. Despite what appears to be a substantial burden of childhood mental health difficulty, the issue remains, for lack of a better phrase, under-prioritized nationally. Meanwhile, a number of higher-income countries have taken a rather different path, actively researching and building games around secific mental health conditions, to the point that psychiatrists in some of these settings have begun incorporating such games into standard treatment protocols. That contrast — games as an established clinical tool in one context, and mental health services that are simply hard to reach in another — is really the starting point for this review.

It would be a mistake, of course, to treat games as uniformly beneficial. Whether playing games helps or harms a child's mental state seems to depend heavily on how much time is spent playing and, perhaps more importantly, on what is actually being played. This ambiguity is precisely why a growing body of researchers has begun asking a more targeted question: not simply whether games are good or bad, but whether particular games, designed or repurposed with specific therapeutic aims, can meaningfully treat particular mental health conditions.

Video games, whatever their reputation, are cognitively demanding in ways that are easy to underestimate. Titles built around strategy, memory, or problem-solving exercise the same mental faculties that more traditional “brain training” activities target, and there is a reasonable case — even if it sounds almost too simple to be true — that regularly playing a handful of such games could help keep the mind active and, over time, guard somewhat against cognitive decline. Chess remains perhaps the clearest example: its demands on strategic foresight are well documented, and regular practice appears to sharpen exactly the kind of thinking it draws on. Sudoku occupies similar territory today, requiring sustained logical reasoning without necessarily looking like “therapy” to the person playing it. Interestingly, some online multiplayer games seem to serve an additional, somewhat unexpected function: they offer a low-pressure social channel for children who struggle with social anxiety, a role quite different from the purely cognitive exercise that games like chess and Sudoku provide.

Several converging social trends appear to be driving a rise in childhood mental health difficulty more broadly — the growth of nuclear (as opposed to extended) family structures, both parents working outside the home, and, somewhat more diffusely, fewer opportunities for children to simply play or talk things through with peers. In Bangladesh specifically, awareness of children's mental health needs remains limited, particularly outside urban centers, and treatment or guidance services are concentrated heavily in Dhaka, leaving rural families with comparatively few options. The net effect is a substantial number of children whose difficulties go unaddressed not for lack of concern, necessarily, but for lack of accessible pathways to help.

This review is organized around a fairly practical aim: to introduce parents, educators, and clinicians to applied games as a treatment avenue that does double duty — offering genuine entertainment value while also engaging a child's capacity to think, strategize, and recall information. One appeal of this approach is that it is not narrowly therapeutic; games that support mental health treatment in clinical populations can, in principle, also benefit children without diagnosed conditions who simply want to keep their minds sharp and, perhaps, reduce their own future risk. By systematically reviewing the existing literature on games and mental health treatment, we hope to make it somewhat easier for readers to identify which specific games might suit which specific needs.

A reasonably diverse literature has begun to accumulate here, spanning several distinct game formats. Tablet-based cognitive and socio-emotional games have been evaluated in resource-constrained educational settings: Rojas-Barahona et al. (2022), for instance, tested a tablet game across sixteen schools serving socioeconomically disadvantaged preschoolers in Chile, assessing outcomes across early literacy, pre-numeracy, cognitive skills, socio-emotional recognition, and mental health symptoms. Narrative-based emotion-regulation games have also drawn attention: David and Magurean (2022) evaluated the REThink game, in which children aged 10 to 16 practiced rapidly identifying happy faces among angry ones, and found modestly faster recovery from dysfunctional negative emotions relative to comparison conditions. Kowal et al. (2021) took a broader, narrative-review approach, surveying commercial titles — Portal 2, World of Warcraft, Pokémon, Final Fantasy, Mario Kart, Candy Crush, Angry Birds, and Boson X among them — for their effects on depressive symptoms and cognitive outcomes, while noting that The Magic Marker, MindLight, and Max appeared specifically to reduce social anxiety.

Physically active games have their own evidence base. Santos et al. (2021) synthesized 17 reviews and meta-analyses examining active video games (AVGs) such as the Wii Balance, Aerobics, Boxing, Golf, Tennis, and Basketball titles, reporting benefits not only for physical fitness — energy expenditure, oxygen intake, heart rate — but also for socializing and self-esteem. Virtual reality has been evaluated somewhat more cautiously: Halldorsson et al. (2021) reviewed VR and applied-game interventions for children and adolescents under 18, finding that applied games generally outperformed VR specifically for anxiety symptoms, with VR showing comparatively limited effect on anxiety and fear outcomes. Head-to-head comparisons with established therapy exist as well: Schoneveld et al. (2020) compared cognitive behavioral therapy against the applied game MindLight over a six-week course, finding CBT somewhat stronger for externalizing symptoms (aggression, hyperactivity, concentration difficulty) but MindLight roughly comparable for internalizing symptoms such as anxiety and low self-efficacy. Kuzu and Durna (2020) conducted a meta-analysis spanning ages 7 to 80, ultimately recommending SPARX for depression, Tetris for post-traumatic stress symptoms, Let's Face It! and the Junior Detective Training Program for autism spectrum disorder, Braingame Brian for ADHD, Lumosity and path-finding tasks for general cognitive functioning, and Guardian Angel for alcohol use disorder. A broader industry perspective, offered by the Financial Times (2019), concluded that exergames suit behavioral activation and social engagement, VR suits exposure-based therapy, CBT- and positive-psychology-based serious games suit structured psychoeducation, and biofeedback games suit relaxation training.

Taken together, this literature suggests that applied games occupy a genuinely distinct niche relative to both conventional entertainment gaming and conventional clinical treatment — neither purely recreational nor purely medical, but something in between that borrows credibility and evidence from both. What remains less clear, and what this review sets out to examine more systematically, is which specific game formats work best for which specific conditions, how strong the underlying evidence actually is, and what a scalable, low-resource-appropriate application of this research might look like — a question of particular relevance in settings such as Bangladesh, where conventional mental health infrastructure remains thin.

2. Methods

To identify the evidence base underlying applied games for children's mental health treatment, we conducted a literature search designed to be reproducible by an independent research team applying the same search strategy and eligibility criteria described below. The overall study-selection process is summarized in Figure 1.

2.2 Search Strategy

Searches were conducted primarily through Google Scholar, restricted to the publication window between 2012 and 2022, using the following search terms, applied both individually and in combination: “Mental health,” “Children mental health,” “Games for mental health problems in children,” and “Games for treatment of mental health problems in children.” Once an initial set of relevant papers had been identified, we conducted targeted follow-up searches using the specific names of games mentioned in those papers, in order to capture additional evaluations of the same interventions that the initial keyword search might have missed. We further hand-searched existing reviews, meta-analyses, and early-phase intervention trials relevant to applied games and children's mental health to supplement the primary database search.

2.3 Eligibility Criteria

Records were included in the review if they met all of the following criteria: (a) the study employed a randomized design; (b) the study examined the effectiveness of a game intended to treat a mental health problem; (c) participants were children or adolescents; (d) the intervention was delivered in a technology-based (digital) format rather than a traditional board or physical game; and (e) the study had been published in a peer-reviewed or otherwise verifiable outlet. Studies that used a game merely as one incidental component of a broader digital CBT program, without treating the game itself as the active therapeutic ingredient, were excluded, as were non-randomized case reports and studies involving exclusively adult populations.

2.4 Study Selection and Screening

The initial search yielded 209 records. Titles and abstracts

Table 1. Summary of representative applied games identified in the review, their primary target mental health condition(s), a brief description of gameplay mechanics, and the therapeutic outcomes reported in the associated studies. Games are grouped loosely by mechanism (narrative CBT-based, biofeedback-based, and virtual-reality-based) to support comparison of therapeutic approach across conditions such as depression, anxiety, ADHD, PTSD, and phobia.

were screened to remove duplicate records and items clearly unrelated to applied games or children's mental health, after which 35 full-text articles were retrieved and assessed against the eligibility criteria above. All 35 of these full-text articles satisfied the inclusion criteria and were retained for narrative synthesis; none were excluded at the full-text stage, reflecting the relatively targeted nature of the keyword and follow-up search strategy. The complete flow of records through identification, screening, and inclusion is depicted in (Figure 1).

2.5 Data Extraction and Synthesis Approach

For each included study, we extracted the game or intervention name, the target mental health condition, the study population and age range, the comparison condition (where applicable), the outcome measures used, and the direction and approximate magnitude of reported effects. Because the included studies varied substantially in design — spanning randomized controlled trials, pilot studies, and reviews of primary trials — we synthesized findings narratively rather than pooling effect sizes statistically, grouping interventions by broad mechanism of action (biofeedback-based, cognitive-training, exergame, virtual reality, and narrative CBT-based games) to allow comparison across conceptually related approaches. A summary of representative games, their target conditions, and their therapeutic mechanisms is provided in Table 1.

3. Results

3.1 Study Selection Overview

The search and screening process, summarized in (Figure 1), yielded 35 eligible studies spanning biofeedback games, cognitive-training games, exergames, virtual reality interventions, and narrative CBT-based serious games. Eight of the included studies specifically examined young people's subjective experiences with these games and their perceived usefulness for addressing mental health concerns, considering a range of applicable titles rather than a single intervention.

3.2 Biofeedback-Based Games

Two biofeedback games recurred across the included studies: Journey to the Wild Divine and Freeze-Framer 2.0, both of which pair relaxation training with real-time visual feedback tied to physiological signals. Children using these interventions showed significantly reduced despair and anxiety relative to a waitlist control group (Knox et al., 2011), and self-report and trait anxiety scores were likewise considerably lower after play compared with waitlist conditions (Santos et al., 2021); a combined approach pairing in-person CBT with a two-electrode biofeedback game additionally improved state anxiety and self-esteem (Santos et al., 2021). By contrast, Max and the Magic Marker, MindLight (delivered via a one-channel dry-sensor EEG headset that modulates an in-game avatar's light output as players navigate a haunted house), and an attention-bias-modification-training (ABMT) game teaching players to orient toward happy rather than unhappy faces did not reduce anxiety over time relative to comparison conditions (Schoneveld et al., 2016).

A related biofeedback title, Dojo, uses heart-rate-variability sensors affixed to the fingertips to scale the difficulty of emotional puzzles following brief training in deep breathing, progressive muscle relaxation, positive thinking, and guided imagery. Dojo did not outperform a non-therapeutic comparison game (Rayman 2) on overall anxiety reduction, though adolescents playing Dojo did show a steeper decline in specific anxiety symptoms (Scholten et al., 2016). PlayMancer, a biofeedback game built around heart rate, galvanic skin response, and breathing frequency, was evaluated as an adjunct to CBT for impulse-related disorders; when combined with CBT for bulimia nervosa and for gambling problems, PlayMancer use was associated with improved emotional control and reduced impulsivity on standardized measures (Tàrrega et al., 2015).

3.3 Cognitive-Training Games

Three trials of cognitive-training games involving structured number- and letter-sequencing tasks were identified, collectively enrolling 31 children with cognitive impairment (Rojas-Barahona et al., 2022; David & Magurean, 2022; Schoneveld et al., 2020). These interventions performed well on cognitive outcome measures, though their direct effect on mood was not systematically examined in any of the three trials.

3.4 Exergames

Four studies examined exergames — titles requiring physical movement — for their effects on depression and anxiety, predominantly in younger populations (Santos et al., 2021; Kowal et al., 2021; Halldorsson et al., 2021; Kuzu & Durna, 2020). Some of this evidence drew on repurposed commercial entertainment titles rather than purpose-built therapeutic games, and one purpose-built exergame produced a substantial reduction in depressive symptoms, with larger effects observed for more entertaining titles than for those with comparatively minimal gaming elements (Financial Times, 2019). These findings warrant some caution, however, since the underlying randomized trials generally relied on small sample sizes.

3.5 Virtual and Augmented Reality Interventions

Virtual reality (VR) and augmented reality (AR) interventions rely on audible, visual, and occasionally other sensory stimuli delivered within an artificial environment to support engagement and, potentially, therapeutic change (Halldorsson et al., 2021; Kuzu & Durna, 2020; Financial Times, 2019). Three VR-based gaming therapies were identified in one review, each reporting favorable intervention effects (Financial Times, 2019), though as detailed further in the Discussion, VR outcomes overall were more modest than those reported for applied games targeting the same conditions.

3.6 Narrative CBT-Based Serious Games

Two further interventions of note emerged from the reviews included here. Several programs adopted a tiered CBT structure delivered within a fantastical narrative, designed to be completed at roughly one level per week on a computer (Fleming et al., 2017); nearly every one of these programs reported favorable or encouraging outcomes, with the exception of a single game with conflicting results that was subsequently withdrawn from further development. SPARX, the most extensively evaluated of these programs, is summarized in detail in (Table 1). SuperBetter, a more recently evaluated example, takes a distinctly gamified approach built around experience points and “leveling up” as users complete brief, frequent tasks throughout the day — a usage pattern common in contemporary consumer apps but, notably, still uncommon among clinically oriented mental health tools. In a randomized controlled trial in which participants played SuperBetter for ten minutes daily over 30 days, participants showed substantially greater reductions in anxiety and depression symptoms than comparison participants (Roepke et al., 2015).

3.7 Entertainment Games and Single-Mechanism Interventions

A separate cluster of studies examined ordinary entertainment titles rather than purpose-built therapeutic games. In one study, students completed a frustrating assignment before being exposed to either a control condition or 45 minutes of a violent video game; those in the violent-game condition reported fewer depressive symptoms immediately afterward (Ferguson & Rueda, 2010), although a related follow-up study found no comparable effect on depression. Separately, mood improved rapidly following short sessions of the casual puzzle game Bejeweled II, with proposed mechanisms including emotional regulation, stress release, and incidental social support (Rossoniello et al., 2009). Journey to the Wild Divine has additionally been tested specifically in children with ADHD, with the experimental group showing fewer ADHD-related behaviors after biofeedback-guided breathing exercises tied to in-game visual and musical rewards (Amon & Campbell, 2008); most ADHD-focused therapeutic games described elsewhere in the literature instead emphasize cognitive remediation and executive-function training. Finally, Tetris — an otherwise unremarkable puzzle title in which players rotate and drop falling tetrominoes to clear horizontal lines — has been studied for its potential to disrupt memory consolidation when played shortly after a triggering event, with preliminary evidence suggesting reduced traumatic flashbacks in PTSD and possible reductions in craving intensity (Holmes et al., 2009).

4. Discussion

4.1 Principal Findings

This review identified 35 eligible studies examining applied games and virtual reality interventions for children's and adolescents' mental health, and a few patterns stand out clearly enough to be worth stating plainly, even while acknowledging that this remains, in a fairly literal sense, an early-stage field. Depression appears to be the condition for which applied games currently have the strongest evidentiary footing: several studies reported meaningful effect sizes for symptom change and remission, and support for SPARX in particular was considerably more consistent than for any other single game (Table 1). Remission rates among adolescents playing SPARX were notably higher than among waitlist controls, and outcomes were broadly comparable to face-to-face counseling, with both approaches producing medium-to-substantial within-group effects roughly in line with those reported for established face-to-face psychological treatments of adolescent depression.

Anxiety outcomes tell a somewhat more tempered story. Pre-post effect sizes in this domain tended to fall in the small-to-medium range, and — perhaps more tellingly — comparisons against non-therapeutic comparison games (as in the Dojo-versus-Rayman 2 trial) frequently failed to show a clear therapeutic advantage for the purpose-built intervention (Scholten et al., 2016). VR interventions followed a related pattern: children and adolescents appeared to tolerate VR settings reasonably well and to remain engaged with treatment, but only two of the three identified VR trials reported symptom outcomes at all, and those that did showed only modest-to-moderate within-group improvement in fear and anxiety symptoms, without statistically outperforming waitlist conditions despite admittedly small trial sizes (Halldorsson et al., 2021).

4.2 Strengths and Methodological Limitations of the Review

This review's principal strengths lie in its approach to identifying and synthesizing evidence specifically on VR and applied-game interventions for children's and young people's mental health, and in its attempt to extract and compare effect-size information across a heterogeneous evidence base rather than simply cataloguing study titles. That said, several limitations temper how confidently these findings should be interpreted. First, effect-size estimates across the included primary studies generally assumed statistical independence between pre- and post-intervention scores, an assumption that, if violated (as is plausible in small, repeated-measures samples), could inflate apparent treatment effects. Second, because included studies tended to assess broad mental health outcomes rather than narrow, disorder-specific symptom clusters, we are not well positioned to draw conclusions about how these games perform against more granular clinical targets. Third, our eligibility criteria excluded studies in which a game functioned merely as a component of a broader digital CBT program rather than as the primary active ingredient, which likely excluded some relevant evidence, if for defensible methodological reasons. Fourth, while we drew on quality-assessment information reported in the included reviews, many underlying primary studies were case studies for which several standard quality criteria did not meaningfully apply and were therefore excluded from summary scoring — meaning that reported quality ratings should be read as “acceptable-to-good relative to study type” rather than as an absolute methodological benchmark. Finally, inconsistent reporting of the specific game or VR elements used in many primary studies raises the possibility that additional relevant trials exist but were not captured by our search strategy, which constrains the generalizability of the conclusions drawn here.

4.3 Practical Implications for Low-Resource Settings

For settings such as Bangladesh, where formal mental health infrastructure for children remains thin outside major urban centers, the practical appeal of applied games lies less in whether they can match the ceiling performance of face-to-face treatment and more in whether they can meaningfully extend the reach of mental health support to children who currently receive none at all. Given that games such as SPARX have shown outcomes broadly comparable to face-to-face counseling for depression (Table 1), and given the low marginal cost of distributing a digital intervention relative to training and deploying clinical staff, applied games appear to represent a genuinely promising, if still early-stage, complement to conventional treatment pathways in resource-constrained contexts.

4.4 Future Directions and Planned Application Development

Building directly on the evidence synthesized here, we plan to develop a mobile application that first determines, through a structured series of screening questions, whether a user may be experiencing a specific mental health difficulty, and then recommends a corresponding evidence-supported game as a treatment option (Figure 2). The application is envisioned around two core functions: a mental-health-analyzer component, in which users respond to a validated set of screening questions that are transmitted to a cloud-based scoring engine, and a recommendation component, which returns a specific suggested game matched to the identified concern. Beyond this immediate application, future research priorities should include independent, adequately powered randomized trials directly comparing game-based and non-game-based treatment options across diverse user groups; closer attention to smartphone-native serious games, which remain comparatively understudied relative to desktop and console titles; and further investigation of socially connected, multiplayer applied games, whose peer-engagement dimension has received little evaluation to date despite its evident relevance to the social-isolation concerns motivating much of this research.

Figure 1. PRISMA-style study selection flow chart. Diagram summarizes the identification of 209 initial records through Google Scholar and hand-searches, screening for duplicates and relevance, full-text assessment of 35 articles against eligibility criteria (randomized design, technology-based game intervention, child/adolescent participants, published outcome), and final inclusion of all 35 studies in the narrative synthesis.

Figure 2. Conceptual design of the proposed mental-health-screening and game-recommendation application. The diagram illustrates the two planned functional components — a cloud-based mental-health-analyzer questionnaire and a corresponding game-recommendation engine — through which users would be screened for potential mental health concerns and matched to an evidence-supported applied game.

5. Conclusion

Taken together, the evidence reviewed here suggests that applied games represent a genuinely viable, if still maturing, therapeutic option for children's mental health problems, with the strongest support currently available for depression-focused interventions such as SPARX and comparatively more modest, mixed support for anxiety- and PTSD-focused titles. Given the small number of randomized controlled trials available for inclusion, these findings should be read as indicative of future potential rather than as definitive proof of efficacy, and considerably more rigorous, adequately powered trials are needed before applied games can be positioned confidently alongside established clinical treatments. Even so, the demonstrated feasibility of multiple distinct approaches — biofeedback, cognitive training, exergaming, VR, and narrative CBT-based design — combined with the pressing need for engaging, scalable, and affordable mental health tools, particularly in under-resourced settings, makes this an area well worth continued investment and independent evaluation.

References