Neurodivergent-Focused Game Accessibility: A Systematic Literature Review

Abstract

Accessible digital games represent an emerging frontier for inclusion, offering both challenges and opportunities to advance the cultural participation of neurodivergent people. This study presents a systematic literature review of empirical research on game accessibility for neurodivergent players, including autistic individuals and persons with intellectual disabilities. Forty-eight studies published between 2014 and 2025 were analysed to examine how accessibility is defined, implemented, and evaluated in game design. The results reveal that accessibility is often framed as a functional or therapeutic adjustment rather than as a social or cultural right. Although growing attention has been paid to sensory and cognitive barriers, few studies adopt neurodiversity-affirming or participatory frameworks. Most remain confined to educational or rehabilitation contexts, with limited involvement of neurodivergent co-designers. The findings call for a redefinition of accessibility as a creative, ethical, and political principle central to inclusive digital culture and the neurodiversity movement.

1. Introduction

Games matter for neurodivergent populations. Calls for neurodivergent accessibility in games have been happening for around a decade, with some early strides in the games’ industry adoption. The calls proposed the task of resolving neurodivergent accessibility in games as an interdisciplinary practice anchored in human–computer Interaction (HCI), inclusion and critical disability studies. This task remains unfulfilled.

Serious games for neurodivergent people have not completely shed medical and psychological models of disability. Implementations remain focused on surface-level interactions rather than placing neurodivergent inclusion at the heart of game design processes. The field continues to give the impression of a succession of somewhat redundant guideline proposals, failing to present a consolidated program for inclusive research and practice. A contribution towards the needed consolidation lies with a systematic literature review of empirical studies of neurodivergent accessibility, including game proposals and implementations, to gauge the state of use of definitions of accessibility, inclusive design, domain-specific guidelines and standards.

Gaming and Neurodiversity

The concept of neurodiversity emerged in the late twentieth century as both a scientific and socio-political response to deficit-based understandings of neurological difference. It challenges traditional medical models that frame autism, attention-deficit/hyperactivity disorder (ADHD), dyslexia, Tourette syndrome, and related conditions as disorders to be cured or normalised. Instead, neurodiversity recognises such variations as natural forms of human cognitive diversity, comparable to biodiversity in ecological systems [1,2]. Within this framework, neurological differences are seen as intrinsic to human variation and as potential sources of creativity, problem-solving, and alternative modes of perception rather than as deviations from an assumed norm.

From its inception, the neurodiversity movement carried a distinctly activist orientation. Originating in online autistic self-advocacy networks, it called for self-determination, social recognition, and an end to paternalistic approaches that excluded neurodivergent people from decisions affecting their lives [1]. This activism aligned with the social model of disability, which locates barriers not within individuals but within environments that fail to accommodate human diversity. The political dimension of neurodiversity thus redefines accessibility as a matter of rights and justice, rather than as a technical add-on to existing systems. Advocates have linked the movement to broader struggles for inclusion and participation in cultural, educational, and technological spheres [3,4].

Within this political and ethical context, accessible gaming becomes more than a question of interface design. It represents a cultural frontier where neurodivergent people seek equitable participation, creative authorship, and visibility. Games serve as both expressive media and social arenas in which norms about cognition, emotion, and communication are enacted. For many neurodivergent players, the right to play and design games parallels the right to access cultural production and representation. The call for accessible gaming therefore intersects with the neurodiversity movement’s demand for inclusive culture, advocating for the dismantling of cognitive barriers that restrict participation in digital leisure and creative economies [5].

Neurodivergent activism has also expanded the meaning of accessibility beyond the accommodation of sensory sensitivities or attention differences. It includes the recognition of alternative play styles, atypical communication preferences, and distinct sensory-emotional experiences of interaction [6]. These insights encourage a reconceptualisation of accessibility as a collaborative and interpretive process, grounded in the lived expertise of neurodivergent communities. Such a perspective transforms accessible design from an act of correction to an act of cultural co-creation [7], positioning neurodivergent players and designers as essential contributors to the evolution of game cultures [6,8].

Heron [9] explains how inaccessibility and exclusion of people with disabilities in digital games (in which we can include neurodivergent players) constitute theft of social integration, cultural capital, speech, and representation leading to real, persistent disempowerment, to the extent that “accessible interfaces are not a niche consideration” and “accessible design is not a luxury” [9]. Access and inclusion in digital games therefore fall under the mandated outcomes of Article 9 of the United Nations Convention on the Rights of People with Disabilities (UNCRPD) [10], namely “access, on an equal basis with others” to “information and communications technologies and systems” and promoting this access early in the design and development process as opposed to an afterthought, as well as full “participation in the community” per Article 19 of the UNCRPD [10].

Themes in barriers for neurodivergent players of digital games include a range of physical barriers, cognitive barriers related to executive dysfunction, to memory problems, to compulsive behaviour, and to expectation mismatch, as well as psychological and emotional barriers related to anxiety and depression and to executive dysfunction. Themes in barriers reported by these players pertaining to game design and features relate to time-pressure, adjustments for controls, difficulty, visuals and audio, as well as the need to focus [11]. However, HCI “research efforts focusing on games for neurodivergent players in particular remain under-explored”, with accessibility in games showing a “strong focus on physical disability” but a lack of design implementation for neurodivergence [6]. The situation has been slow to improve, has only improved to a limited degree, and still led by concerns with controls and interface, both in guidelines and in implementation [12].

In summary, this review Systematic Literature Review (SLR) intends to address the persistent gap between growing calls for neurodivergent accessibility in games and the fragmented ways in which accessibility is currently defined, implemented, and evaluated in empirical research. Through the systematic mapping and analysis of how accessibility is conceptualised, operationalised, and tested across studies, it aims to clarify the current state of the field, identify recurring limitations, and outline directions for more coherent, participatory, and neurodiversity-affirming approaches to game design. In this sense, it explored the main Research Question (RQ1): “How has accessible game design been studied and implemented to support neurodivergent players?”. Therefore, to better explore this main question, it will comprise a set of more specific questions, namely:

• RQ2—What are the most common accessibility barriers faced by neurodivergent players in digital games?
• RQ3—Which game design features and adaptations are most frequently proposed or implemented to enhance accessibility for neurodivergent groups?
• RQ4—How do different game genres address the needs of neurodivergent players?
• RQ5—To what extent are neurodivergent players involved in the game design process?
• RQ6—What gaps exist in current research on accessible game design for these populations, and what future directions are suggested?

2. Notes on Language and Conceptual Framing

Before presenting this study and its findings and recognising the connections between language, action, power, identity and culture, we find it important to clarify the language used throughout this paper, as there has been an ongoing debate about the use of identity-first versus person-first language in research [13,14].

Although part of the eligibility criteria was that participants were either self-identified or were formally diagnosed, we adopted a non-diagnostic-based language in an effort to distance ourselves from the medical model of disability and move closer to the social model.

In line with this, when referring to people with intellectual disabilities, we will use a person-first language, as previous research [15] has informed us that this is the preference of these individuals. On the contrary, the Autistic community has a preference for identity-first language [16]; so, when referring specifically to this group, an identity-first language will be adopted. Even though they have different views on language, when referring to these two groups, we will adopt the term ‘neurodivergent(s)’, as it encompasses individuals who have cognitive functions that differ from the dominant societal view of ‘normal’ [2], including, but not restricted to people with intellectual disabilities, autistic, dyslexic, ADHD and epileptic individuals.

However, we also recognise that language is associated with identity [17] and that identity is something specific to each person. Although this may be the language adopted by the majority, we also acknowledge that it does not mean it is adopted by all. It is also important to mention that a specific disability language and positionality statement is also described in the final part of this article.

Although this review adopts the term ‘neurodivergent’ to reflect a broad and inclusive conceptual framing, the empirical scope of this review is intentionally narrower. In line with the inclusion criteria, the studies reviewed predominantly focus on autistic individuals and people with intellectual disabilities, as these groups represent the most consistently addressed populations in the game accessibility literature. The use of an umbrella term is therefore conceptual rather than exhaustive, and findings should not be assumed to generalise to all neurodivergent conditions. This boundary is maintained to ensure analytical coherence while acknowledging the broader diversity encompassed by neurodivergence, through the previously mentioned activist-driven orientation [1].

3. Methods

The present study, as previously mentioned, adopted an SLR research design as a strategy to broadly map the evidence in the field in which it is both dense and dispersed. For such purpose, PRISMA guidelines [18] were adopted, and can be followed through in the developed checklist (see https://www.doi.org/10.6084/m9.figshare.31007179, accessed on 6 January 2026).

3.1. Eligibility Criteria

Studies were eligible for inclusion if they explicitly involved autistic individuals and/or people with intellectual disabilities, either through formal diagnoses or self-identification as neurodivergent. Eligible studies had to address digital games designed for or studied with these populations and be reported as peer-reviewed empirical research (quantitative, qualitative, or mixed-methods). Case studies of game development processes were also included if they incorporated neurodivergent participants. To ensure relevance to contemporary accessibility debates, only publications from 2007 onwards, following the adoption of the UNCRPD [10], and written in English were considered.

Exclusion criteria encompassed studies that focused exclusively on neurotypical participants or on disabilities unrelated to the scope of this review (like physical, visual, or hearing impairments). Also excluded were studies centred solely on non-digital games (such as board or physical games), and papers addressing only general “game-based learning” without specific adaptations or considerations for neurodivergent players.

3.2. Information Sources

The literature search was conducted across five electronic databases and research platforms. The following sources were consulted: ACM Digital Library (last searched 8 April 2025), Google Scholar (9 April 2025), B-On (Biblioteca do Conhecimento Online; 9 April 2025), ProQuest (9 April 2025), and ResearchGate (9 April 2025).

For Google Scholar and ResearchGate, searches were conducted through a free mapping strategy, initially by using the keyword “Games Accessibility.” Results were then manually filtered to include only those related to neurodivergence and associated conditions (like autism, ADHD, dyslexia, or intellectual disability). This exploratory approach was chosen to complement the more structured searches run in subscription databases and to ensure that emerging or interdisciplinary works, which are often less systematically indexed, were not overlooked.

The inclusion of B-On (https://www.b-on.pt/en/, accessed on 9 April 2025) was particularly relevant. B-On is the Portuguese national consortium that provides unified access to a wide range of international publishers’ collections, including journals, conference proceedings, and e-books across disciplines. Through B-On, institutions in Portugal gain access to most of the major publishers (e.g., Elsevier, Springer, IEEE, Wiley), making it a crucial entry point to international peer-reviewed literature.

3.3. Search Strategy

The search strategy was designed to capture peer-reviewed studies addressing digital game accessibility in relation to neurodivergent populations. Boolean strings combined terms for accessibility, neurodivergence, and game development/design, with minor adaptations across databases to account for indexing rules (e.g., truncation). A publication date filter (from 2007 onwards) was applied to align this review with the adoption of the UNCRPD (UN, 2006) and subsequent developments in accessibility practices. Only English-language publications were considered. A more general search strategy was used in the databases in which truncation was not supported (

Table 1. General Boolean search string used in the systematic review (non-truncated version).

Search String:
(“accessibility” OR “cognitive accessibility” OR “psychological accessibility”)
AND (“neurodivergent people” OR “neurodiverse people” OR “neurodiversity” OR “autism” OR “ADHD” OR “hyperactivity” OR “dyslexia” OR “dyspraxia” OR “Tourette’s syndrome” OR “OCD” OR “anxiety disorders” OR “bipolar disorder” OR “intellectual disability” OR “intellectual disabilities”)
AND (“digital game development” OR “video game development” OR “videogame development” OR “game development” OR “game design” OR “gaming industry”)

This version was applied in databases that do not support truncation symbols.

) and an adapted one for the remaining databases (

Table 2. General Boolean search string used in the systematic review (truncated version).

Search String:
(“accessibility” OR “cognitive accessibility” OR “psychological accessibility”)
AND (“neurodivergent people” OR “neurodiverse people” OR “neurodiversity” OR “autism” OR “ADHD” OR “hyperactivity” OR “dyslexia” OR “dyspraxia” OR “Tourette’s syndrome” OR “OCD” OR “anxiety disorde*” OR “bipolar disorde*” OR “intellectual disabilit*”)
AND (“digital game development” OR “video game development” OR “videogame development” OR “game dev*” OR “game design” OR “gaming industry”)

This version was applied in databases that support truncation symbols.

).

Search strings were adapted to the specific syntax and functionalities of each database. While the conceptual blocks (e.g., game-related terms, accessibility, neurodivergence) were kept consistent across sources to ensure conceptual coherence, field tags, operators, and limits were adjusted according to database requirements. Table 1 presents the core Boolean logic adopted. Conversely, Table 2 presents the truncated version of that same query, implemented in databases that technically allow it. Some more database-specific implementations followed the technical constraints of each platform.

For Google Scholar, searches were conducted using simplified keyword combinations derived from the main Boolean blocks due to platform constraints. The results were sorted by relevance, and screening was conducted sequentially until thematic saturation was reached, defined as the point at which no new eligible studies were identified across consecutive result pages. In line with common systematic review practices, the first 300 results were screened.

ResearchGate was used exclusively as a supplementary discovery tool. Items retrieved via ResearchGate were only included if they could be independently verified as peer-reviewed publications, based on confirmation of publication in indexed journals or conference proceedings through the publisher’s website or bibliographic databases (SCImago Journal & Country Rank, https://www.scimagojr.com, accessed on 9 April 2025) was used as a source for confirming the peer-reviewed status whenever possible.). Preprints, unpublished manuscripts, and non–peer-reviewed materials were excluded.

3.4. Selection and Data Collection Process

Screening at the title and abstract level was conducted by one senior researcher, with all decisions discussed and validated in consultation with another scholar to ensure consistency. The eligibility assessment of full-text reports was then performed by two junior researchers, who discussed all unclear cases and “maybes” with a senior scholar. Final inclusion decisions were reached through iterative discussion to ensure consistency.

3.5. Data Items and Synthesis

Data were extracted into a structured Excel sheet created for this review. Extraction was conducted by four junior researchers and supervised by two senior scholars with whom all entries were subsequently discussed and validated to ensure consistency. No automation tools were used at any stage. For each included study, information was collected on bibliographic details, study characteristics, participant profiles, study aims, game-related features, accessibility measures, and the role of neurodivergent participants in the research process. Outcomes of interest were those relating to the operationalisation of accessibility in digital game design and use, with particular attention to cognitive and psychological dimensions. All results compatible with these domains were considered, and no restrictions were applied to measures, time points, or analyses.

Depending on the nature of the data, the results were synthesised either through descriptive statistics, calculated using JASP (v0.19.3), or through critical content analysis [19], ensuring both quantitative and qualitative dimensions of the evidence base were adequately captured.

3.6. Study Risk of Bias Assessment

To assess the risk of bias across the included studies, we adopted a tiered classification framework adapted from [20], which provides guidance for synthesising qualitative and quantitative health evidence. This approach allowed us to move beyond binary notions of study quality and instead capture nuances of methodological robustness and relevance to neurodivergent accessibility in digital game development. Each study was independently assessed and classified into one of three tiers:

• Tier 1—Low relevance/weak methods: These studies were characterised by limited methodological transparency, theoretical or conceptual approaches without empirical grounding, or marginal relevance to neurodivergent accessibility in digital game design.
• Tier 2—Moderate methods/some relevance: These contributions displayed adequate but not rigorous methods, with partial attention to accessibility for neurodivergent populations. While informative, their findings were considered limited in terms of transferability or depth of evidence.
• Tier 3—Strong methods/clear accessibility framing: These studies demonstrated robust methodological approaches (qualitative, quantitative, or mixed), provided empirical data or detailed design evaluations, and framed accessibility explicitly in relation to neurodivergent populations.

For studies where reporting was ambiguous or incomplete, classification decisions were made conservatively, based on available information. Where methodological rigour could not be confidently determined, this study was assigned to the lower tier. This approach ensured transparency and consistency in assessing potential bias while acknowledging the heterogeneity of the research approaches represented in this review.

4. Results

4.1. Study Selection

The systematic search across five databases yielded a total of 2695 records (ACM Digital Library = 365; Google Scholar = 260; B-On = 1785; ProQuest = 278; ResearchGate = 7). After the removal of 1119 duplicates, 1576 records were screened at the title and abstract level. Of these, 1490 records were excluded for not meeting the inclusion criteria.

A total of 86 full-text reports were retrieved and assessed for eligibility. Following full-text screening, 38 reports were excluded for the following reasons: duplicates (n = 9), not empirical studies (n = 9), involving out-of-scope populations (n = 7), not focused on digital games (n = 6), not addressing game accessibility or design (n = 6), and language restriction (Portuguese only; n = 1).

The final sample comprised 48 studies, which were included in this review. The study selection process is summarised in this study’s flow diagram (Figure 1).

Some reports initially appeared to meet the inclusion criteria but were excluded upon closer inspection. For instance, certain studies discussed accessibility but focused exclusively on non-digital games, and were therefore removed at the eligibility stage. Others addressed neurodiversity more broadly but without linking it to digital game design or accessibility, and thus did not align with the scope of this review.

4.2. Study Characteristics

The sample of 48 studies included a total of 2399 participants, with an average sample size of 54.5 per study ($SD$ = 104.1). The study with the larger sample included 623 individuals [21] and the smaller sample was studied by Sancho et al. [22], with a total of three participants. The different sample sizes, alongside other studies’ characteristics, are summarised in

Table 3. Overview of studies’ characteristics (N = 48).

Study	Country	Type/Methodological Approach	Sample Size	Condition(s)
[23]	Several Countries	Empirical Study (Mixed Methods)	60	Autism
[24]	Brazil	Empirical Study (Quantitative)	38	Autism
[25]	Brazil	Empirical Study (Qualitative)	18	ADHD
[26]	Italy	Empirical Study (Mixed Methods)	69	Autism
[27]	Spain	Empirical Study (Mixed Methods)	10	Intellectual Disability
[28]	Portugal	Empirical Study (Quantitative)	11	Intellectual Disability
[29]	Taiwan	Empirical Study (Quantitative)	40	Intellectual Disability
[30]	South Korea	Empirical Study (Mixed Methods)	8	Intellectual Disability
[31]	Brazil	Empirical Study (Quantitative)	37	Intellectual Disability
[32]	Qatar	Empirical Study (Quantitative)	77	Intellectual Disability
[33]	The Netherlands	Conceptual Paper (Qualitative)	n/a	Multiple Conditions
[34]	Brazil	Design Case Study (Qualitative)	17	Autism
[35]	Belgium	Empirical Study (Qualitative)	15	ADHD
[36]	United States	Empirical Study (Qualitative)	11	Other
[37]	Ecuador	Design Case Study (Qualitative)	n/a	Dyslexia
[38]	Ecuador	Design Case Study (Qualitative)	n/a	Autism
[39]	South Korea	Empirical Study (Mixed Methods)	27	Autism
[40]	Belgium	Design Case Study (Qualitative)	88	Multiple Conditions
[41]	France	Empirical Study (Quantitative)	70	Autism
[42]	USA	Empirical Study (Mixed Methods)	65	Autism
[43]	China	Empirical Study (Quantitative)	110	ADHD
[44]	Canada	Empirical Study (Quantitative)	5	Autism
[45]	USA	Empirical Study (Mixed Methods)	7	Autism
[46]	Pakistan	Empirical Study (Qualitative)	23	Autism
[22]	Several Countries	Design Case Study (Qualitative)	3	Intellectual Disability
[47]	Portugal	Empirical Study (Qualitative)	80	Multiple Conditions
[48]	Several Countries	Empirical Study (Quantitative)	178	Dyslexia
[49]	Germany	Empirical Study (Qualitative)	5	Dyslexia
[50]	Several Countries	Empirical Study (Quantitative)	313	Dyslexia
[51]	Several Countries	Empirical Study (Qualitative)	5	Dyslexia
[52]	Mexico	Design Case Study (Qualitative)	9	Multiple Conditions
[53]	Several Countries	Empirical Study (Quantitative)	43	Dyslexia
[54]	Ecuador	Empirical Study (Mixed Methods)	47	Multiple Conditions
[55]	Greece	Empirical Study (Mixed Methods)	120	ADHD
[56]	UK	Empirical Study (Mixed Methods)	12	Intellectual Disability
[57]	Egypt	Empirical Study (Mixed Methods)	7	Autism
[58]	Portugal	Empirical Study (Mixed Methods)	31	Intellectual Disability
[59]	Portugal	Empirical Study (Mixed Methods)	14	Intellectual Disability
[60]	Portugal	Empirical Study (Quantitative)	31	Intellectual Disability
[61]	Portugal	Empirical Study (Quantitative)	15	Intellectual Disability
[62]	Mexico	Design Case Study (Quantitative)	12	ADHD
[21]	France	Empirical Study (Mixed Methods)	623	ADHD
[63]	China	Empirical Study (Quantitative)	20	ADHD
[64]	USA	Empirical Study (Qualitative)	3	Multiple Conditions
[65]	South Korea	Empirical Study (Mixed Methods)	4	Autism
[66]	China	Empirical Study (Qualitative)	14	Autism
[67]	Taiwan	Empirical Study (Qualitative)	4	ADHD
[68]	Croatia	Design Case Study (Qualitative)	n/a	Multiple Conditions

.

As expressed in Table 3, the studies in the sample were published between 2014 and 2025, and conducted across a broad range of geographical contexts. The largest number of studies originated from Portugal (n = 6; 12.5%), followed by Brazil (n = 5; 10.4%), South Korea (n = 4; 8.3%), and the United States (n = 4; 8.3%). It is also relevant to highlight the presence of six studies simultaneously conducted in several countries. Individual studies were also conducted in Belgium, Canada, China, Croatia, Ecuador, Egypt, France, Germany, Greece, Italy, Mexico, Pakistan, Qatar, Spain, Taiwan, the Netherlands, and the United Kingdom. (2.1% each)

In terms of type of contribution, the majority were empirical studies (n = 38; 79.2%), complemented by design case studies (n = 7; 14.6%) and a smaller number of conceptual papers (n = 1; 2.1%). Methodologically, the studies were diverse: quantitative approaches (n = 16; 33.3%) and qualitative approaches (n = 15; 31.3%) were frequent, with mixed methods (n = 13; 27.1%) also well represented. Three studies (6.3%) were categorised as design case studies adopting primarily qualitative strategies, and one as conceptual work.

The sample sizes varied considerably, ranging from 3 to 623 participants, with some design studies reporting no direct participant numbers. Conditions addressed included autism (n = 17; 35.4%), Intellectual Disability (n = 15; 31.3%), ADHD (n = 9; 18.8%), dyslexia (n = 7; 14.6%), and multiple conditions (n = 7; 14.6%). Some studies involved smaller or non-specified groups (like the study by Žilak et al. [68]).

Publication years were unevenly distributed. The earliest study was published in 2014, with a steady growth thereafter. Peaks were observed in 2022 (n = 9; 18.8%), 2023 (n = 7; 14.6%), and 2024 (n = 9; 18.8%), reflecting increasing scholarly attention to accessibility for neurodivergent populations in game design. The most recent works included those from 2025 (n = 3; 6.3%).

Aims of the Studies

The aims of the studies included in this review were diverse but can be grouped into several broad orientations, as shown in

Table 4. Studies in the sample grouped by the strand/category of their specific aim (N = 48).

Strand	Number of Studies (%)	Included Studies
Assessment and screening	10 (20.8%)	[21,39,41,42,45,48,49,50,51,68]
Co-design and participatory inquiry	5 (10.4%)	[28,34,35,36,47]
Context and engagement (sociocultural)	1 (2.1%)	[46]
Guidelines/methods/frameworks	7 (14.6%)	[26,37,38,40,58,59,60]
Inclusive play and sensory design	3 (6.3%)	[44,64,66]
Interventions/training development	22 (45.8%)	[22,23,24,25,27,29,30,31,32,33,43,52,53,54,55,56,57,61,62,63,65,66]

Percentages represent the proportion of studies (N = 48) addressing each thematic strand. Some studies contributed to more than one strand.

.

The aims of the 48 included studies clustered into six main strands (see Table 4). A first strand comprised works on co-design and participatory inquiry (n = 5; 10.4%), which sought to involve neurodivergent participants directly in the design process, for example by eliciting perspectives of teenagers with attention-deficit/hyperactivity disorder [35] or engaging children with intellectual disabilities, autism, and aphasia in participatory co-creation [28,34,36,47].

The largest group of studies focused on the development and piloting of game-based interventions or training programmes (n = 22; 45.8%). These ranged from mathematics learning games for ADHD [25] and VR-based cognitive training for young adults with intellectual disabilities [61] to serious games supporting emotion recognition in autism [24] and daily routine memorisation for Down syndrome [31].

A third strand encompassed the use of games as tools for assessment and screening (n = 10; 20.8%). These included language-independent web-based games to predict or screen for dyslexia [48,49,51], mobile tools for assessing executive functioning in autism [42], and game-based performance tasks to measure representational flexibility in autistic adolescents [45].

Another group of studies addressed the elaboration of guidelines, methods, and frameworks for accessibility (n = 7; 14.6%). Examples include the creation of evidence-based design guidelines for adults with intellectual disabilities [60], participatory action research for accessible gaming [58,59], and methods for developing accessible serious games for dyslexia and autism [37,38].

Three studies focused on inclusive play and sensory design (n = 3; 6.3%), such as investigations of AR tabletop play with children with autism [66] and VR games designed for neurodiverse children to address sensory diversity [64].

Finally, one study (2.1%) emphasised the role of sociocultural and institutional context in shaping game engagement, highlighting how local factors influenced children with autism in Pakistan [46].

4.3. Risk of Bias in Studies

The assessment of methodological quality and potential risk of bias conducted using the tiered framework adapted from [20], revealed notable variation across the included studies. Of the 48 studies, 13 (27.1%) were classified as Tier 1, indicating weak methodological rigour, limited empirical grounding, or marginal relevance to neurodivergent accessibility in digital game development. Twenty-three studies (47.9%) were classified as Tier 2, reflecting moderate methodological robustness and partial consideration of accessibility concerns. Finally, 12 studies (25.0%) achieved Tier 3 status, demonstrating strong methodological quality, empirical grounding, and clear framing of accessibility in relation to neurodivergent populations.

Overall, nearly half of the sample was situated in Tier 2, suggesting that although accessibility for neurodivergent players has gained some methodological attention, it often remains inconsistently operationalised. The relatively small proportion of Tier 3 studies highlights the need for more rigorous and explicitly accessibility-oriented research within this domain. Conversely, the considerable presence of Tier 1 studies highlights the persistence of conceptual or theoretical contributions that, while valuable in scope, provide limited empirical evidence to inform practice or policy.

4.4. Results of Syntheses

4.4.1. Views on Accessibility in the Sample of Studies

Most studies did not adopt a particular or clear definition of accessibility (n = 44; 91.67%), meaning that this concept was not specifically defined in terms of the developed research. The remaining four studies (8.33%) clearly stated the adopted definition of accessibility in the scope of the study, as presented in

Table 5. Definitions of accessibility adopted in the sample of studies (n = 4).

Study	Definition of Accessibility
[37]	“Accessibility is the characteristics implemented to objects, systems or environments to allow for their use despite the disability or condition of a person.”
[60]	Both studies adopted a definition emerging from the work of [69], which defines accessibility as the “design of products, services, and environments that are usable and navigable by as many people as possible, including those with disabilities or other diverse needs”.
[59]	Same definition as [60]—see above.
[68]	“Accessibility is a measure of the extent to which a product, service, or environment is suitable for all users, including people with disabilities and the elderly.”

Only four studies explicitly defined accessibility. Two of them [59,60] referred to Iwarsson and Ståhl’s framework of design usability and inclusivity [69].

.

The four studies that did define accessibility adopted diverse perspectives, as presented in Table 5. One framed accessibility as a functional characteristic of systems or environments that enables use regardless of disability or condition [37]. Others emphasised a social or contextual approach, viewing barriers as constructed by societal constraints and mitigated through personalised technological support [30]. A further set of two studies drew on inclusive design principles, highlighting accessibility as the design of products and environments that are usable by the widest range of people, including those with disabilities [59,60]. Finally, one study conceptualised accessibility in terms of measurable suitability, stressing its role in ensuring inclusivity across populations, including people with disabilities and older adults [68].

Most specifically regarding the definition of cognitive accessibility, only two studies [59,60] clearly defined the concept, both based on previous work from [70]. According to this vision, it “aims to make information and communication more understandable, intuitive, easy to process, and interactive for neurodiverse people”.

Conversely, a larger number of studies (n = 23; 47.9%) adopted any guidelines, frameworks, or theories to support the enhancement in cognitive accessibility in the proposed game or game-based intervention.

Some works adopted participatory or values-oriented approaches, such as accessibility and theme-driven guidelines co-created with young people with attention-deficit/hyperactivity disorder [35] the Socially Aware Design and VCIA model [25], exergame design principles from Hernandez et al. [71] as applied by Castelhano and Roque [28], and design frameworks for involving neurodiverse children [47,72,73,74]. Participatory design work for people with aphasia was also referenced, namely Moffatt et al. [75], applied by Hymes et al. [36].

Several studies drew on general accessibility and ergonomic standards, including ISO 9241-210 on human-system interaction [50], and the W3C note Making Content Usable for People with Cognitive and Learning Disabilities [26]. Broader tools such as Includification [76], Google Material Design Guidelines, or Apple’s Human Interface Guidelines were also applied [22,31].

A third group of studies reported educational and cognitive learning theories, including Mayer’s Cognitive Theory of Multimedia Learning and Skinner’s operant conditioning model [32], and Keller’s ARCS model of instructional design [29]. In health-sport contexts, frameworks for personalised and adaptive training [77] were integrated [52].

Finally, some works relied on condition-specific guidelines, such as autism software design principles [57], ADHD guidelines from Vonthron et al. [21] and Powell et al. [78], the human-centred design adaptations for autism [23,79,80], and recommendations for Down syndrome [22,31]. Additional examples included MagicBLOCKS guidelines [66], multisensory VR design for cognitive training [61], and assessment frameworks for mobile serious games [37,38].

In sum, while nearly half of the studies incorporated formal or informal frameworks, their application was fragmented and context-specific, with no single dominant reference point across the literature. This reflects both innovation and the lack of consensus on standard practices for operationalising cognitive accessibility in game development.

4.4.2. Accessibility for Neurodivergent Players

To better contextualise the sample of studies regarding the way games were framed, it is relevant to highlight that most of them adopted games for “serious purposes” (n = 32; 66.7%), meaning that they are used as tools to promote learning, skill acquisition well-being, or any other outcome that is external to the gameplay. Eight studies (16.7%) combined both a serious and an entertainment gaming driven approach, while five studies (10.4%) were focused on entertainment only. Moreover, three studies (6.3%) were unclear or did not specify the framing or type of use of the games adopted.

Regarding the addressed game genres (meaning the genre of the games developed or adapted for accessibility purposes on the scope of each study) puzzles were the most frequent (n = 17; 35.4%). The second most common situation was the creation or adaptation of multiple game genres in the same study (n = 14; 29.2%). Simulation and action games were adopted in four studies each (8.3%), while one study (2.1%) was centred around a Role Playing Game (RPG). There were also eight studies (16.7%) that lacked enough details in their report that allow for a coherent categorisation of the game genre(s).

As a strategy to understand the intersection between these two axes the analysis represented in

Table 6. Distribution of game genre(s) addressed by games’ framing (N = 48), with percentages calculated for the total of each game framing category.

Games’ Framing	Puzzle (n; %)	Multiple Genres (n; %)	Simulation (n; %)	Action/ Adventure (n; %)	RPG (n; %)	Other/Not Specified (n; %)
Both approaches (n = 8)	2 (25.0%)	2 (25.0%)	2 (25.0%)	0 (0.0%)	0 (0.0%)	2 (25.0%)
Entertainment only (n = 5)	0 (0.0%)	3 (60.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	2 (40.0%)
Serious purposes (n = 32)	14 (43.8%)	7 (21.9%)	2 (6.3%)	4 (12.5%)	1 (3.1%)	4 (12.5%)
Unclear/Not specified (n = 3)	1 (33.3%)	2 (66.7%)	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
Total (N = 48)	17 (35.4%)	14 (29.2%)	4 (8.3%)	4 (8.3%)	1 (2.1%)	8 (16.7%)

Percentages were calculated for each game framing category (N = 48). Categories are not mutually exclusive, as some studies addressed multiple genres.

was developed.

The distribution of game genres varied substantially according to the framing of the games (Table 6). Studies that framed games for serious purposes dominated the sample (n = 32; 66.7%) and most frequently employed puzzle games (43.8%), followed by multiple genres (21.9%) and action/adventure titles (12.5%). Simulation and role-playing games appeared less frequently in this group (6.3% and 3.1%, respectively). By contrast, studies with an entertainment-only framing (n = 5; 10.4%) showed no reliance on puzzle games, but instead concentrated on multiple genres (60%) and non-specified/other genres (40%).

Across the 48 studies, mechanics designed or adapted for cognitive accessibility revealed several recurring design strategies, categorised in

Table 7. Distribution of game mechanics by analytical category (N = 48).

Category	Examples of Mechanics	Representative Studies
Memory, attention, and concentration training	Memory card/puzzle tasks; progressive sequences of colours/sounds; auditory/visual cues; inhibitory control games; rhythm-based tasks; repetition and reinforcement	[21,23,36,37,41,42,43,45,48,50,51,61]
Adaptive difficulty and personalisation	Dynamic difficulty adjustment; real-time adaptation to player performance; avatar customisation; tutorials and simplified instructions; configurable parameters (colour, contrast, speed, font, voice)	[21,26,34,52,56]
Multimodal and sensory interaction	Use of visual, auditory, and tactile cues; eye-blink selection; gesture-based control (tap, drag, pinch, air gestures); integration of EEG, VR, or AR sensors; multisensory feedback (sound, visuals, haptics)	[22,31,32,34,44,62,66]
Narrative and representation	Inclusive or relatable storylines (e.g., ADHD representation, The Little Prince, emotion-based narratives); role-play and empathy-building scenarios	[24,30,33,35,44]
Motivation, feedback, and rewards	Immediate feedback (positive/negative reinforcement); achievements and rewards to sustain engagement; playful challenges; clear guidance	[30,41,53,55]
Social and collaborative play	Cooperative mechanics (joint attention tasks; co-play roles; inclusive robotic play); communication tools (teleport, voice chat); multiplayer rhythm or performance patterns	[28,47,64,65]
Therapeutic and cognitive-behavioural grounding	Mechanics based on behavioural reinforcement; stress regulation and mentalisation; self-regulation tasks (e.g., closing eyes to trigger alpha waves); emotional recognition tasks	[24,33,55,62]

. A strong emphasis was placed on memory, attention, and concentration, often operationalised through puzzle games, rhythm-based challenges, or inhibitory control tasks. These mechanics reflect attempts to scaffold core cognitive processes frequently identified as support needs among neurodivergent players.

Equally prominent were adaptive and personalised systems, ensuring that difficulty, presentation, and interaction could be adjusted to player abilities. This included real-time adaptation to performance, adjustable sensory parameters (colour, contrast, speed), and simplified tutorials, which reduce cognitive load and improve usability.

Many studies integrated multimodal and sensory interaction, highlighting the importance of flexibility in input and feedback modalities. Mechanisms such as eye-blink selection, gesture-based control, or multisensory VR/AR environments broadened accessibility for diverse users, particularly those with motor or perceptual challenges.

Narrative framing and representation were also central. Games frequently used relatable or inclusive storylines (like representing ADHD experiences, embedding familiar children’s stories, or centring on emotions), which functioned as motivational anchors and helped sustain engagement.

Motivational structures and feedback mechanisms ensured that players received continuous reinforcement, with immediate responses to actions, achievements, or adaptive guidance reducing frustration and maintaining flow. Finally, several works embedded collaborative and therapeutic dimensions. Cooperative play mechanics promoted social communication, while others drew on behavioural reinforcement or therapeutic frameworks for stress regulation and self-control.

4.4.3. Implementing and Testing Accessibility

Regarding the process of implementing accessibility, the predominant accessibility features that were implemented, adapted, or analysed in the sample of studies were related to customisation options (e.g., font, difficulty, pacing), which happened in 17 studies (35.4%). Conversely, the least type of implemented or analysed accessibility measures were related to navigation aids and UI simplifications (n = 6; 12.5%). The full analysis of accessibility measures is presented in

Table 8. Predominant accessibility features adapted, implemented, or analysed in the sample of studies (N = 48).

Predominant Accessibility Features Adapted, Implemented, or Analysed	n (%)
Customisation options (e.g., font, difficulty, pacing)	17 (35.4%)
Feedback systems (visual, auditory, text-based)	10 (20.8%)
Narrative or emotional content accommodations	8 (16.7%)
Other	7 (14.6%)
Navigation aids and UI simplifications	6 (12.5%)

Percentages were calculated for the total number of studies (N = 48). Categories are not mutually exclusive.

.

As expressed (Table 8), seven studies (14.6%) presented accessibility features classified as “other”. From this subsample, it is possible to highlight the predominance of studies in which accessibility was promoted through hardware-based solutions, meaning that the interaction barriers were solved through the usage of peripherals, specific controllers, or even the adoption of different, easier or more personalised, gaming platforms. This included a diverse range of solutions, like the implementation of bodily-controlled interactions [28], social robots [47], adapted mobile solutions [22,65], or commercial movement-based controllers, like Microsoft Kinect [66].

While most studies included neurodivergent individuals in the sample (n = 39; 81.3%), only three studies (6.3%) considered them as co-designers, meaning that those individuals were fully and horizontally included in the game design and development process. Most studies included participants as playtesters only, without specifying procedural aspects (n = 24; 50.0%), followed by nine (18.8%) studies that included them as interviewees and/or filling questionnaires. Five studies (10.4%) included participants through inclusive design framings, meaning that—although not in a fully equitable way—their feedback and creative inputs were part of the process. Conversely, seven studies (14.6%) framed participants in completely passive roles regarding the research and game design processes.

Regarding the effectiveness of the users’ feedback in the cognitive accessibility enhancement of the game, the vast majority of the studies (n = 40; 83.3%) documented it as effective. Nevertheless, there were studies that did not approach this question (n = 5; 10.4), or in which the effectiveness was unclear (n = 3; 6.3%). The inclusion of these individuals in the accessible game-based studies and interventions was seen as having secondary impacts for them in 43 studies (89.6%), with only five studies (10.4%) not approaching this question. Regarding this, most studies either referenced multiple secondary outcomes from accessible gaming interventions (n = 15; 31.3%), or specifically focused on how they promoted learning (n = 15; 31.3%). Other relevant secondary outcomes were: well-being (n = 5; 10.4%); mental health (n = 4; 8.3%); self-determination (n = 2; 4.2%); physical health-related aspects (n = 1; 2.1%); or stigma reduction (n = 1; 2.1%).

The patterns reported above are presented descriptively as they appear in the reviewed studies and form the empirical basis for the critical interpretation developed in the Discussion.

5. Discussion

In the following sections, we first synthesise patterns and tendencies that emerge directly from the reviewed literature, and subsequently interpret these findings through a critical, neurodiversity-informed lens. Where interpretative or normative claims are advanced, they are explicitly grounded in, and distinguished from, the empirical trends identified in the synthesis.

5.1. How Has Accessible Game Design Been Studied and Implemented to Support Autistic Players and Individuals with Intellectual Disabilities? (RQ1)

Accessible game design for neurodivergent players remains in an emergent stage of consolidation. Across the reviewed studies, accessibility was predominantly conceptualised as a technical or functional characteristic of systems rather than as a comprehensive socio-technical process embedded within inclusive design cultures. Only a small subset of studies explicitly defined accessibility, most often relying on ergonomic or usability-based formulations, like in the studies from Iwarsson and Ståhl [69] or Sousa et al. [59,60]. This limited conceptual clarity mirrors broader tendencies in other HCI research, where accessibility is frequently operationalised through the optimisation of user interfaces rather than as an ethical or participatory imperative [81].

The majority of works in our sample investigated accessibility through game-based interventions targeting skill development or cognitive training for specific conditions such as autism, ADHD, or intellectual disability. These studies were typically driven by therapeutic or pedagogical aims, reflecting the persistence of a medical or deficit-based framing of neurodivergence, and aligned with previous results from [82]. Few studies approached accessible game design as a cultural and political practice oriented toward empowerment and representation. The dominance of serious game frameworks demonstrates how accessibility has largely been instrumentalised for rehabilitation or learning outcomes rather than integrated into mainstream entertainment or expressive game design. This bias was also previously documented by Sousa [82] and by Hassan [83].

From a methodological standpoint, the SLR results indicate that accessible game design has been explored through a wide variety of approaches but with limited coherence, ranging from controlled experimental testing to participatory workshops. Approximately half of the studies employed mixed or quantitative designs to assess usability, performance, or engagement, yet these approaches rarely captured the lived experiences or cultural meanings of play for neurodivergent individuals. The small proportion of design case studies and participatory works signals an ongoing tension between experimental rigour and participatory inclusiveness. This finding corroborates the argument that neurodiversity research in games often prioritises measurement over meaning, reducing accessibility to an engineering problem rather than a matter of justice, previously brought by Spiel and Gerling [6].

5.2. What Are the Most Common Accessibility Barriers Faced by Autistic Players and Individuals with Intellectual Disabilities in Digital Games? (RQ2)

Considering the SLR results, barriers encountered by neurodivergent players were multidimensional and intertwined cognitive, sensory, and emotional aspects. Studies frequently reported obstacles related to excessive cognitive load, complex navigation, time pressure, or unmodifiable difficulty levels. These barriers can exacerbate executive dysfunction and anxiety, reducing the sense of competence and autonomy that games typically foster. Furthermore, inconsistent or overstimulating audiovisual feedback, such as flashing lights or dense soundscapes, often created sensory overload for autistic players or those with attention-deficit/hyperactivity disorder.

Another recurrent barrier expressed in the results was the lack of customisation options that would allow players to adapt interfaces and feedback to their sensory or attentional profiles. Despite being widely recommended in accessibility guidelines [78], such options were seldom implemented beyond colour or contrast adjustments. Social barriers also emerged: multiplayer environments were often inaccessible due to complex communication systems, absence of inclusive moderation tools, or inadequate support for collaborative interaction.

The predominance of studies focusing on single diagnostic categories in the sample contributed to a fragmented understanding of accessibility barriers. Cross-cutting challenges that affect multiple neurodivergent groups, such as unpredictability, pacing, and comprehension of implicit rules, seem to remain underexplored. The evidence indicates that accessibility for neurodivergent players continues to be treated as condition-specific rather than as a universal design principle grounded in cognitive diversity.

5.3. Which Game Design Features and Adaptations Are Most Frequently Proposed or Implemented to Enhance Accessibility for These Neurodivergent Groups? (RQ3)

Across the reviewed corpus, recurring accessibility features included customisable settings, multimodal feedback systems, and adaptive difficulty mechanisms. These elements were designed to address differences in sensory processing, attention span, and cognitive flexibility. Adaptive systems that adjusted gameplay parameters in real time were particularly valued for maintaining engagement and minimising frustration. Games offering multiple feedback modalities, like visual, auditory, and tactile, also seemed to contribute to more inclusive experiences, consistent with universal design principles.

Narrative and representational inclusivity appeared as another important dimension in the sample. Several studies introduced relatable protagonists or scenarios depicting neurodivergent experiences, which can enhance motivation and self-identification. This aligns with recent activist and scholarly calls to expand accessibility beyond mechanics to narrative and aesthetic representation [84,85]. However, such examples remained exceptional rather than normative.

Mechanics emphasising memory, attention, or emotion regulation were prevalent in the analysed studies, particularly within serious games. These features often aimed to train cognitive functions rather than simply accommodate diversity, once again illustrating the dominance of remediation frameworks, as described in previous studies [82]. Despite this, the increasing integration of multimodal and sensor-based interfaces (e.g., eye-tracking, VR, or motion control) represents a promising direction for inclusive design when grounded in user-led processes. Without direct involvement of neurodivergent players, technological sophistication alone cannot guarantee accessibility.

5.4. How Do Different Game Genres Address the Needs of Autistic Players and Individuals with Intellectual Disabilities? (RQ4)

The predominance of puzzle and simulation genres in the sample reflects a focus on cognitive training rather than entertainment or creative exploration. Puzzle-based games were frequently selected for their structured challenges, perceived suitability for skill assessment, and ease of adaptive scaling. Simulation environments, including VR and AR formats, provided opportunities for safe rehearsal of social or daily living skills. Nevertheless, such genres can risk reinforcing therapeutic paradigms that limit neurodivergent players to functional improvement contexts.

Entertainment-oriented genres such as action, adventure, or role-playing games are underrepresented in research, according to the SLR results. When they did appear, their design rarely considered sensory overload, reaction time variability, or complex narrative comprehension. Yet these genres hold substantial potential for fostering agency, creativity, and social belonging if inclusive design principles are embedded from the outset [58,59]. The gap between serious and entertainment games in terms of accessibility parallels broader divisions between assistive and mainstream technologies.

According to the produced insights, cross-genre experimentation that blends cognitive accessibility with expressive potential remains rare. The scarcity of inclusive mainstream titles can be seen as underlining a structural issue in the gaming industry, where accessibility initiatives often originate from academic or therapeutic projects rather than commercial production pipelines. Bridging this gap requires a shift from accessibility as post-hoc adaptation toward accessibility as creative practice.

5.5. To What Extent Are Autistic Individuals and Those with Intellectual Disabilities Involved in the Game Design Process? (RQ5)

The findings reveal a persistent underrepresentation of neurodivergent players as active contributors in game design. Only a small minority of studies adopted full co-design methodologies, whereas the majority involved participants merely as playtesters or evaluators. This pattern limits the transformative potential of participatory design, which has been shown to yield more meaningful and context-sensitive accessibility outcomes in previous research [35,73].

Certain notions of tokenistic participation were also observed in studies describing limited consultation or feedback sessions without equitable decision-making power. Such approaches risk perpetuating paternalistic dynamics where researchers define accessibility needs on behalf of participants. Genuine participatory frameworks, as defined in the field of critical disability studies [86], require iterative collaboration, shared authorship, and recognition of neurodivergent expertise.

Notwithstanding these limitations, a few exemplary works demonstrated the benefits of inclusive design. Participatory projects involving children with intellectual disabilities or autism often reported enhanced engagement, creative ownership, and improved usability outcomes. These results corroborate evidence from broader disability research indicating that co-creation not only improves product quality but also contributes to empowerment and social capital [87]. Expanding participatory and emancipatory research traditions in game design thus represents a necessary condition for achieving substantive, not merely procedural, accessibility.

5.6. What Gaps Exist in Current Research on Accessible Game Design for These Populations, and What Future Directions Are Suggested? (RQ6)

The synthesis of evidence highlights several critical gaps. Conceptually, there seems to be an urgent need for consensus on definitions of cognitive accessibility and inclusive design that are grounded in neurodiversity theory rather than medical diagnostic categories. Without a shared conceptual foundation, the field risks perpetuating fragmented and discipline-specific discourses. Methodologically, greater integration of qualitative and participatory methods is necessary to complement the dominance of quantitative evaluation. Longitudinal designs capable of capturing sustained impacts of accessible play are particularly scarce [83,88].

In terms of practical application, few studies examined scalability or transferability of accessible features across contexts and populations. The absence of comparative research across genres or platforms prevents the development of generalisable design guidelines. Moreover, accessibility continues to be treated as an isolated research goal instead of a cross-cutting dimension of ethics, representation, and culture. The inclusion of neurodivergent perspectives within the mainstream game industry also remains limited, reflecting the broader structural inequities in creative labour previously described in research [89].

Future research should therefore prioritise multi-stakeholder collaborations linking academia, industry, and neurodivergent communities. Such collaborations should embrace accessibility as a process of negotiation between cognitive diversity, aesthetic freedom, and technological constraint. A neurodiversity-affirming agenda for game design would view accessibility not as remediation but as creative enrichment that benefits all players.

6. Conclusions

This review demonstrates that accessible game design for neurodivergent players has advanced in both scope and complexity but remains limited in conceptual coherence and participatory grounding. The main findings illustrating this conclusion are systematised in

Table 9. Synthesis of research questions, key findings, and identified gaps in the literature.

Research Question	Key Findings from the SLR	Identified Gaps and Future Directions
RQ1	Accessibility is mainly framed as a functional or therapeutic adjustment neglecting a certain “right to playfulness” of neurodivergent individuals	Need for cultural, rights-based, and neurodiversity-affirming framings of accessibility
RQ2	Cognitive, sensory, and emotional barriers are most frequently reported	Limited cross-condition and intersectional analyses of accessibility barriers
RQ3	Customisation options and adaptive difficulty mechanisms are the most commonly implemented features	Lack of attention to narrative, representational, and expressive dimensions of accessibility
RQ4	Puzzle and serious games dominate the literature	Underrepresentation of entertainment-oriented and mainstream game genres
RQ5	Minimal involvement of neurodivergent players as co-designers or in other positions that allow them to participate in the creative processes	Need for participatory, co-creative, and emancipatory design methodologies
RQ6	Conceptual approaches and methodological practices remain fragmented	Need for coherent frameworks and longitudinal, mixed-method research designs

. Most research continues to position accessibility as an ancillary feature of usability or therapeutic intervention rather than as an inherent element of cultural inclusion. The field has made tangible progress in identifying sensory, cognitive, and emotional barriers, yet it still struggles to translate these insights into holistic frameworks that recognise neurodivergent players as agents, not merely as beneficiaries.

The evidence suggests that digital games have significant potential to enhance autonomy, self-determination, and social participation when accessibility is approached as a design ethos rather than as a set of compensatory adaptations. Nevertheless, this requires a shift from the instrumental use of games for cognitive training only toward their recognition as expressive, relational, and cultural artefacts. The findings also underscore the importance of moving beyond condition-specific approaches toward a more integrated understanding of cognitive diversity.

6.1. Implications

The implications of this review extend across theoretical, methodological, and practical domains. Theoretically, this study reinforces the value of neurodiversity as a lens that reframes accessibility from deficit correction to creative pluralism. Game design that genuinely reflects neurodivergent perspectives contributes to broader social justice goals aligned with the United Nations Convention on the Rights of Persons with Disabilities [10], particularly the principles of participation and equality in cultural life.

Methodologically, the synthesis calls for more participatory and reflexive research strategies. Empirical studies must expand their focus beyond performance metrics to include lived experience, identity, and agency. Co-design methodologies, when genuinely inclusive, enable mutual learning between designers and participants, producing richer and more contextually valid insights. Such approaches align with emancipatory traditions in disability research that view participants as knowledge co-producers rather than subjects of observation [87].

Practically, the findings highlight the need for standardised frameworks and actionable design guidelines tailored to neurodivergent accessibility. Collaboration between academic researchers, developers, and neurodivergent communities can help establish benchmarks for inclusive practice. Industry adoption will depend on embedding accessibility at the earliest stages of production and recognising it as both an ethical obligation and a creative opportunity. Educational initiatives for designers and developers should also include training on cognitive accessibility and participatory ethics to sustain systemic change.

6.2. Limitations and Future Directions

This review’s conclusions are constrained by several factors, including the restriction to English-language, peer-reviewed studies and the heterogeneity of research methods and reporting standards across the sample. These limitations reduce comparability and may obscure important regional or community-based initiatives. The predominance of short-term interventions and the scarcity of participatory approaches also limit understanding of the long-term impact and sustainability of accessible design practices.

Future research should prioritise longitudinal, mixed-method, and participatory studies that include neurodivergent players as co-creators throughout the design process. Expanding collaboration between academic researchers, game developers, and advocacy groups can generate more comprehensive frameworks for cognitive and sensory accessibility. It is equally important to explore mainstream and entertainment contexts, where accessibility can function as a catalyst for creative innovation rather than as a specialised niche. A future agenda grounded in neurodiversity and inclusion will advance both the quality of accessible games and the recognition of neurodivergent voices within digital culture.