Systematic Review of Interventions for Twice-Exceptional Autistic Learners

Abstract

Twice-exceptional autistic learners (2e-ASL), those who meet criteria for autism and giftedness, are vulnerable to under-identification and inadequate support. Although interest is growing, evidence for interventions remains limited. This systematic review synthesised intervention studies involving 2e-ASL (4–18 years), including interventions designed specifically for 2e-ASL (2e-Specific) and interventions for Autistic learners that included 2e-ASL (2e-Inclusive). A PRISMA-guided search (2000–2025) identified 51 eligible studies. Forty-eight were 2e-Inclusive; only three were 2e-Specific. In the 2e-inclusive literature, 2e-ASL participants were rarely distinguished analytically and were typically subsumed under broad labels with very wide IQ ranges (often spanning ≥ 50 points within a single sample). Social (42.94%) and emotional (45.09%) outcomes were most frequently targeted, followed by intellectual outcomes (17.64%); behavioural (13.72%), academic (5.88%), language (3.92%), and sensory/perceptual (3.92%) domains were addressed far less often. Interventions were predominantly delivered in controlled settings like clinical or university settings (56.86%), with fewer school-based (35.29%) and limited home (3.92%) or community (3.92%) implementations. Effective programmes, regardless of domain, were consistently structured, explicit, and scaffolded, using visual supports, modelling, guided practice, and opportunities for rehearsed generalisation. 2e-Specific interventions were strength-based and interest-aligned, supporting creativity, advanced reasoning, and challenging, meaningful tasks; these were associated with gains in engagement, motivation, and behavioural regulation as well as targeted skills. The findings suggest that, although the evidence base is growing, it remains fragmented. The review highlights the need for clearer definitions of 2e-ASL, more consistent reporting of learners’ cognitive profiles, and a stronger focus on real-world application to strengthen the evidence base. Interventions that adopt strength-based, structured, and scalable approaches appear particularly promising for supporting meaningful, generalisable, and sustained outcomes for 2e-ASL.

1. Introduction

Autism is a neurodevelopmental condition characterised by differences in social communication and interaction, alongside restricted or repetitive patterns of behaviour, interests, or activities (American Psychiatric Association, 2022). While autism is defined within a single diagnostic framework, it is widely understood as a highly heterogeneous spectrum, with substantial variation in cognitive profiles, developmental trajectories, and everyday functioning (American Psychiatric Association, 2022; Centers for Disease Control and Prevention (CDC), 2014). One of the most important sources of this variability is intellectual functioning, which strongly influences learning, adaptive skills, and psychosocial outcomes across development (Charman et al., 2011; Wolff et al., 2022).

Within this broad spectrum, some autistic individuals also demonstrate exceptional intellectual, academic, or creative abilities. These individuals are commonly described as twice-exceptional (2e), referring to the co-occurrence of giftedness and disability (Reis et al., 2014). Giftedness is typically understood as performance, or the potential for performance, that is well above that of age-matched peers in areas such as intellectual reasoning, creativity, or the arts, and is associated with the need for differentiated educational experiences (National Association for Gifted Children (NAGC), 2019, p. 1). When giftedness co-occurs with autism, the resulting profile, often termed twice-exceptional autistic (2e-AS), is characterised by uneven or asynchronous development, in which advanced cognitive strengths exist alongside challenges in social communication, executive functioning, and adaptive behaviour (Assouline et al., 2012; Foley Nicpon et al., 2011).

Interest in twice-exceptional autistic learners (2e-ASL) has grown in recent years, moving beyond simple recognition toward more detailed descriptions of their cognitive and psychosocial profiles (Lim, 2021; Gelbar et al., 2021). Many demonstrate strengths in verbal reasoning, pattern recognition, creativity, and academic achievement, while also experiencing difficulties in areas such as processing speed, executive functioning, and everyday adaptive skills (Cain et al., 2019; Doobay et al., 2014; Michaelson et al., 2021). Rather than acting as a protective factor, higher IQ can intensify the mismatch between cognitive potential and real-world performance (Charman et al., 2011; Cain et al., 2019; Michaelson et al., 2021).

High intellectual ability also affects how autism is recognised. Autistic individuals with higher IQ are more likely to develop compensatory strategies that mask social-communication differences, contributing to delayed or missed diagnoses (Ratto et al., 2018; Wolff et al., 2022). Population-based studies show that individuals with IQ scores in the gifted range are far more likely to reach adolescence or adulthood without an autism diagnosis, reducing access to early and appropriate supports (Katusic et al., 2021; Michaelson et al., 2021). Despite their cognitive strengths, 2e-ASL are not shielded from psychosocial risk. Compared to gifted peers without autism, they experience higher rates of anxiety, depression, and social isolation, with evidence suggesting that emotional distress increases rather than decreases at higher levels of intellectual ability (Casten et al., 2023; Foley Nicpon et al., 2010; Michaelson et al., 2021).

Together, these findings point to a clear need for interventions that are responsive to the distinctive, asynchronous profiles of 2e-ASL. In response, recent scholarship has proposed conceptual frameworks, dual differentiated instructional models, and strengths-based approaches intended to support this population better (Austermann et al., 2025; Lim, 2021; Reis & Renzulli, 2025). These approaches emphasise recognising strengths, aligning learning with interests, and providing structured support for areas of difficulty. Despite growing conceptual clarity, empirical evidence to guide interventions remains limited. A systematic review published in 2022 found no empirically evaluated interventions specifically designed for 2e-ASL, highlighting a substantial gap between theory and practice (Gelbar et al., 2021). Much of the existing literature remains descriptive, focusing on identification challenges rather than on tested intervention strategies. At the same time, the broader autism intervention literature has expanded rapidly but rarely examines moderators such as intellectual ability, and many studies restrict sample heterogeneity in ways that limit moderation analyses (Klinger et al., 2021; Wolery & Garfinkle, 2002). As a result, 2e-ASL are likely included within general autism intervention studies, yet their specific needs, strengths, and outcomes are seldom examined separately. However, research and practice consistently indicate that autistic learners with high cognitive ability are frequently under-identified as 2e, or are retrospectively identified, and are often included within broader autism or 2e samples without explicit reporting or tailored intervention design (Antshel et al., 2007; Foley Nicpon et al., 2011; Reis et al., 2014; Ronksley-Pavia, 2015). As a result, intervention studies that include 2e-ASL are rarely synthesised in relation to their dual profiles, leaving a substantial body of potentially relevant evidence unexamined (Gelbar et al., 2021; Kasari & Smith, 2013).

Accordingly, a systematic synthesis is needed to bridge this gap. By examining interventions that either explicitly target 2e-ASL or implicitly include them within broader high-ability autism samples, the present review seeks to clarify how existing interventions address the unique, asynchronous profiles of 2e-ASL, identify intervention characteristics associated with positive outcomes, and inform the development of integrated and scalable supports. This gap has important implications for practice. Without a clear synthesis of how existing interventions include, accommodate, or impact 2e-ASL, clinicians and educators lack guidance on whether commonly used autism interventions are appropriate for this group, or which intervention features may best support their distinctive profiles.

The present review extends prior work by adopting a broader inclusion framework that captures both interventions specifically designed for 2e-ASL and interventions that included 2e-ASL within more general samples. This approach reflects the realities of service provision and research practice, where twice-exceptionality is frequently under-identified, inconsistently reported, or identified retrospectively. By doing so, this review offers a more comprehensive and ecologically valid understanding of how existing interventions function for 2e-ASL and identifies opportunities for adaptation, scalability, and future research.

Research Aim and Questions

This review aims to analyse the available evidence on interventions that have included or targeted 2e-AS children and adolescents (age 4–18 years) by examining the available evidence on interventions that have included or targeted this population. Specifically, it explores three key questions:

• What interventions were implemented that either explicitly targeted 2e-ASL learners or included 2e-ASL learners, and how were these interventions characterised in terms of intervention type, outcome, setting, and duration?
• How effective were these interventions in improving developmental, behavioural, and academic outcomes for 2e-ASL students?
• What recommendations can be drawn for designing scalable, evidence-based supports?

2. Methods

2.1. Review Framework

This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; Page et al., 2021) and was prospectively registered with PROSPERO (CRD420250650864). Study selection was guided by the Population, Intervention, Comparison, Outcomes, and Study design (PICOS) framework.

2.2. Conceptual Definitions

2e-ASL were defined as individuals meeting diagnostic criteria for autism spectrum disorder (DSM-5, American Psychiatric Association, 2022) and demonstrating high intellectual ability. Intellectual ability was operationalised as IQ ≥ 115 or evidence of superior academic achievement. A threshold of IQ ≥ 115 was adopted to capture a broader 2e-ASL population, as stricter cut-offs risk excluding students who may still require targeted support (Foley Nicpon et al., 2011; Reis et al., 2014; Ronksley-Pavia, 2015). This aligns with evidence that giftedness and disability can mask each other (Burger-Veltmeijer et al., 2011; Gelbar et al., 2021) and that autistic youth with above-average IQ are often under-identified (Wolff et al., 2022).

2.3. Eligibility Criteria

Studies were included if they: (a) reported on children or adolescents identified as both autistic and gifted. Giftedness was defined based on reported IQ scores, evidence from specific IQ or IQ ranges consistent with a high-ability profile, or explicit identification of giftedness or 2e within the study; (b) examined an intervention in educational, clinical, home, or community settings; and (c) presented empirical findings on at least one developmental, behavioural, or academic outcome. Only peer-reviewed English-language articles were included. Excluded were studies without explicit reference to both autism and giftedness, conceptual/review-only papers, and unpublished grey literature. Full eligibility criteria are listed in Appendix A (Table A1).

2.4. Search and Screening Procedures

The PRISMA flow diagram (Figure 1) outlines the study selection process. Eight databases (MEDLINE, PubMed, CINAHL, ERC, ERIC, PsycINFO, SCOPUS, Web of Science) were searched using terms across four domains: AS (e.g., “autism,” “Asperger’s”), giftedness (e.g., “gifted,” “talented”), interventions (e.g., “program,” “treatment”), and settings (e.g., “school,” “clinic,” “home”). Boolean operators and truncation were used, and terms were adapted for each database. No date restrictions were applied. The full search strategy is provided in the Appendix B (Table A2). The final database search was conducted in February 2025. Reference lists of included studies and prior reviews were also hand-searched. Search results were imported into Zotero for duplicate removal, then uploaded to Covidence for screening. Two reviewers independently screened titles/abstracts and full texts; disagreements were resolved through discussion or consultation with a third reviewer.

Figure 1. PRISMA flow diagram depicting study selection.

2.5. Data Extraction, Quality Appraisal, and Synthesis

Data extraction was conducted by the first reviewer (A.R.). A structured form was used to extract study details (author, year, country, design, sample size, age, IQ classification), intervention characteristics (type, duration, delivery, setting), and outcomes (academic, developmental, behavioural). For mixed samples, only data specific to 2e-ASL participants were included. Methodological quality was assessed using the Mixed Methods Appraisal Tool Version 2018 (MMAT, Hong et al., 2018) by two independent reviewers (A.R., C.B.), with consensus reached through discussion.

Given substantial heterogeneity across studies in intervention design, participant characteristics, outcome measures, and reporting practices, particularly the frequent absence of precise or disaggregated IQ data, statistical pooling was not appropriate. A narrative synthesis approach was therefore adopted. Studies were organised by intervention focus, delivery setting (school, clinical/research, home, or community), and outcome domain. Separate narrative syntheses were conducted for (a) studies in which 2e-ASL were embedded within broader autism samples, and (b) studies explicitly designed for 2e-ASL. Within each synthesis, findings were examined at two levels. First, intervention characteristics associated with positive outcomes were analysed within outcome domains (e.g., social, emotional, academic, behavioural, intellectual). Second, cross-cutting intervention characteristics associated with positive effects across outcome domains were identified. Where reported, the potential moderating role of IQ and related indicators of high ability (e.g., language ability) on intervention response was also considered.

3. Results

3.1. Study Selection

The final search identified 7611 records. After removing duplicates and conducting preliminary screening, 605 full-text articles were assessed for eligibility. Of these, 511 were excluded, and 43 studies couldn’t be retrieved, leaving 51 studies for the final synthesis.

The most common reason for exclusion was insufficient reporting of IQ data (n = 206; 40.31%), where studies only provided vague descriptors such as “IQ above 70” or “IQ above 100.” Other major reasons included participants not meeting the giftedness threshold of IQ ≥ 115 (n = 51; 9.98%), grey literature such as theses, book chapters, or conference proceedings (n = 176; 34.44%), and studies limited to adult samples (n = 21; 4.11%). Additional exclusions were due to inaccessible full texts (n = 43; 8.41%), non-English publications (n = 24; 4.7%), and ineligible designs such as editorials or reviews (n = 29; 5.67%). All included studies were peer-reviewed and reported extractable data consistent with the eligibility criteria.

3.2. Descriptive Characteristics of Included Studies

Table 1 provides an analysis of the 51 included studies, revealing several key trends regarding research focus, setting, and methodological detail.

Table 1. Characteristics of included study (n = 51).

		N (%)
Study sample	2e-Specific	48 (94.12%)
	2e-Inclusive	3 (5.88%)
IQ mentioned	Only the IQ range	34 (66.67%)
	Specific IQ mentioned	14 (27.45%)
	Gifted/2e	3 (5.88%)
Year of publication	2000–2010	17 (33.33%)
	2011–2015	12 (23.53%)
	2016–2020	11 (2.57%)
	2021–2024	11 (21.57%)
Setting	School	18 (35.29%)
	Home	2 (3.92%)
	Community	2 (3.92%)
	Clinical/Research setting	29 (56.86%)
Type of Intervention	Single-domain studies	29 (56.86%)
	Multi-domain studies	22 (43.13%)
Outcome	Social	27 (52.94%)
	Emotional	23 (45.09%)
	Intellectual	11 (21.56%)
	Behavioural	8 (15.68%)
	Language	3 (5.88%)
	Academic	4 (7.84%)
	Sensory/Perceptual	2 (3.92%)
	Physical	0

3.2.1. Study Sample

Only three (5.88%) were 2e-specific, with interventions explicitly developed for 2e populations. Of these, two studies specifically targeted 2e-ASL, while one focused on a broader 2e population that included a 2e-ASL subgroup. The remaining studies (94.12%, n = 48) were 2e-inclusive, meaning they included 2e participants within the study sample but did not design interventions specifically for them.

The results suggest that interventions specifically designed for 2e-ASL remain limited, with most evidence derived indirectly from broader autism-focused studies. As a result, it is difficult to determine whether existing approaches adequately address the distinct cognitive and developmental profiles associated with 2e-ASL.

3.2.2. IQ Reporting

Among the studies that were 2e-inclusive, participant IQ reporting varied considerably across studies. Most provided only an IQ range (66.67%, n = 34), while a smaller number reported specific IQ scores (27.45%, n = 14). This inconsistency limits the ability to extract comparable data across studies. Precise IQ reporting could have supported more accurate synthesis and clearer evaluation of how outcomes may differ across levels of cognitive ability.

The lack of consistent IQ reporting also makes it difficult to examine whether intervention outcome differs according to level of giftedness or specific cognitive strengths.

3.2.3. Year of Publication

A total of 51 studies published between 2000 and 2025 met the inclusion criteria. Research activity was relatively consistent across this period. Approximately one-third of the studies (33.3%) were published between 2000 and 2010, with steady output in subsequent years. More recently, 29.4% of studies were published between 2021 and 2024, indicating continued and growing interest in this area. Notably, all three interventions specifically designed for the 2e population were published within this most recent period, suggesting an emerging focus on tailored approaches for 2e individuals.

3.2.4. Research Settings

Interventions were delivered across a range of settings. Most of the interventions were delivered in controlled settings like clinical or university-based research lab contexts (56.86%). School-based interventions were the next-largest group (35.29%), while home and community settings were rarely used, together comprising less than 7.84% of the included studies.

The predominance of highly controlled research settings may limit understanding of how interventions translate into everyday educational and family contexts. The relative absence of home and community-based approaches also highlights a gap in ecologically valid and generalisable intervention research for 2e-ASL.

3.2.5. Intervention Type and Outcomes

In terms of intervention design, some studies focused on a single outcome domain (e.g., social or intellectual), while others adopted a multi-domain approach, targeting combinations such as socio-emotional or socio-emotional–intellectual outcomes. Studies were relatively evenly distributed between single-domain interventions (n = 29) and multi-domain interventions (n = 22).

Social outcomes were the most commonly assessed (n = 27), followed by emotional outcomes (n = 23). Intellectual outcomes were examined in a smaller number of studies (n = 11). In contrast, behavioural, academic, language, and sensory/perceptual outcomes were addressed less frequently. None of the included studies focused on physical outcomes.

This pattern suggests that intervention research has primarily prioritised social and emotional functioning, reflecting dominant priorities within broader autism intervention literature. In comparison, domains more closely associated with 2e, such as advanced academic development, intellectual strengths, and asynchronous cognitive profiles, appear comparatively underexplored. The limited focus on academic, language, sensory, and physical outcomes also highlights important gaps in understanding the broader developmental needs of 2e-ASL.

3.3. IQ Characteristics of Participants

Table 2 shows that across the 14 studies that reported individual IQ scores, verbal, performance, and full-scale indices show wide variability. Mean VIQ (78.95) and PIQ (73.49) fall in the low-average range, while mean FSIQ (80.75) is just below the typical population average (100). Minimum scores dip to the borderline-deficient range (VIQ = 51, PIQ = 48, FSIQ = 35) and maximum scores reach the highly gifted spectrum (VIQ = 168, PIQ = 147, FSIQ = 160). The overall proportion of 2e-ASL among these participants is 32.9%, indicating that roughly one-third of the sample is identified as gifted even though there is a broad IQ distribution.

Table 2. Descriptive Statistics of IQ Ranges and Percentage of Gifted Population from the 14 Studies Providing Separate IQ Data.

IQ Measure	Mean of Lowest Scores	Absolute Min	Mean of Highest Scores	Absolute Max	Overall Range
VIQ (N = 27)	78.95	51	127.15	168	117
PIQ (N = 21)	73.49	48	129.43	147	99
FSIQ (N = 38)	80.75	35	131.35	160	125

The wide IQ ranges reported within individual studies may limit understanding important differences in intervention outcome associated with cognitive ability and uneven developmental profiles. The substantial heterogeneity present within the included samples illustrates the difficulty of identifying and interpreting outcomes for 2e-ASL when they are embedded within broader autism cohorts.

3.4. Characteristics and Outcomes of Interventions That Were Inclusive of 2e-ASL

Table 3 included synthesis of intervention studies with mixed samples that either (a) reported IQ-specific findings confirming the inclusion of 2e-ASL, or (b) reported IQ ranges encompassing the high-IQ levels relevant to 2e-ASL. This approach allowed the identification of studies likely to include 2e-ASL even when they were not explicitly labelled as 2e-AS and were embedded within broader autism samples. Studies were organised by intervention focus and outcome domain. Table 3 presents the narrative synthesis, summarising intervention focus, intervention type, representative studies, and the key characteristics of interventions associated with positive outcomes within each domain.

Table 3. Intervention Types, Outcome Focus, and Key Characteristics of Included Interventions.

Intervention Focus	Intervention Type	Example Interventions (Citations)	Key Characteristics of the Interventions
Sensory/Perceptual
Foundational neural and perceptual mechanisms (neural self-regulation, neuroplasticity, multisensory temporal integration)	Technology-mediated physical training interventions	Mu-Rhythm Neurofeedback Training (Datko et al., 2018); Audiovisual Temporal Binding Training (Feldman et al., 2023)	Target low-level body functions rather than overt behaviour; Use neurofeedback or computer-based perceptual learning; Highly structured, repetitive training paradigms; Explicit, real-time performance or neural feedback; Delivered in controlled settings; Limited functional generalisation; moderated by baseline cognitive and language ability
Social
Conversational competence (conversation initiation/maintenance, joining groups, social planning) Social communication and pragmatic skills (joining in, maintaining interaction, greetings, group participation)	Manualised social skills programmes (CBT-informed); Modelling-based and technology-supported interventions	PEERS for Adolescents (Chang et al., 2014; Hill et al., 2017; Lordo et al., 2017; Lao et al., 2024); SCORE Skills Strategy Training (Webb et al., 2004); Outpatient Social Skills Group (Barry et al., 2003); Social Stories + Video Modelling (Sansosti & Powell-Smith, 2008); Social Skills Training ± Parent/Teacher Involvement (Dekker et al., 2019)	Explicit teaching of conversational rules and social scripts; Structured role-play and rehearsal; Video or narrative modelling of target verbal or non-verbal behaviours; Feedback and reinforcement; Parent-assisted homework to support generalisation; Explicit demonstration before real-world contexts; Structured practice with feedback
Play, interaction, and joint engagement (reciprocal play, collaboration, sustained interaction)	Peer-mediated and play-based interventions	Preschool Peer Social Intervention (Bauminger-Zviely et al., 2020); SCIPYC (Minne & Semrud-Clikeman, 2012); Peer-Mediated Recess/Recess Pals (Rodríguez-Medina et al., 2016); Music-Mediated Intervention (MacDonald-Pregent et al., 2023)	Peers as models and interaction partners; Structured and semi-structured play activities; Adult facilitation and scaffolding; Emphasis on reciprocal engagement and shared activity; Practice in naturalistic settings (playground, classroom)
Social cognition (Theory of Mind, perspective-taking, social cue interpretation, social judgement)	CBT-based social–cognitive interventions	SCI/SCI-E/SCI-A (Stichter et al., 2010, 2012, 2016; Schmidt et al., 2011); Social Adjustment Enhancement Intervention (Solomon et al., 2004); Cognitive–Behavioural Social–Emotional Program (Bauminger, 2002); CB-E Social Skills Training (Bauminger, 2007)	Explicit instruction in social reasoning and mental-state understanding; Emotion and cue recognition training; Scaffolded discussion and guided practice; Repeated application to social scenarios; Integration of cognitive and behavioural components
Social reciprocity and flexible responding (adaptive participation across contexts)	Multi-component and ecological social interventions	AR + Social Stories + CT Games (I.-J. Lee & Hsu, 2024); CMAR-ST + CMMR-ST (I.-J. Lee & Huang, 2024); Collaborative Tech + CBT (Bauminger-Zviely et al., 2013); Summer Treatment Program (Mitchell et al., 2015)	Multi-modal instruction (visual, digital, physical activities); Structured frameworks for complex social situations; Emphasis on flexibility and real-world application
Emotional
Emotional understanding (emotion recognition, categorisation, interpretation, emotional awareness)	CBT-based and rule-based emotional learning interventions	CB-E Social Skills Training (Bauminger, 2007); Social Adjustment Enhancement Intervention (Solomon et al., 2004); SCI/SCI-E/SCI-A (Stichter et al., 2010, 2012, 2016); SCIPYC (Minne & Semrud-Clikeman, 2012); JeStiMulE Serious Game (Serret et al., 2014); Secret Agent Society (Sofronoff et al., 2017)	Explicit instruction in emotion recognition and interpretation; Rule-based categorisation of emotions; Visual, multisensory, or play-based supports; Modelling and guided practice with feedback; Repeated exposure to emotional stimuli
Emotional regulation (arousal control, anxiety reduction, emotional coping, adaptive responding)	CBT-informed and mindfulness-based regulation interventions	CBT for Anger Management (Sofronoff et al., 2007); Soles of the Feet Mindfulness (Shah et al., 2022); Music-Mediated Intervention (MacDonald-Pregent et al., 2023); Junior Detective Training Program (Beaumont & Sofronoff, 2008); CBT for Anxiety in AS (Sofronoff et al., 2005); MCIT (White et al., 2009); Coping Group—Fighting Worry & Facing Fears (J. A. Reaven et al., 2009); Facing Your Fears (J. Reaven et al., 2012a); Facing Your Fears—Adolescent (J. Reaven et al., 2012b); PEERS—emotional coping component (Lordo et al., 2017)	Explicit teaching of regulation and coping strategies; Behavioural rehearsal, role-play, and guided practice; Attention-redirection and somatic grounding techniques; Graded exposure for anxiety reduction; Structured, scaffolded learning with visual supports; Parent or adult involvement to support generalisation
Behavioural
Independent functioning in daily routines (daily transitions, daily living skills, organisational behaviour, self-monitoring)	Behavioural skills training and self-management interventions	PDA/iPod Prompting (Palmen et al., 2012); Surviving & Thriving in the Real World (Duncan et al., 2018); File Box System + Self-Monitoring (Dorminy et al., 2009)	Targets functional, observable behaviours in everyday routines; Structured routines and task breakdown; Use of prompts, visual supports, digital aids, or organisational systems; Self-monitoring to promote independence and maintenance; Systematic reduction of adult prompting and supervision; Repeated practice in naturalistic settings
Reduction of off-task, disruptive, or challenging behaviour (off-task behaviour, solitary behaviour, disruptive behaviour, immediate self-control)	Behavioural regulation interventions	Peer-Mediated Recess—behavioural component (Rodríguez-Medina et al., 2016); Soles of the Feet Mindfulness (Shah et al., 2022); Parent–Child Interaction Therapy (Solomon et al., 2008)	Focus on decreasing frequency and intensity of observable problem behaviours; Use of reinforcement and clear behavioural expectations; Immediate regulation strategies (e.g., mindfulness-based attention redirection); Real-time feedback or coaching; Measurement of behaviour change in context
Appropriate behavioural responding and regulation (appropriate responses, anger-related behaviour, adaptive responding)	CBT-informed behavioural regulation interventions	CBT for Anger Management (Sofronoff et al., 2007); Parent–Child Interaction Therapy (Solomon et al., 2008)	Explicit teaching of appropriate behavioural responses; Behavioural rehearsal, modelling, and role-play; Practice across settings (home and school); Emphasis on replacing maladaptive behaviour with adaptive responses
Academic
Strategic engagement with academic tasks, including self-regulated learning behaviours and effective test-taking behaviours to improve performance and reduce test anxiety	Comprehensive, strategy-based academic training programmes.	PIRATES test-taking strategy (Songlee et al., 2008); Discourse Comprehension (Åsberg & Sandberg, 2010); Small group question answer training (Palmen et al., 2008)	Explicit instruction in academic strategies; mnemonic devices; modelling and thinking-aloud demonstrations; verbal rehearsal; role-play and game-based practice; guided and advanced practice with feedback; self-management and self-monitoring strategies
Intellectual
Executive functioning, cognitive flexibility, and problem-solving (planning, organisation, inhibition, flexibility, goal-setting, real-life problem solving)	Cognitive–behavioural and CBT-based executive-function curricula; Metacognitive and psychoeducational group interventions	Cognitive–behavioural and CBT-based executive-function curricula (Kenworthy et al., 2014; Pugliese et al., 2024; Stichter et al., 2010, 2012, 2016; Solomon et al., 2004). Metacognitive and psychoeducational group interventions (Goodman et al., 2017). Behaviourally structured, school-style programmes (Mitchell et al., 2015)	Explicit instruction in EF concepts (planning, flexibility, goal-setting, self-monitoring); direct teaching of the skill, why it matters, and how to apply it; structured problem-solving frameworks; modelling and thinking-aloud demonstrations; visual and verbal cueing; scaffolded and guided practice with systematic fading; repeated rehearsal across tasks; goal-plan-do-check routines; behaviour management and reinforcement systems; parent and teacher training to support generalisation across settings
Language
Higher-level language processes (discourse comprehension; grammatical form acquisition)	Instruction-based language interventions	Discourse Comprehension (QAR-based) (Åsberg & Sandberg, 2010); Explicit–Implicit vs. Implicit-only Grammar Instruction (Bangert et al., 2019)	Adult-guided or computer-mediated delivery; Explicit, structured teaching of rules, strategies, or linguistic forms; Direct strategy instruction for discourse comprehension and direct rule explanation for grammar; Implicit exposure (e.g., modelling, recasting) used but insufficient alone; Scaffolding through modelling, guided practice, and corrective feedback; Highly structured tasks with repeated practice and controlled input; Clearly defined, narrowly targeted learning objectives

Note: Bold text denotes the intervention domain. Interventions, example studies, and key characteristics have been synthesised and grouped according to the domain targeted by the intervention.

Across studies, interventions most frequently targeted social domains, followed by emotional and intellectual domains, with fewer studies addressing behavioural, academic, language, or sensory/perceptual outcomes. None of the interventions focused on physical outcomes. Most interventions were delivered in clinical or school settings, with comparatively limited implementation in home or community contexts.

3.4.1. Sensory/Perceptual Outcomes

Only a small number of studies (3.92%) addressed sensory/perceptual outcomes, focusing on underlying neural or sensory mechanisms rather than observable functional behaviour. These interventions employed technology-mediated, mechanistic training approaches, such as neurofeedback and computer-based perceptual learning, delivered in controlled settings (Datko et al., 2018; Feldman et al., 2023). Across these studies, improvements were observed at the level of neural activation or perceptual processing; however, behavioural generalisation was limited. Importantly, intervention impact was moderated by baseline cognitive characteristics, with higher nonverbal IQ predicting greater gains and transfer in perceptual learning tasks (Feldman et al., 2023).

Findings suggest that sensory/perceptual interventions may be more effective when aligned with learners’ cognitive strengths, particularly nonverbal reasoning abilities. However, the limited evidence base and lack of demonstrated functional generalisation highlight the need for further research examining how gains in perceptual processing translate into meaningful real-world outcomes for 2e-ASL.

3.4.2. Social Outcomes

Social outcomes represented the most extensively studied domain. Interventions targeted multiple layers of social functioning, ranging from foundational conversational skills and pragmatic communication to peer interaction, play, and higher-order social cognition. Across studies, social interventions were predominantly explicit, structured, and rule-based, frequently incorporating modelling, role-play, guided rehearsal, and performance feedback (Webb et al., 2004; Stichter et al., 2010, 2012, 2016).

Manualised social skills programmes demonstrated consistent improvements in conversational competence and social participation, particularly when structured practice was paired with adult or parent-supported generalisation (Chang et al., 2014; Hill et al., 2017; Lordo et al., 2017; Dekker et al., 2019). Peer-mediated and play-based interventions extended social learning into naturalistic contexts, supporting reciprocal engagement and generalisation beyond the intervention setting (Bauminger-Zviely et al., 2020; Rodríguez-Medina et al., 2016). Multi-component and technology-supported social interventions emphasised flexibility and contextual adaptation, though outcomes varied across studies (Bauminger-Zviely et al., 2013; I.-J. Lee & Hsu, 2024; I.-J. Lee & Huang, 2024).

Findings suggest that social interventions were most effective when they combined explicit instruction with opportunities for supported practice and real-world application.

3.4.3. Emotional Outcomes

Interventions targeting emotional outcomes included both emotional understanding and emotional regulation interventions. Emotional understanding interventions focused on supporting emotion recognition, categorisation, and interpretation. These interventions explicitly taught participants how to recognise and interpret emotional cues, often breaking emotions down into clear, structured, and learnable categories. They commonly incorporated visual supports, multisensory activities, or play-based approaches to enhance engagement and accessibility. In addition, interventions often included modelling of emotional responses, guided practice, and feedback to support skill development. Repeated exposure to a range of emotional situations was used to reinforce learning and strengthen understanding over time. Interventions targeting emotional understanding used explicit instruction and categorisation strategies, supported by visual, multisensory, or game-based materials (Bauminger, 2007; Serret et al., 2014).

Emotion regulation interventions focused on helping individuals manage and respond to their emotional experiences in adaptive ways. These interventions were largely CBT-informed, targeting anxiety, anger, and emotional coping through structured skill instruction, behavioural rehearsal, and graded exposure and included mindfulness-based regulation strategies (Sofronoff et al., 2005, 2007; J. A. Reaven et al., 2009; J. Reaven et al., 2012a; White et al., 2009). Across studies, improvements in emotional regulation were most consistently reported when interventions incorporated scaffolded practice and caregiver involvement, supporting application beyond the therapy context.

Emotional interventions appeared most effective when emotional concepts were explicitly taught and repeatedly practised across supported contexts.

3.4.4. Behavioural Outcomes

In the included studies, behavioural interventions focused on building independence in everyday routines, reducing challenging behaviours, and promoting appropriate behavioural responding and regulation. These interventions commonly targeted areas such as organisational behaviour, transitions, and self-monitoring, as well as the reduction of off-task or disruptive behaviours. They were typically delivered through behavioural skills training and self-management approaches (Dorminy et al., 2009; Palmen et al., 2012; Duncan et al., 2018).

Across studies, effective behavioural interventions shared an emphasis on task structure, visual or digital supports, and adult prompts, with behaviour change measured in naturalistic environments. Mindfulness- and CBT-based approaches targeting behavioural regulation showed reductions in problem behaviours, particularly when strategies were practised across settings (Shah et al., 2022; Solomon et al., 2008).

Findings suggest that behavioural interventions were most effective when supports were explicit, structured, and consistently reinforced across contexts. The importance of self-monitoring and scaffolded regulation strategies may be particularly relevant for 2e-ASL, who may experience uneven executive functioning despite areas of cognitive strength.

3.4.5. Academic Outcomes

They were addressed in a limited number of studies and focused on strategic engagement with academic tasks rather than subject-specific content. These interventions aimed to develop specific skills that support learning and performance in academic settings. Interventions emphasised explicit instruction in learning strategies, self-regulation, and test-related coping, incorporating modelling, rehearsal, and guided practice. Reported gains were most evident in strategic behaviour and engagement rather than broad academic achievement.

The relatively limited focus on academic outcomes is notable given that advanced academic ability is a defining feature of many 2e-ASL. This may suggest that intervention research has prioritised remediation of difficulties over the development and support of intellectual and academic strengths.

3.4.6. Intellectual Outcomes

Interventions within the intellectual domain predominantly target executive functioning, cognitive flexibility, and problem-solving as interrelated cognitive processes supporting independent functioning and adaptive learning. Across studies, these outcomes are addressed primarily through explicit cognitive–behavioural and curriculum-based behavioural interventions, often incorporating ABA-informed principles such as structured routines, reinforcement, and systematic skill practice, and delivered in structured group formats within school or clinical settings (Kenworthy et al., 2014; Pugliese et al., 2024; Stichter et al., 2010, 2012, 2016; Solomon et al., 2004; Mitchell et al., 2015), with additional use of metacognitive psychoeducational approaches to support self-awareness and strategy use (Goodman et al., 2017).

Findings suggest that intellectual interventions were most effective when higher-order thinking skills were explicitly scaffolded and linked to practical strategy use. However, despite the relevance of advanced cognitive ability within 2e-ASL, relatively few interventions directly targeted gifted-level reasoning, creativity, or intellectual enrichment.

3.4.7. Language Outcomes

These were examined in a small subset of studies and focused on higher-level language processes, including discourse comprehension and grammatical learning. Interventions relied on explicit instructional approaches, delivered through adult-guided or computer-based formats. Explicit teaching of rules and strategies was associated with stronger immediate gains than implicit exposure alone, though maintenance and generalisation were variable and often moderated by baseline language ability (Åsberg & Sandberg, 2010; Bangert et al., 2019).

The variability in generalisation highlights the importance of considering individual language and cognitive profiles when designing intervention supports.

3.5. Characteristics and Outcomes of Interventions Specifically Developed for 2e-ASL

Table 4 summarises intervention studies that explicitly identified participants as 2e-ASL and implemented interventions specifically designed for this population. Only three interventions within the included literature were explicitly designed for 2e learners. Of these, two interventions were specifically developed for 2e-ASL (Kahveci & Güneyli, 2023; Gaber, 2022), while one study targeted a broader 2e sample, within which autistic learners constituted a subgroup (Aldawsari et al., 2023). In all three studies, positive effects were observed within the targeted outcome domains, including intellectual, academic, behavioural, and engagement-related outcomes.

Table 4. Studies with 2e-specific interventions participants, IQ, intervention names, delivery, and outcomes.

Intervention (Name + Citation)	Domain	Intervention Description	Setting of Delivery	Duration/Frequency	Key Findings	Efficacy Signal
CBC + Visual Schedules + Curriculum Adaptation (Kahveci & Güneyli, 2023)	Language + Behavioural + Academic	Strength-based, individualised intervention involving needs assessment, visual-schedule workbooks, and curriculum adaptations aligned with student strengths, alongside instruction in self-monitoring and on-task strategies; implemented collaboratively across intervention, maintenance, and generalisation phases.	School-home conjoint	Multi-week phased	On-task behaviour or ability to remain focused on assigned tasks improved ↑; academic engagement imrpoved ↑; strongest when work is interest-linked/challenging	Positive
Training to develop attitude toward creativity (Gaber, 2022)	Intellectual (creativity attitude)	Creativity-focused training program that uses a structured, multi-stage curriculum and interactive techniques like brainstorming, free play, and dialogue to encourage the production of original ideas and foster positive emotional attitudes toward their talents	School	26 sessions (3×/week; 40 min)	Creativity attitude improved ↑; gains maintained at 2-month follow-up; high engagement	Positive
SCAMPER technique program (Aldawsari et al., 2023)	Intellectual (productive thinking)	Arts-based, strength-focused intervention that uses structured, question-driven techniques to stimulate idea generation, creative synthesis, and critical analysis in artistic work	School	4 main phases in 12 stages	Productive-thinking improved ↑ (ASD p = 0.006); ASD subgroup improved least vs. LD/ADHD	Positive

Across all three studies, interventions were strength-based by design which focused on improving interests, creativity, and cognitive strengths rather than prioritising remediation of autistic difficulties (Gaber, 2022; Kahveci & Güneyli, 2023; Aldawsari et al., 2023). Curriculum adaptation and creative challenge were central mechanisms which were promoted either through interest-linked academic tasks and visual organisation supports (Kahveci & Güneyli, 2023) or arts-based and creativity-focused programmes that encouraged idea generation, originality, and reflective thinking (Gaber, 2022; Aldawsari et al., 2023). In contrast to deficit-oriented skill training, these interventions positioned 2e-ASL as active producers of ideas, rather than recipients of corrective instruction. Rather than remediating deficits, these interventions explicitly focused on learners’ existing strengths, interests, and talents as primary mechanisms for engagement and skill development.

School-based creativity interventions were specifically developed for 2e-ASL and focused on productive thinking, originality, and attitudes toward creativity (Gaber, 2022; Aldawsari et al., 2023). These programmes used structured yet flexible activities, including brainstorming, free play, dialogue, and question-driven techniques (e.g., SCAMPER), to encourage idea generation and creative expression. Both studies reported positive gains in creativity-related outcomes, with high levels of engagement. Gains were maintained beyond the immediate intervention period where follow-up data were available. However, when autistic learners were analysed as a subgroup within a broader 2e sample (Aldawsari et al., 2023), improvements, although statistically significant, were more modest compared to peers with other exceptionalities, suggesting possible differences in responsiveness within 2e population, which implies the need for further investigation.

Across studies, positive outcomes were reported consistently, including improvements in on-task behaviour and academic engagement (Kahveci & Güneyli, 2023), attitudes toward creativity and sustained motivation (Gaber, 2022), and productive thinking skills (Aldawsari et al., 2023). Importantly, these gains were not limited to cognitive outputs but extended to behavioural regulation and emotional engagement, suggesting that strength-based approaches may indirectly support areas of difficulty by enhancing autonomy, competence, and motivation.

However, findings also indicate heterogeneity in responsiveness within autistic samples. In the SCAMPER intervention, autistic participants showed smaller gains than peers with other neurodevelopmental profiles, despite overall positive effects (Aldawsari et al., 2023), highlighting the need for further refinement and individualisation even within strengths-based models.

Three interventions specifically designed for 2e-ASL share a common profile: they are interest-driven, creativity-oriented, cognitively challenging, and scaffolded rather than corrective.

3.6. Effective Intervention Characteristics for 2e-ASL

Intervention characteristics were examined across outcome domains to identify features consistently associated with positive effects across the selected intervention studies.

Across outcome domains, interventions that showed positive effects for 2e-ASL tended to share a common set of design features. Most effective programmes were highly structured, explicit, and carefully scaffolded, regardless of whether they targeted social, emotional, cognitive, academic, or behavioural outcomes (Chang et al., 2014; Stichter et al., 2016; Kenworthy et al., 2014; Sofronoff et al., 2017). A consistent element across these studies was the use of explicit instruction, in which skills were clearly defined, their purpose explained, and application demonstrated. This approach was particularly prominent in interventions addressing social cognition, executive functioning, and language, where abstract concepts benefit from clear, rule-based explanations (Bauminger, 2007; Stichter et al., 2010; Åsberg & Sandberg, 2010).

Visual supports and modelling played a central role across intervention types. Many programmes relied on visual schedules, scripts, video modelling, worked examples, or structured frameworks to reduce cognitive load and make expectations more concrete (Sansosti & Powell-Smith, 2008; Kenworthy et al., 2014; Dekker et al., 2019). These supports were especially evident in social-communication, behavioural, and executive-function interventions, where step-by-step guidance and visual cues helped learners organise their responses and complete tasks more independently.

Another shared feature of effective interventions was the emphasis on guided practice with feedback. Interventions routinely provided repeated opportunities to rehearse target skills in a supported way, alongside timely feedback and reinforcement (Webb et al., 2004; Stichter et al., 2012; Sofronoff et al., 2007). Role-play, structured rehearsal, and guided problem-solving were commonly used to help consolidate learning before skills were applied in less structured, real-world situations, particularly in social and emotional domains (Solomon et al., 2004; Bauminger, 2002).

Interventions that appeared more functionally meaningful were often embedded within everyday contexts. Programmes delivered in schools, playgrounds, or routine daily activities were more likely to support generalisation than those implemented solely in controlled clinical or laboratory settings (Bauminger-Zviely et al., 2020; Rodríguez-Medina et al., 2016; Mitchell et al., 2015). Involvement of parents and teachers further strengthened outcomes by extending learning beyond the intervention sessions and promoting consistency across settings (J. A. Reaven et al., 2009; Dekker et al., 2019; Lordo et al., 2017).

Technology-enhanced approaches were used across several domains but did not dominate the intervention landscape. When effective, technology typically functioned as a supporting element, enhancing visualisation, structure, or feedback rather than serving as the primary mechanism of change (Sansosti & Powell-Smith, 2008; Serret et al., 2014; I.-J. Lee & Hsu, 2024).

A different pattern emerged in the small subset of interventions explicitly designed for 2e-ASL. These studies consistently adopted strength-based and interest-aligned approaches, using creativity, advanced reasoning, or academic strengths as starting points for engagement and learning (Gaber, 2022; Aldawsari et al., 2023; Kahveci & Güneyli, 2023). Creativity- and arts-based programmes employed structured techniques to encourage higher-order thinking and foster positive attitudes toward ability, while interest-based curriculum adaptations improved on-task behaviour and academic engagement when tasks were sufficiently challenging and meaningful.

The synthesis of 2e-specific interventions suggests that these interventions are most effective for 2e-ASL when they combine explicit teaching, visual scaffolding, repeated guided practice, and delivery in meaningful contexts, with additional benefits when learners’ strengths and interests are actively incorporated.

3.7. Quality Appraisal

The evidence base demonstrates moderate to good methodological quality, supported by consistent internal validity but limited external generalisability. High-fidelity RCTs (e.g., Kenworthy et al., 2014; Sofronoff et al., 2005; Dekker et al., 2019; Pugliese et al., 2024) achieve good quality, whereas small quasi-experimental or single-case studies remain of moderate quality due to scope and scale constraints.

4. Discussion

The systematic review aimed to synthesise evidence on interventions for 2e-ASL to understand how these learners are supported. The current review of 51 studies (2000–2025) highlights an emerging but fragmented evidence base on interventions for 2e-AS children and adolescents. While findings demonstrate effects across developmental, academic and behavioural domains, the literature remains constrained by inconsistent definitions of 2e-ASL, limited 2e-ASL representation, lack of interventions specifically designed for the population and a predominant focus on clinical settings.

4.1. Empirical Interventions (RQ: 1)

Most of the interventions included in this review did not explicitly target 2e-ASL. Instead, these learners were typically grouped within samples described as “high-functioning” or “aspergers” labels that were used inconsistently and often without clear operational definitions. Although intellectual functioning is widely recognised as a major source of heterogeneity within autism (Charman et al., 2011; Wolff et al., 2022), it was rarely treated as such in intervention design or analysis. Across studies, terms such as “high-functioning” were frequently applied to broad groups defined by minimal criteria (e.g., IQ > 70) or by very wide, sometimes unspecified IQ ranges. As a result, samples often included learners with markedly different cognitive profiles, ranging from below-average ability to highly gifted levels. Reported IQ ranges were often exceptionally wide, with differences of up to 50 points in verbal IQ, nearly 60 points in performance IQ, and over 50 points in full-scale IQ. In practice, this meant that 2e-ASL were rarely identified or analysed as a distinct group, and giftedness was largely incidental rather than intentionally supported within interventions.

The majority of interventions focused on social and emotional skills, particularly social communication, peer interaction, and emotional understanding. This emphasis aligns with the core diagnostic features of autism (American Psychiatric Association, 2022) and reflects longstanding priorities in autism intervention research. Interventions were most often delivered through structured, skills-based programmes or naturalistic, interaction-focused approaches. Delivery settings were predominantly clinical or university-based, with far fewer interventions implemented in schools, homes, or community contexts. Programme duration and intensity varied widely, and many studies provided limited detail regarding dosage or implementation fidelity.

A clear distinction emerged between interventions developed for the broader autistic population and those designed specifically for 2e-ASL. A smaller subset of studies stood out because they were designed specifically for 2e-ASL. These interventions differed qualitatively from the broader literature. Rather than focusing primarily on remediating social or emotional difficulties, they foregrounded learners’ intellectual strengths, creativity, interests, and advanced cognitive processes. Learners were encouraged to engage deeply with challenging, meaningful tasks, generate ideas, and contribute actively, rather than being positioned as passive recipients of corrective training.

This strengths-based orientation closely reflects recent theoretical and educational models proposed for 2e-ASL, which emphasise interest alignment, differentiated instruction, creativity, and talent development alongside structured support for areas of difficulty (Lim, 2021; Austermann et al., 2025; Reis & Renzulli, 2025). Importantly, the findings also extend previous reviews suggesting a lack of empirically evaluated interventions specifically designed for 2e-ASL (Gelbar et al., 2021), indicating that a small but emerging evidence base is beginning to develop.

4.2. Effectiveness of Interventions (RQ: 2)

Overall, interventions showed positive effects, but it is difficult to draw firm conclusions about effectiveness for 2e-ASL specifically. Because most studies grouped learners with a wide range of cognitive abilities together, outcomes were typically averaged across heterogeneous samples. This approach limits meaningful differences in how learners respond to intervention and makes it difficult to determine who benefits, under what conditions, and why. This is a particularly significant limitation given the distinctive asynchronous profile of 2e-ASL, who simultaneously present with advanced cognitive strengths, such as verbal reasoning and pattern recognition alongside challenges in social communication, executive functioning, and adaptive behaviour (Assouline et al., 2012; Cain et al., 2019). As noted earlier, higher IQ does not function as a straightforward protective factor; rather, it can intensify the mismatch between cognitive potential and real-world performance (Charman et al., 2011; Cain et al., 2019; Michaelson et al., 2021), making undifferentiated grouping especially problematic for this population

Only a small number of studies in this review examined cognitive factors as moderators of intervention outcomes, but these studies provide important insights. Higher IQ was associated with stronger gains in structured, explicit programmes targeting socio-emotional understanding and cognitive skills (Solomon et al., 2004). Receptive language ability predicted responsiveness to grammar instruction (Bangert et al., 2019) and to perceptual learning tasks (Feldman et al., 2023). In contrast, lower verbal IQ was linked to greater improvements in joint engagement and social reciprocity within more naturalistic, interaction-based interventions (MacDonald-Pregent et al., 2023). Together, these findings suggest that IQ and language ability influence intervention responsiveness, but not in a uniform way. Their impact depends on the nature of the intervention and the context in which it is delivered. Without more consistent moderator analyses that account for within-group variability, the intervention needs of 2e-ASL will continue to be masked within averaged group outcomes.

Interventions designed specifically for 2e-ASL demonstrated benefits beyond cognitive outcomes. Strength-based and interest-aligned approaches were associated with improvements in engagement, motivation, and behavioural regulation. These findings are particularly noteworthy for this population for several interconnected reasons. First, 2e-ASL faces elevated risks of underachievement, disengagement, and reduced self-efficacy, in part because deficit-focused models emphasise areas of difficulty while overlooking advanced abilities (Lim, 2021; Austermann et al., 2025). Strength-based approaches directly counter this by supporting intervention in areas of genuine competence, which may reduce experiences of failure and support motivation. Second, interest-alignment as an intervention feature maps onto well-documented patterns in autism, where focused interests serve as sources of motivation, emotional regulation, and identity, and can function as productive entry points for broader skill development (Reis & Renzulli, 2025). For 2e-ASL specifically, aligning intervention with areas of strength and interest addresses both dimensions of their dual profile simultaneously, supporting the gifted learner’s need for appropriate challenge while also supporting autistic patterns of deep engagement. Third, given the psychosocial risks including higher rates of anxiety, depression, and social isolation among 2e-ASL compared to gifted peers without autism, interventions that improve engagement and behavioural regulation carry particular clinical significance for this group (Casten et al., 2023; Foley Nicpon et al., 2010).

These findings suggest that appropriately challenging, meaningful, and strength-based interventions may support not only targeted skills but also broader dimensions of learning, engagement, and adaptive functioning in 2e-ASL.

4.3. Recommendations (RQ: 3)

Taken together, the findings point to several clear directions for future research and practice. First, there is a pressing need for more consistent definitions and clearer identification of 2e-ASL in intervention research. Without this clarity, it remains difficult to compare findings across studies or to develop a cumulative evidence base.

Second, cognitive diversity must be treated as a meaningful factor in intervention design, rather than as a background characteristic. IQ and language ability shape how learners engage with interventions and how they respond and should be examined systematically as moderators of outcomes.

Third, greater attention is needed to real-world settings, particularly schools and community contexts. While controlled environments can support initial learning, autistic learners might require explicit support to generalise skills beyond the intervention context.

Finally, this review supports a shift towards strength-based, scalable interventions for 2e-ASL learners. Approaches that combine clear structure with appropriate intellectual challenge, meaningful tasks, and learner interests appear better suited to supporting long-term engagement, learning, and wellbeing than those focused solely on remediation.

Overall, the effectiveness of interventions for 2e-ASL depends not only on which skills are targeted, but on whether learners’ intellectual strengths and cognitive diversity are recognised and valued. Designing interventions that acknowledge and work with this complexity is essential for achieving meaningful outcomes for 2e-ASL.

5. Conclusions

This systematic review synthesised intervention research involving 2e-ASL children and adolescents and highlights a developing but fragmented evidence base.

Regarding the first research question, findings suggest that interventions including 2e-ASL were highly heterogeneous in their design, outcomes, settings, and duration, with learners rarely identified as a distinct population and instead commonly embedded within broader autism samples. Across studies, interventions most frequently targeted social and emotional domains and predominantly relied on structured, explicit, and scaffolded approaches.

In response to the second research question on intervention effectiveness, findings indicate generally positive outcomes across developmental, behavioural, and academic domains. However, firm conclusions regarding effectiveness specifically for 2e-ASL remain limited due to inconsistent identification practices, heterogeneous samples, and limited examination of moderators.

Importantly, the small number of interventions designed specifically for 2e-ASL adopted strength-based approaches that focused on intellectual ability, creativity, and learner interests, and were associated with improvements in engagement, motivation, and behavioural regulation. These findings suggest that recognising and incorporating cognitive strengths may enhance the relevance and effectiveness of interventions for this population.

In response to the third research question, which sought to draw recommendations for designing scalable evidence-based supports, the review identifies several priorities for practice and future research. This highlights the need for clearer identification of 2e-ASL, more consistent reporting of cognitive characteristics, and the development of interventions that integrate both support for autistic differences and opportunities to nurture gifted potential. Strength-based, interest-aligned, and appropriately differentiated approaches show particular promise and warrant further empirical investigation as a foundation for scalable, evidence-informed support for this population.