A Review Of The Sex Differences Of Restricted And Repetitive Behaviors And Interests In Autism

Restricted and repetitive behaviors and interests (RRBIs) are core features of autism spectrum disorder. They include stereotyped movements, rigid routines, fixated interests, and differences in sensory responses.

Recent research suggests that RRBIs may manifest differently between autistic males and females. For instance, males might show more obvious repetitive behaviors or interests in mechanical objects, while females may have more socially-oriented or camouflaged RRBIs.

Studying these sex differences is crucial for several reasons: it can improve diagnostic accuracy, especially for females who are often underdiagnosed; it can lead to more tailored interventions and support strategies; and it deepens our understanding of autism’s diverse presentations.

This knowledge is vital for creating more inclusive diagnostic criteria and ensuring that all autistic individuals receive appropriate recognition and support.

Edwards, H., Wright, S., Sargeant, C., Cortese, S., & Wood‐Downie, H. (2024). Research review: A systematic review and meta‐analysis of sex differences in narrow constructs of restricted and repetitive behaviours and interests in autistic children, adolescents, and adults. Journal of Child Psychology and Psychiatry, 65(1), 4-17. https://doi.org/10.1111/jcpp.13855

Key Points

• The systematic review and meta-analysis found significant sex differences in stereotyped behaviors and restricted interests, with autistic males displaying higher rates compared to autistic females.
• No significant sex differences were found for sensory experiences or insistence on sameness.
• Qualitative differences were observed in how RRBIs manifest between sexes, particularly in the types of restricted interests.
• The research highlights the importance of fine-grained analysis of narrow RRBI constructs rather than only examining broad constructs.
• Sex differences in RRBIs could contribute to the under-recognition and late diagnosis of autism in females.
• The findings have implications for improving autism assessment and diagnostic tools to better capture how autism may present differently in females.
• Limitations included heterogeneity between studies, underrepresentation of individuals with lower IQ, and wide age ranges that prevented analysis of age as a moderating factor.
• The study underscores the need for more nuanced, sex-informed approaches to autism research, diagnosis, and support.

Rationale

The rationale for this study was to address gaps in the existing literature on sex differences in restricted and repetitive behaviors and interests (RRBIs) in autism.

Previous research and reviews have typically focused on broad constructs of autism symptomatology, including RRBIs as an overall domain (van Wijngaarden-Cremers et al., 2014).

However, there was a lack of systematic examination of sex differences at the level of narrow RRBI constructs based on the DSM-5 diagnostic criteria.

The authors note that consistently exploring sex differences only at the broad construct level may miss potential subtle differences in specific RRBI subdomains.

This is particularly important given emerging evidence of a potential “Female Autism Phenotype” that may manifest differently from traditional (male-centric) conceptualizations of autism (Hull et al., 2020).

Understanding nuanced sex differences in RRBIs could help explain the underdiagnosis of autism in females.

Additionally, while previous reviews had found that autistic females generally display fewer RRBIs than males at the broad construct level (Lai et al., 2015; van Wijngaarden-Cremers et al., 2014), studies examining narrow RRBI constructs produced more mixed results.

For example, some research found no sex differences in specific RRBI behaviors like stereotyped language or resistance to change (Mclennan et al., 1993), while others reported higher rates of sensory issues in autistic females (Lai et al., 2011).

The rationale was further strengthened by the inclusion of sensory symptoms as an RRBI subdomain in the DSM-5, which had not been systematically synthesized in terms of sex differences.

Overall, a fine-grained analysis of sex differences across narrow RRBI constructs was needed to clarify inconsistencies in the literature and potentially inform more sensitive diagnostic approaches.

Method

The study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

The review protocol was prospectively registered on PROSPERO (registration number: CRD42021254221).

Search strategy and terms

The authors searched six electronic databases on May 25, 2021: APA PsychInfo, Medline, ERIC, Science Direct, PsycArticles, and CINAHL Plus with Full Text.

Search terms were based on DSM-5 autism symptom subdomains of restricted, repetitive patterns of behavior, interests, or activities.

Population terms included “autism spectrum disorder” and “autism spectrum condition”. Comparator terms were “sex” and “gender”.

Outcome terms included “repetitive behavi?r*”, “restricted interest”, “insistence on sameness”, “sensory”, and “circumscribed interest*”.

The full search strategy is provided in the supplementary materials. Reference lists of included studies were also hand-searched.

The search was re-run on October 21, 2022 to identify any additional relevant articles.

Inclusion and exclusion criteria:

Inclusion criteria:

• Cross-sectional, peer-reviewed studies
• Including autistic males and females of any age
• Including an outcome measure of RRBI subdomains based on DSM-5 criteria
• Studies with participants with high autistic traits or self-identifying as autistic

Exclusion criteria:

• Studies not including a measure reporting a DSM-5 RRBI subdomain
• Studies with very small samples of autistic females (6 or fewer)
• Non-English language studies (due to financial constraints)

The authors provide a detailed rationale for the inclusion/exclusion criteria in Table 1 of the paper.

Statistical measures

Four random-effects meta-analyses were performed using Comprehensive Meta-Analysis software, one for each narrow RRBI construct.

Standardized mean differences (SMD) were calculated between autistic males and females. Where studies used multiple measures, the measure most closely reflecting the narrow construct was selected.

Publication bias was assessed through visual inspection of funnel plots and Egger’s test. Heterogeneity was evaluated using chi-squared tests and the I2 statistic.

The authors planned moderator analyses for age groups and IQ levels, but were unable to conduct these due to limitations in the included studies.

A post-hoc sensitivity analysis of high-quality studies and a meta-regression with publication year as a covariate were conducted.

Results

The systematic review included 46 studies in the narrative synthesis, with 25 of these included in the meta-analyses.

Key findings for each RRBI subdomain were:

Stereotyped or repetitive motor movements, use of objects, or speech:

• Meta-analysis found significantly higher rates in autistic males compared to females (SMD = 0.21, 95% CI [0.09, 0.33], p = .001)
• Narrative synthesis identified males as displaying more preoccupations with parts of objects, stereotyped object use, and motor mannerisms

Highly restricted, fixated interests that are abnormal in intensity or focus:

• Meta-analysis found significantly higher rates in autistic males (SMD = 0.18, 95% CI [0.07, 0.29], p = .001)
• Narrative synthesis highlighted qualitative differences:
  • Males showed greater interest in object-related topics (e.g. technology, mechanics, transport)
  • Females showed greater interest in topics with social qualities or related to living things (e.g. psychology, animals, arts and crafts)

Insistence on sameness, inflexible adherence to routines, or ritualized patterns of verbal/nonverbal behavior:

• Meta-analysis found no significant sex difference (SMD = 0.01, 95% CI [-0.03, 0.05], p = .68)
• Some studies in narrative synthesis reported higher rates of specific behaviors (e.g. distress at small changes) in females

Hyper- or hypo-reactivity to sensory input or unusual interests in sensory aspects of the environment:

• Meta-analysis found no significant overall sex difference (SMD = -0.09, 95% CI [-0.27, 0.09], p = .31)
• Narrative synthesis identified some differences in specific sensory domains (e.g. higher taste sensitivity in females, more visual interests in males)

Heterogeneity and publication bias:

• Significant heterogeneity was found for stereotyped behaviors, restricted interests, and sensory experiences analyses
• Publication bias was detected for the restricted interests analysis

The authors were unable to conduct planned moderator analyses for age and IQ due to limitations in the included studies.

A sensitivity analysis of high-quality studies produced similar results to the main analyses.

Insight

This systematic review and meta-analysis provides important insights into sex differences in restricted and repetitive behaviors and interests (RRBIs) in autism.

By examining narrow constructs of RRBIs rather than broad domains, the study reveals a more nuanced picture of how these core autism features may manifest differently in males and females.

The finding that autistic males display higher rates of stereotyped behaviors and restricted interests compared to females is particularly informative.

This aligns with and extends previous research suggesting fewer overall RRBIs in autistic females (van Wijngaarden-Cremers et al., 2014).

However, by focusing on specific RRBI subdomains, this study shows that sex differences are not uniform across all repetitive and restricted behaviors.

The qualitative differences in restricted interests between males and females provide crucial insight into potential biases in autism identification and diagnosis.

The tendency for autistic females to have more socially-oriented or developmentally normative interests (e.g. animals, relationships) compared to the object-focused interests often associated with male autism presentations (e.g. mechanics, transportation) could contribute to females being overlooked for autism assessment.

This expands on previous work suggesting that the female autism phenotype may be less readily recognized by clinicians and educators (Bargiela et al., 2016).

The lack of significant sex differences in sensory experiences and insistence on sameness is also informative, suggesting these features may be equally prevalent in autistic males and females.

This highlights the importance of thoroughly assessing these domains in all individuals, regardless of sex.

These findings extend previous research by demonstrating that sex differences in autism are not monolithic across symptom domains.

They underscore the need for more sensitive, sex-informed approaches to autism assessment and diagnosis.

Future research should further explore how these RRBI differences manifest across development and in individuals without formal autism diagnoses.

Additionally, investigating the underlying neurobiological and sociocultural factors contributing to these sex differences would be valuable.

Strengths

The study had several methodological strengths:

1. Comprehensive search strategy: The authors searched multiple databases and hand-searched reference lists, increasing the likelihood of capturing all relevant studies.
2. Adherence to systematic review guidelines: The study followed PRISMA guidelines and pre-registered the protocol, enhancing transparency and reproducibility.
3. Focus on narrow constructs: By examining specific RRBI subdomains rather than broad constructs, the study provided a more nuanced understanding of sex differences in autism.
4. Combination of narrative synthesis and meta-analysis: This approach allowed for both quantitative comparison and qualitative exploration of sex differences.
5. Quality assessment: The authors used the Newcastle-Ottawa Quality Assessment Scale to evaluate study quality and conducted sensitivity analyses based on study quality.
6. Consideration of publication bias: The use of funnel plots and Egger’s test to assess publication bias strengthens the reliability of the findings.
7. Inclusion of diverse study designs: The review incorporated both studies of clinically diagnosed individuals and those with high autistic traits, broadening the applicability of the findings.

Limitations

Several limitations should be considered when interpreting the results:

1. Language restriction: Only English-language studies were included due to financial constraints, potentially missing relevant research in other languages.
2. Sample representativeness: A significant proportion (36%) of included studies were rated as “unsatisfactory” in quality, often due to unrepresentative samples or lack of validated autism diagnosis confirmation. This may limit generalizability, particularly to individuals with co-occurring intellectual disabilities.
3. Focus on diagnosed individuals: Most included studies featured participants with existing clinical diagnoses, limiting generalizability to potentially autistic individuals without formal diagnoses.
4. Heterogeneity: Significant heterogeneity was found in the meta-analyses for restricted interests, stereotyped behaviors, and sensory experiences. This suggests substantial variability between studies that could not be fully explained.
5. Inability to analyze moderating factors: Due to limitations in the included studies, the authors were unable to conduct planned moderator analyses for age and IQ, which may have accounted for some of the observed heterogeneity.
6. Lack of longitudinal data: The cross-sectional nature of included studies prevents examination of how sex differences in RRBIs may change across development.
7. Binary sex focus: The review did not include studies examining autism in gender-diverse individuals, limiting understanding of RRBI presentations across the full gender spectrum.

These limitations highlight the need for future research with more representative samples, longitudinal designs, and consideration of potential moderating factors like age and cognitive ability.

Clinical Implications

The findings of this study have significant implications for autism research, clinical practice, and diagnostic processes:

1. Diagnostic sensitivity: The identified sex differences in stereotyped behaviors and restricted interests suggest that current diagnostic tools and criteria may be biased towards male autism presentations. This could contribute to the underdiagnosis of autism in females, particularly those without additional cognitive or behavioral difficulties. Clinicians and researchers should consider developing or adapting assessment tools to better capture female-typical RRBI presentations.
2. Clinical awareness: Professionals involved in autism assessment and diagnosis need to be aware of the potential for qualitative differences in how RRBIs manifest between males and females. For example, recognizing that females may have more socially-oriented restricted interests that might be mistaken for typical behavior.
3. Fine-grained assessment: The study highlights the importance of examining narrow constructs of RRBIs rather than relying solely on broad domain scores. Clinicians should consider assessing each RRBI subdomain separately to avoid overlooking subtle sex differences.
4. Sensory and sameness behaviors: The lack of significant sex differences in sensory experiences and insistence on sameness emphasizes the importance of thoroughly assessing these domains in all individuals, regardless of sex.
5. Research methodology: Future autism studies should strive for more representative samples, including individuals across the IQ spectrum and with diverse gender identities. Longitudinal research is needed to understand how sex differences in RRBIs may change across development.
6. Educational implications: The findings suggest that educational staff may benefit from training on recognizing diverse autism presentations, particularly in relation to restricted interests that may appear more socially typical in females.
7. Theoretical understanding: The observed sex differences contribute to our understanding of the potential “Female Autism Phenotype” and highlight the need for more nuanced, sex-informed models of autism.
8. Lifespan considerations: The potential for age-dependent manifestations of RRBIs (e.g., how restricted interests may be perceived differently at various life stages) underscores the need for clinicians to consider developmental context when assessing autism.

These implications collectively point to the need for a more individualized, sex-informed approach to autism assessment, diagnosis, and support that considers the diverse ways in which core features like RRBIs may manifest across different populations.

References

Primary reference

Edwards, H., Wright, S., Sargeant, C., Cortese, S., & Wood‐Downie, H. (2024). Research review: A systematic review and meta‐analysis of sex differences in narrow constructs of restricted and repetitive behaviours and interests in autistic children, adolescents, and adults. Journal of Child Psychology and Psychiatry, 65(1), 4-17. https://doi.org/10.1111/jcpp.13855

Other references

Bargiela, S., Steward, R., & Mandy, W. (2016). The experiences of late-diagnosed women with autism spectrum conditions: An investigation of the female autism phenotype. Journal of Autism and Developmental Disorders, 46(10), 3281-3294.

Hull, L., Petrides, K. V., & Mandy, W. (2020). The female autism phenotype and camouflaging: A narrative review. Review Journal of Autism and Developmental Disorders, 7(4), 306-317.

Lai, M. C., Lombardo, M. V., Pasco, G., Ruigrok, A. N., Wheelwright, S. J., Sadek, S. A., … & Baron-Cohen, S. (2011). A behavioral comparison of male and female adults with high functioning autism spectrum conditions. PloS one, 6(6), e20835.

Lai, M. C., Lombardo, M. V., Auyeung, B., Chakrabarti, B., & Baron-Cohen, S. (2015). Sex/gender differences and autism: setting the scene for future research. Journal of the American Academy of Child & Adolescent Psychiatry, 54(1), 11-24.

Mclennan, J. D., Lord, C., & Schopler, E. (1993). Sex differences in higher functioning people with autism. Journal of autism and developmental disorders, 23(2), 217-227.

van Wijngaarden-Cremers, P. J., van Eeten, E., Groen, W. B., Van Deurzen, P. A., Oosterling, I. J., & Van der Gaag, R. J. (2014). Gender and age differences in the core triad of impairments in autism spectrum disorders: a systematic review and meta-analysis. Journal of autism and developmental disorders, 44(3), 627-635.

Keep Learning

Socratic questions for a college class to discuss this paper

1. How might societal gender norms and expectations influence the manifestation and recognition of restricted interests in autistic individuals?
2. What ethical considerations should researchers and clinicians keep in mind when studying and discussing sex differences in autism?
3. How could the findings of this study inform the development of more inclusive autism screening tools? What challenges might arise in creating such tools?
4. In what ways might the observed sex differences in RRBIs impact the daily lives and experiences of autistic individuals? How could this inform support strategies?
5. How do the limitations of this study, particularly the focus on diagnosed individuals, potentially skew our understanding of autism in the broader population?
6. What role might camouflaging or masking behaviors play in the observed sex differences in stereotyped behaviors? How could future research address this?
7. How might the findings of this study challenge or support current theories about the neurobiology of autism?
8. In what ways could the qualitative differences in restricted interests between males and females reflect broader societal influences rather than inherent sex differences?