Camouflaging Autistic Traits Questionnaire (CAT-Q)

Abstract

The Camouflaging Autistic Traits Questionnaire (CAT-Q) is a widely recognized, self-report psychometric instrument designed to quantify the behavioral and cognitive strategies employed by individuals to camouflage, mask, or compensate for autistic traits during social interactions. Developed by Laura Hull, William Mandy, Meng-Chuan Lai, Simon Baron-Cohen, and colleagues (2019), the CAT-Q addresses a critical gap in clinical and developmental psychometrics by operationalizing the phenomenon of social camouflaging. The instrument consists of 25 items evaluated along a 7-point Likert scale ranging from 1 (“Strongly Disagree”) to 7 (“Strongly Agree”).

Through rigorous exploratory and confirmatory factor analyses, the CAT-Q yields a robust three-factor structure representing three distinct yet correlated dimensions: Compensation (learning and practicing strategies to overcome social-communication difficulties), Masking (actively concealing autistic characteristics to project a non-autistic persona), and Assimilation (attempting to blend into social environments and avoiding the exposure of social vulnerabilities). Psychometric evaluations demonstrate excellent internal consistency across both autistic and neurotypical populations, with total scale Cronbach’s alpha values typically exceeding α = .90, accompanied by stable test-retest reliability over three-month intervals. The CAT-Q exhibits strong convergent validity with measures of autistic traits such as the Autism-Spectrum Quotient (AQ), as well as significant correlations with internalizing mental health symptoms, including generalized anxiety and major depression. By providing an empirical framework for measuring compensatory social behaviors, the CAT-Q plays a vital role in identifying missed or late-diagnosed autism, clarifying sex and gender differences in the phenotypic presentation of autism, and elucidating the psychological costs of prolonged behavioral masking.

Keywords

autistic camouflaging, masking, compensation, assimilation, autism spectrum disorder, CAT-Q, late diagnosis, female autism phenotype, psychometrics, mental health

Authors

The Camouflaging Autistic Traits Questionnaire was developed and psychometrically validated by a collaborative team of developmental psychologists, neuroscientists, and psychiatrists specializing in neurodevelopmental conditions:

• Laura Hull, Ph.D. — Department of Clinical, Educational and Health Psychology, University College London (UCL), London, United Kingdom.
• William Mandy, Ph.D., DClinPsy — Professor of Neurodevelopmental Conditions, Department of Clinical, Educational and Health Psychology, University College London (UCL), London, United Kingdom.
• Meng-Chuan Lai, M.D., Ph.D. — Centre for Addiction and Mental Health (CAMH), The Hospital for Sick Children, Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada; and Autism Research Centre, Department of Psychiatry, University of Cambridge, Cambridge, United Kingdom.
• Simon Baron-Cohen, Ph.D., FBA, FMedSci — Professor of Developmental Psychopathology, Director of the Autism Research Centre (ARC), Department of Psychiatry, University of Cambridge, Cambridge, United Kingdom.
• Carrie Allison, Ph.D. — Director of Strategy, Autism Research Centre (ARC), Department of Psychiatry, University of Cambridge, Cambridge, United Kingdom.
• Paula Smith, Ph.D. — Autism Research Centre (ARC), Department of Psychiatry, University of Cambridge, Cambridge, United Kingdom.
• K. V. Petrides, Ph.D. — Professor of Psychology and Psychometrics, UCL Psychometric Laboratory, University College London (UCL), London, United Kingdom.

Purpose

The primary purpose of the Camouflaging Autistic Traits Questionnaire (CAT-Q) is to systematically identify, assess, and quantify the behavioral modifications, explicit cognitive strategies, and psychological mechanisms used by autistic individuals—as well as individuals within the general population who possess elevated autistic traits—to hide their natural communicative, behavioral, or emotional differences in social contexts. Clinically, the instrument was engineered to explain why a substantial subpopulation of autistic individuals, particularly cisgender females, non-binary individuals, and adults with preserved intellectual and structural language functioning, experience delayed identification or frequent misdiagnosis.

Prior to the formal operationalization of camouflaging, clinicians and diagnostic assessment tools primarily relied upon outward, observable behavioral deficits in reciprocal social interaction and nonverbal communication. Consequently, individuals who developed sophisticated cognitive workarounds—such as pre-scripting conversational topics, memorizing appropriate facial expressions from popular media, or maintaining deliberate eye contact despite physical discomfort—routinely fell below conventional diagnostic thresholds on instruments like the Autism Diagnostic Observation Schedule (ADOS). The CAT-Q illuminates this critical discrepancy between an individual’s underlying autistic cognitive architecture and their superficial, highly effortful social presentation.

Beyond diagnostic considerations, the CAT-Q serves as an essential research and clinical evaluation tool for mental health correlates. Extensive qualitative and quantitative evidence demonstrates that sustained, long-term camouflaging is cognitively exhausting and emotionally taxing. High CAT-Q scores are strongly predictive of autistic burnout, chronic anxiety, clinical depression, identity confusion, and elevated suicidality. Clinicians utilize the scale to evaluate the internal psychological strain experienced by patients, helping guide affirming therapeutic interventions, reduce pathological masking, and foster psychological wellbeing.

Psychological Construct

The construct of social camouflaging in autism refers to the discrepancy between an individual’s internal neurodivergent cognitive-emotional profile and their external behavioral presentation in social spaces. Camouflaging encompasses both deliberate and automatic behaviors aimed at managing the impressions formed by others, navigating neurotypical social conventions, and protecting oneself from social stigma, alienation, or discrimination. The CAT-Q deconstructs this overarching psychometric construct into three distinct, interdependent latent factors:

1. Compensation

The Compensation subscale assesses explicit, active cognitive strategies engineered to overcome, bypass, or counterbalance core social and communication differences. Autistic individuals frequently experience challenges with spontaneous non-verbal communication, implicit theory of mind, and natural conversational pacing. To compensate, an individual employs conscious intellectual effort, learning social etiquette algorithmically rather than intuitively.

Behavioral manifestations captured by this dimension include:

• Explicitly scripting social conversations prior to engaging with peers or colleagues (e.g., preparing conversational trees or practicing icebreaker topics).
• Studying the communicative patterns, gestures, and body language of neurotypical peers, actors, or public figures to replicate them during social encounters.
• Echoing phrases, vocal cadences, or behavioral mannerisms observed in films, literature, or television shows to appear conversationally fluent and socially conventional.

2. Masking

The Masking subscale measures the active suppression, concealment, or inhibition of natural autistic behaviors, emotional states, and communicative styles to project a non-autistic persona. Unlike compensation, which involves learning and adopting additive social behaviors, masking primarily functions through behavioral suppression and performance monitoring.

Key indicators within this dimension include:

• Suppressing motor stereotypies (stimming), such as hand-flapping, rocking, or pacing, when in the presence of others.
• Forcing eye contact or actively tracking where one looks during interactions, despite finding sustained direct gaze physically overwhelming or sensory-aversive.
• Constantly monitoring one’s facial expressions and body posture to simulate warmth, engagement, or relaxation, even when experiencing internal distress or sensory overload.

3. Assimilation

The Assimilation subscale gauges the respondent’s deep psychological experience of attempting to blend into social environments, achieve social acceptance, and avoid exposing social vulnerabilities. This dimension reflects feelings of alienation, self-alienation, social detachment, and the sense that social interaction requires donning an artificial persona or playing an exhausting dramatic role.

Characteristic experiences within this dimension include:

• Feeling that social encounters require “performing” a part or wearing a metaphorical mask rather than engaging authentically.
• Relying on social anchors—such as close friends, partners, or structured activities—to buffer or navigate complex social environments.
• Forcing oneself to engage in social contexts out of external obligation while feeling fundamentally alienated, misunderstood, or unable to be oneself.

Theoretical Framework

The psychometric architecture of the CAT-Q is grounded in several convergent sociological, developmental, and neurocognitive theories. Most prominently, the construct draws upon Erving Goffman’s foundational sociological framework of dramaturgical analysis and impression management (1959). Goffman postulated that social actors constantly engage in front-stage performances designed to conform to normative social scripts, carefully managing audience impressions while reserving unmasked authenticity for back-stage environments. In the context of autism, this framework highlights how individuals who lack instinctive alignment with dominant neurotypical norms perform calculated, high-effort “front-stage” roles to mitigate social penalties and secure inclusion.

From a neurocognitive perspective, the CAT-Q aligns with models of executive functioning and compensatory cognitive scaffolding. Neurodevelopmental research led by Livingston and Happé (2017) posits that compensatory mechanisms allow individuals with atypical social cognition to achieve typical behavioral output. Rather than executing social interactions through automated socio-emotional circuitry (e.g., spontaneous mentalizing), compensatory camouflaging relies heavily upon top-down executive control, verbal mediation, explicit memory, and working memory. Consequently, this cognitive override is resource-intensive and vulnerable to breakdown under conditions of stress, fatigue, or cognitive overload.

The framework also integrates modern neurodiversity perspectives and the Double Empathy Problem formulated by Damian Milton (2012). This paradigm asserts that social interaction breakdowns between autistic and neurotypical individuals do not represent an intrinsic deficit within the autistic person, but rather a bidirectional mismatch in communication styles, emotional expressivity, and mutual understanding. Within a predominantly neurotypical society, the burden of bridge-building falls disproportionately on the autistic individual, who must acquire compensatory strategies and suppress natural traits to survive socially and professionally. The CAT-Q operationalizes the cumulative psychological labor generated by this asymmetric relational dynamic.

Validity

The psychometric validity of the CAT-Q has been evaluated across multiple diverse, cross-national clinical and non-clinical cohorts, demonstrating exceptional psychometric properties:

Construct and Factorial Validity

During its initial validation by Hull et al. (2019), the scale was administered to a substantial international sample comprising autistic adults (n = 354) and non-autistic adults (n = 478). Exploratory factor analysis (EFA) on an initial pool of candidate items yielded an unambiguous three-factor solution accounting for substantial variance. Subsequent confirmatory factor analysis (CFA) verified that the hierarchical three-factor model—comprising Compensation, Masking, and Assimilation loading onto a global camouflaging factor—exhibited superior model fit compared to unidimensional or alternative two-factor structures.

Convergent and Divergent Validity

The CAT-Q demonstrates significant convergent validity with established measures of the broader autism phenotype. Hull et al. (2019) observed moderate-to-strong positive correlations between CAT-Q total scores and the 10-item Autism-Spectrum Quotient (AQ-10; r = .38 to .49, p < .001). Furthermore, camouflaging scores are positively associated with social anxiety (measured via the Liebowitz Social Anxiety Scale) and generalized psychological distress (measured via the GHQ-12 and DASS-21).

Divergent validity has been confirmed through weak or non-significant correlations with unrelated constructs, such as spatial reasoning abilities and non-verbal IQ. Additionally, research across independent international validation studies (e.g., in Italian, Japanese, French, and Swedish adaptations) demonstrates that autistic women and non-binary individuals consistently score higher on the CAT-Q than autistic men, mirroring well-documented clinical patterns of the female autism phenotype.

Criterion and Predictive Validity

Criterion-related validity is evidenced by the scale’s sensitivity in differentiating autistic individuals who received late diagnoses (in adulthood) from those diagnosed early in childhood. Late-diagnosed autistic adults demonstrate significantly elevated total CAT-Q scores, particularly on the Compensation and Assimilation subscales, underscoring the instrument’s power in capturing the masking behaviors that historically obscured their neurodevelopmental profile from clinicians and educators.

Reliability

The CAT-Q exhibits high internal consistency and temporal stability across clinical and non-clinical populations:

• Internal Consistency: In the original validation cohort (Hull et al., 2019), the total CAT-Q demonstrated excellent internal reliability, with a Cronbach’s alpha of α = .94 in the autistic group and α = .93 in the non-autistic comparison group. The subscales demonstrated robust internal consistency across groups:
  • Compensation: α = .91 (autistic), α = .89 (non-autistic)
  • Masking: α = .85 (autistic), α = .84 (non-autistic)
  • Assimilation: α = .92 (autistic), α = .91 (non-autistic)

  Subsequent studies using McDonald’s omega have reaffirmed these values, with composite reliability coefficients routinely exceeding ω = .90 for the total scale.

• Test-Retest Reliability: Temporal stability was assessed across a 3-month follow-up period in a subsample of participants. The intraclass correlation coefficient (ICC) for the overall CAT-Q total score was r = .86 (p < .001), indicating strong longitudinal reliability and stability of self-reported camouflaging over time. Subscale test-retest coefficients remained high: Compensation (r = .84), Masking (r = .78), and Assimilation (r = .85).

Factor Analysis

The psychometric derivation of the CAT-Q involved an iterative, multi-stage structural equation modeling strategy conducted by Hull and colleagues (2019):

Exploratory Factor Analysis (EFA)

An initial pool of 48 candidate items derived from qualitative studies was administered to a calibration sample. Principal axis factoring with Promax (oblique) rotation was conducted, given the theoretical expectation that dimensions of camouflaging are correlated. Examination of the scree plot, Kaiser-Guttman retention criterion (eigenvalues > 1.0), and parallel analysis converged on a distinct three-factor architecture. Items exhibiting poor primary loadings (< .40) or severe cross-loadings (> .30 on secondary factors) were systematically eliminated, reducing the scale to 25 optimal items.

Confirmatory Factor Analysis (CFA)

The refined 25-item structure was subsequently cross-validated in an independent validation sample using maximum likelihood estimation. The three-factor second-order model (where the 25 items load onto three first-order factors—Compensation, Masking, and Assimilation—which in turn load onto a single higher-order camouflaging factor) demonstrated excellent fit indices:

• Comparative Fit Index (CFI): .947 (acceptable fit threshold ≥ .90; excellent ≥ .95)
• Tucker-Lewis Index (TLI): .941
• Root Mean Square Error of Approximation (RMSEA): .056 (90% CI [.052, .060], indicating close model fit < .06)
• Standardized Root Mean Square Residual (SRMR): .058 (indicating acceptable fit < .08)

Standardized item factor loadings across the three dimensions were robust, ranging from .51 to .87 on Compensation, .48 to .82 on Masking, and .57 to .89 on Assimilation. Measurement invariance testing across sex (male vs. female) and diagnostic status (autistic vs. non-autistic) confirmed configural, metric, and scalar invariance, establishing that CAT-Q scores can be meaningfully compared across demographic and clinical groups.

Instrument / Measurement Tool

• Instrument Name: Camouflaging Autistic Traits Questionnaire (CAT-Q)
• Scale Type: Self-report psychometric questionnaire
• Target Population: Adults (aged 16+) with or without a formal autism spectrum diagnosis
• Item Count: 25 items
• Response Scale: 7-point Likert scale (1 = Strongly Disagree, 2 = Disagree, 3 = Somewhat Disagree, 4 = Neither Agree nor Disagree, 5 = Somewhat Agree, 6 = Agree, 7 = Strongly Agree)
• Subscales:
  • Compensation (9 items): Items 1, 4, 5, 8, 11, 14, 17, 20, 23
  • Masking (8 items): Items 2, 6, 9, 12, 15, 18, 21, 24
  • Assimilation (8 items): Items 3, 7, 10, 13, 16, 19, 22, 25
• Scoring and Reverse-Coding:
  • Items are scored positively from 1 to 7 according to the respondent’s rating.
  • Five items are reverse-scored: Items 3, 12, 19, 22, and 24. For these items, scoring is inverted: 7 = Strongly Disagree, 6 = Disagree, 5 = Somewhat Disagree, 4 = Neither Agree nor Disagree, 3 = Somewhat Agree, 2 = Agree, 1 = Strongly Agree.
  • Total Score: Calculated by summing all 25 items (ranging from 25 to 175). Higher scores denote greater degrees of social camouflaging.
  • Subscale Scores: Calculated by summing the items within each domain: Compensation (range 9–63), Masking (range 8–56), Assimilation (range 8–56).
• Normative Reference Values: In the original validation study, autistic females exhibited the highest mean total score (M = 124.4, SD = 21.6), followed by autistic males (M = 110.1, SD = 25.7), non-autistic females (M = 97.4, SD = 23.3), and non-autistic males (M = 86.8, SD = 24.1). Clinically, a total score of ≥ 100 is frequently utilized as an empirical marker of elevated camouflaging behavior.

Permissions & Fee and Test Year

The Camouflaging Autistic Traits Questionnaire was officially published in 2019 by Laura Hull and colleagues in the Journal of Autism and Developmental Disorders. The research was supported by the Medical Research Council (MRC) and the Autism Research Trust.

In accordance with open-science initiatives, the CAT-Q is released under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0). It is freely accessible for non-commercial clinical use, professional psychological assessment, and academic research without fee or royalty payment, provided that appropriate credit is attributed to the original authors and the original publication is cited.

References

• Goffman, E. (1959). The presentation of self in everyday life. Anchor Books. https://monoskop.org/images/1/19/Goffman_Erving_The_Presentation_of_Self_in_Everyday_Life.pdf
• Hull, L., Mandy, W., Lai, M. C., Baron-Cohen, S., Allison, C., Smith, P., & Petrides, K. V. (2019). Development and validation of the Camouflaging Autistic Traits Questionnaire (CAT-Q). Journal of Autism and Developmental Disorders, 49(3), 819–833. https://doi.org/10.1007/s10803-018-3792-6
• Hull, L., Petrides, K. V., Allison, C., Smith, P., Baron-Cohen, S., Lai, M. C., & Mandy, W. (2017). “Putting on my best normal”: Social camouflaging in adults with autism spectrum conditions. Journal of Autism and Developmental Disorders, 47(8), 2519–2534. https://doi.org/10.1007/s10803-017-3166-5
• Livingston, L. A., & Happé, F. (2017). Conceptualising compensation in neurodevelopmental disorders: Reflections from autism spectrum disorder. Neuroscience & Biobehavioral Reviews, 80, 729–742. https://doi.org/10.1016/j.neubiorev.2017.06.005
• Milton, D. E. (2012). On the ontological status of autism: The ‘double empathy problem’. Disability & Society, 27(6), 883–887. https://doi.org/10.1080/09687599.2012.710008

Items of the Scale