Masking Is Not a Social Skill: What Camouflaging Research Actually Measures

People call it a social skill when an autistic adult makes the room comfortable. Camouflaging research measures something quieter: the strategies used to hide or compensate for autistic traits, and the cost of doing that work in public.

This is Autism Digest’s first research brief. It stays inside published numbers. It does not offer a cure, a training program, or a new way to diagnose anyone.

What masking and camouflaging are

In 2017, Laura Hull and colleagues asked 92 autistic adults about “putting on my best normal.” Social camouflaging is the use of strategies to make autism less visible in social situations. The short-term cost those adults named most often was exhaustion. The mask hid the work. It did not remove it.

The same research group later built a questionnaire so that work could be counted instead of only described. The Camouflaging Autistic Traits Questionnaire, or CAT-Q, is a 25-item self-report scale. It is not a social-skills inventory. It asks how often a person compensates, hides, or tries to blend in.

ICD-11 asks clinicians to be aware of masking or coping behaviours when they assess autism. Awareness is not the same as treating the mask as progress.

The CAT-Q’s three parts

The CAT-Q has three factors. Compensation (9 items) covers strategies used to get around social and communication difficulties: scripts, copied gestures, studying how other people talk. Masking (8 items) covers strategies used to hide autistic characteristics or present a non-autistic persona: monitoring the face, forcing the body to look relaxed. Assimilation (8 items) covers the work of fitting in during social situations that do not feel like the self: performing “normal,” avoiding interaction, needing support just to stay in the room.

The CAT-Q’s three factors, with item counts and one-line definitions. Source: Hull et al. 2019.

On a 7-point Likert scale, autistic adults in the exploratory sample (N = 200) scored higher than non-autistic adults (N = 202) on every factor. Assimilation showed the widest gap: 5.29 (SD 1.15) against 3.32 (SD 1.27). Compensation was 4.42 (1.31) against 2.89 (1.27). Masking sat closer: 4.55 (1.35) against 4.29 (1.10). Totals were 4.79 (0.99) against 3.48 (1.04). Autistic participants scored significantly higher on the total and on all three factors.

Exploratory-sample means on Compensation, Masking, and Assimilation. Error bars are standard deviations. Source: Hull et al. 2019, Table 3.

Masking is the factor that looks most shared. Looking “fine” is not an autism-only habit. The larger gaps sit in compensation and assimilation — building a social performance, and surviving a room that still does not fit.

Who was studied

These are adult self-report samples, recruited online. The CAT-Q development paper used a full sample of 354 autistic and 478 non-autistic adults. People who were self-diagnosed were excluded from the autistic group. The 2020 gender analysis used 306 autistic and 472 non-autistic adults. Cage and Troxell-Whitman surveyed 262 autistic adults. Recruitment ran through UK research databases and social media. That is not a clinic waiting room, a classroom, or a global census.

Read every number in this brief as a description of those adults, on those questionnaires, in those papers.

The mental-health cost

Cage and Troxell-Whitman asked 262 autistic adults to complete the CAT-Q and the DASS-21. They scored the same 25 CAT-Q items as a sum (range 25–175), not as a 1–7 item mean. The CAT-Q total mean was 116.12 (SD 20.48). On the DASS-21, mean depression was 19.68 (10.99), anxiety 15.53 (9.64), and stress 24.06 (9.25).

They grouped people by how consistently they camouflaged: consistently low (n = 68), switchers (n = 78), and consistently high (n = 78). The low group had significantly lower stress than both switchers and the high group, and lower anxiety than the high group. Depression did not differ significantly between groups.

Classified group sizes and the published DASS-21 comparisons. Group means are not shown because they were not in the verified figures for this brief. Source: Cage & Troxell-Whitman 2019.

That is not a ladder you can use to rank a person. It is one comparison inside one adult sample: more consistent camouflaging lined up with more stress, and the high group carried more anxiety than the low group. Depression did not sort the same way.

Cassidy and colleagues (2018) treated camouflaging as a risk marker in research on suicidality in autistic adults. This brief does not repeat a percentage from that paper. The marker is the point. A person who “appears fine” is not automatically safe.

Why people said they mask

Ninety-one people in the Cage sample wrote in other reasons for camouflaging. The coded themes, by how often they appeared, were fitting in or passing (42), avoiding retaliation or bullying (34), concerns about impression (32), habit (16), and internalised stigma (12).

Other-reasons coded from 91 write-ins. Source: Cage & Troxell-Whitman 2019.

Those are not five personality types. They are reasons adults gave for doing expensive work. Fitting in sat next to staying safe. A workplace that rewards the first one is often the same room that made the second one necessary.

What this is not

This is not a cure, and it is not a girls-only story.

In Hull, Lai, and Mandy (2020), autistic females scored higher than autistic males on total CAT-Q after age and autistic-like traits were controlled (partial η2 = 0.08). The difference showed up in Masking (η2 = 0.05) and Assimilation (η2 = 0.06), not in Compensation. Non-autistic adults showed no gender difference. In Cage and Troxell-Whitman, raw CAT-Q totals were 118.90 (SD 18.83) for females and 114.25 (SD 21.36) for males; that gap was not significant (p = 0.13). Men mask. A higher mean in autistic women is not a rule about who is allowed to be tired.

The CAT-Q is not a diagnostic test. It does not prove someone is autistic, and a low score does not prove they are not. It is a research measure of self-reported camouflaging behaviours in adults who could complete an online questionnaire. It does not speak for children, for people who do not use written self-report, or for anyone outside these samples.

It is also not a training target. Teaching a student or an employee to raise their CAT-Q score is teaching them to spend more of the thing these papers keep linking to exhaustion, stress, and anxiety.

What to do with this

If someone “appears fine” at work, believe the evening. The meeting that looked easy may have used compensation, masking, and assimilation at once. The colleague who is articulate on Zoom and silent by dinner is not being inconsistent. They are off shift.

At home, treat the crash as information, not as a character problem. Food, quiet, stims, and no recap of the day are not rewards for good behavior. They are how a nervous system comes off a performance. The same adult who can host a birthday may need the next morning empty.

Do not demand unmasking as proof of authenticity. Avoiding retaliation and bullying was the second most common write-in reason. Safety is a real reason to keep the costume on. The useful move is smaller: stop treating the costume as the skill. Reduce the number of rooms that require it. Write “appears fine” out of the performance review, the present levels, and the family story that says this person is fine because the room was fine.

Sources

Laura Hull, K. V. Petrides, Carrie Allison, Paula Smith, Simon Baron-Cohen, Meng-Chuan Lai, and William Mandy published “Putting on My Best Normal”: Social Camouflaging in Adults with Autism Spectrum Conditions in the Journal of Autism and Developmental Disorders in 2017. The sample was 92 autistic adults. The short-term cost named most often was exhaustion. https://doi.org/10.1007/s10803-017-3166-5

Laura Hull, William Mandy, Meng-Chuan Lai, Simon Baron-Cohen, Carrie Allison, Paula Smith, and K. V. Petrides published Development and Validation of the Camouflaging Autistic Traits Questionnaire (CAT-Q) in the Journal of Autism and Developmental Disorders in 2019 (volume 49, pages 819–833). The full sample was 354 autistic and 478 non-autistic adults. The 25-item scale has three factors: Compensation (9 items), Masking (8 items), and Assimilation (8 items). Exploratory-sample means are reported in Table 3. https://doi.org/10.1007/s10803-018-3792-6 Open access: https://pmc.ncbi.nlm.nih.gov/articles/PMC6394586/

Laura Hull, Meng-Chuan Lai, Simon Baron-Cohen, Carrie Allison, Paula Smith, K. V. Petrides, and William Mandy published Gender Differences in Self-Reported Camouflaging in Autistic and Non-Autistic Adults in Autism in 2020 (volume 24, issue 2, pages 352–363). The sample was 306 autistic and 472 non-autistic adults. https://doi.org/10.1177/1362361319864804

Eilidh Cage and Zoe Troxell-Whitman published Understanding the Reasons, Contexts and Costs of Camouflaging for Autistic People in the Journal of Autism and Developmental Disorders in 2019. The sample was 262 autistic adults. https://doi.org/10.1007/s10803-018-03878-x Open access: https://pmc.ncbi.nlm.nih.gov/articles/PMC6483965/

Sarah Cassidy, Louise Bradley, Rebecca Shaw, and Simon Baron-Cohen published Risk Markers for Suicidality in Autistic Adults in Molecular Autism in 2018. The paper treats camouflaging as a risk marker. This brief does not quote a percentage from that study. https://doi.org/10.1186/s13229-018-0226-4