Autism in adult women and social masking

Social masking covers the strategies used to hide autistic traits or compensate for their effects during an interaction. Hull and colleagues described it in 2017 from the accounts of 92 adults and drew out a three-stage sequence: a motivation, a set of techniques, a set of consequences. The motivation most often cited was not concealment but access: holding a place in a group, keeping a job, making friends.

The questionnaire built from those accounts, published in 2019 by the same team, keeps twenty-five items across three factors. It was validated with 354 autistic and 478 non-autistic adults, with an internal consistency of 0.94. The three factors explain what the word covers better than any definition does.

The descriptions collected in qualitative studies look alike from one country to the next. They are not traits, they are actions, and that is what makes them recognisable.

• Rehearsing a conversation in advance: anticipating questions, practising answers, preparing a line for getting out.
• Copying reactions observed elsewhere: an intonation, a laugh at the right moment, an expression learned by watching other people.
• Monitoring your own face and body throughout the exchange, the way you proofread a text while writing it.
• Forcing eye contact you believe is expected, and losing the thread of what is being said while you do.
• Holding back a repetitive movement that settles you, until you are on your own again.
• Collapsing after an evening that went well, unable to speak or move for hours afterwards.
• Making a career of observation: reading widely about social rules, keeping written ones, revising before a dinner.

The last item explains the apparent paradox. Someone who masks heavily can come across as entirely at ease in company, and genuinely be so for an hour. The cost does not show inside the interaction: it arrives afterwards, and it is almost never reported back to the people who were there.

The most telling figure comes from a meta-analysis published in 2017 by Loomes, Hull and Mandy, which pooled 54 studies and more than 13.7 million participants. The male-to-female ratio varies there according to how people were found, and that gap is the useful information.

The reading is direct. Search in the general population and you find roughly three men to one woman. Start from people already identified and you find four and a half. The gap between those two figures is not a biological difference: it is the share of women who meet the criteria without having been recognised.

A Dutch survey published in 2013 covering 2,275 people points the same way and adds the timing: depending on age group and on the categories used at the time, women were identified later than men. The qualitative work of Bargiela, Steward and Mandy, published in 2016 with fourteen women aged 22 to 30 recognised late, describes the mechanism from inside: they report having pretended to be normal, and judge that their gender led the professionals they met to look elsewhere.

The 92 adults interviewed by Hull and colleagues in 2017 named exhaustion first, then something harder to put into words: no longer being sure who you are after years of presenting an edited version of yourself. That second cost recurs in nearly every qualitative study published since.

Cage and Troxell-Whitman surveyed 262 autistic adults in 2019. Two patterns stood out as the heaviest: masking everywhere, and masking in some contexts only while being unable to let go elsewhere. Women more often cited what the authors called conventional reasons, such as getting by in a formal setting like work.

A larger study published in 2021 by Hull and her team linked self-reported masking to more symptoms of generalised anxiety, social anxiety and depression. Two qualifications matter: once autistic traits and age were accounted for, the additional effect remained small, and no interaction with gender was found. In other words this is not a female phenomenon, it is a phenomenon more often described by women.

• Fatigue that a good night does not clear, because it follows social load rather than physical load.
• A gap between the impression given and the actual state, which makes asking for help harder to phrase.
• Effort that is invisible and therefore never credited, by those around you or by the professionals you meet.
• The risk of giving up whole categories of social situation rather than keep paying for them.

It changes one specific thing: it moves the question. A questionnaire written around autistic traits asks what you feel in noise, facing a change, inside an exchange. A questionnaire written around masking asks something else: what you do so that none of it shows, and what that costs you afterwards. Someone who has been compensating for thirty years often answers the two very differently.

That is also the limit of the exercise. A self-observation questionnaire cannot confirm or rule out a condition: it asks nothing about how far back the difficulties go, nothing about what else might explain them, and it cannot see what you do not yet have words for. What it usefully produces are sentences you can take to an appointment.

1. For two weeks, note what you prepare before an exchange and what you do afterwards. The afterwards is the most informative part, and the most often forgotten.
2. Answer the questionnaire built around social masking, then the one about autistic traits in adulthood. Comparing the two readings is worth more than a single score.
3. Take those notes to your family doctor rather than the score. Both questionnaires are offered at the foot of this page.