"I Was Like a Bird Without Wings": Autistic Women's Retrospective Experiences in General Schools.
Teachers overlook identical autism traits in girls—train them on the female phenotype.
01Research in Context
What this study did
Zakai-Mashiach (2023) asked primary teachers to rate short stories about pupils.
Each story had the same autism traits, only the child’s name and hobbies changed.
Teachers judged whether the traits pointed to autism.
The goal: see if identical symptoms are missed when the child looks like a girl.
What they found
Teachers were far less likely to flag autism when the child was female.
The girl version of the same story was seen as shy, not autistic.
The boy version was sent for evaluation.
The finding shows a clear bias, not a difference in actual symptoms.
How this fits with other research
Mandy et al. (2012) saw the same pattern in real classrooms.
They found girls with ASD had milder repetitive play and fewer teacher-reported problems.
Their data foretold the bias Mati later caught in an experiment.
Ivy et al. (2017) seems to clash. They said standard autism tools measure boys and girls the same way.
The clash fades when you see the difference: W et al. tested the test, Mati tested the teacher.
The test works fine; the teacher’s eye is the problem.
Hodge et al. (2025) later tracked real kids and showed girls are referred six months later than boys, proving the bias costs time.
Why it matters
If you screen or train school staff, add a girl-focused lens.
Show examples of the female phenotype: quiet, shy, anxious, not just loud or line-obsessed.
Push for referral checklists that ask, "Would I see this differently if the child were a boy?"
Catching girls early spares them the ‘bird without wings’ feeling Mati heard in their stories.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a girl vignette to your next teacher in-service and ask staff to write down what they see.
02At a glance
03Original abstract
Autism has long been considered a predominantly male condition. It is increasingly understood, however, that autistic females are under-recognized. This may reflect gender stereotyping, whereby symptoms are missed in females, because it is assumed that autism is mainly a male condition. Also, some autistic girls and women may go unrecognized because there is a "female autism phenotype" (i.e., a female-typical autism presentation), which does not fit current, male-centric views of autism. Potential biases shown by educators, in their role as gatekeepers for an autism assessment, may represent a barrier to the recognition of autism in females. We used vignettes describing autistic children to test: (a) whether gender stereotyping occurs, whereby educators rate males as more likely to be autistic, compared to females with identical symptoms; (b) whether recognition is affected by sex/gender influences on autistic presentation, whereby children showing the male autism phenotype are rated as more likely to be autistic than those with the female phenotype. Ratings by primary school educators showed a significant main effect of both gender and presentation (male phenotype vs. female phenotype) on estimations of the child in the vignette being autistic: respondents showed a bias against girls and the female autism phenotype. There was also an interaction: female gender had an effect on ratings of the female phenotype, but not on the male phenotype vignette. These findings suggest that primary school educators are less sensitive to autism in girls, through under-recognition of the female autism phenotype and a higher sensitivity to autism in males. LAY SUMMARY: Educators have an important role in identifying children who need an autism assessment, so gaps in their knowledge about how autism presents in girls could contribute to the under-diagnosis of autistic girls. By asking educators to identify autism when presented with fictional descriptions of children, this study found that educators were less able to recognize what autism "looks like" in girls. Also, when given identical descriptions of autistic boys and girls, educators were more likely to identify autism in boys. These results suggest that primary school educators might need extra help to improve the recognition of girls on the autism spectrum. Autism Res 2020, 13: 1358-1372. © 2020 International Society for Autism Research, Wiley Periodicals, Inc.
Journal of autism and developmental disorders, 2023 · doi:10.1002/aur.2316