‘If I behave like the stupidly kind character maybe I will stop being accidentally rude to people’: does reading fiction inform the social understanding and masking behaviours of autistic females?

ABSTRACT

Girls are consistently found to have higher literacy levels than their male peers and to consume more social narrative fiction. This study conducted interviews with four autistic girls to better understand how reading fiction can inform the camouflaging behaviours of autistic girls. Participants described how imitating characters embodying desired values, such as kindness, provided them with a practical behavioural template. However, literature is only a simulation of reality, and participants described social “failures” where the tone of the imitated text did not align with their social reality.

KEYWORDS:

• Female autism
• gender
• reading
• literacy
• masking

Introduction

Autism

Autistic children attending a mainstream school face a complex social environment (Dean et al. 2023, 2), where they are observed by both peers and educators. As Foucault argued, “the judges of normality are present everywhere” (Foucault 1975, 1418), creating social pressure for individuals to “fit in”. Expectation to conform to the social norm is amplified in the mainstream school environment, manifested through school uniform regulations, behavioural rules, and the students’ own social expectations for peers (Shacter and Jovenen 2018). An environment designed for neurotypical needs “does not effectively accommodate autistic characteristics” (Kang et al. 2023, 1), and can make schools a challenging setting for autistic students (Bower 2021, 917).

Autism is characterised by “differences in social communication and interaction, focused interests and sensitivity to sensory stimulation” (Perry et al. 2022, 800). Diagnosis is based on behavioural observation (Ochoa-Lubinoff, Makol, and Dillon 2023, 800), and this process is complicated by comorbid conditions such as anxiety, ADHD, and depression, which can obscure autistic behaviours. Whilst autism is diagnosed at a higher rate in males than females (MacFayden et al. 2023, p81), autism research has overwhelmingly used male samples, resulting in a male-centric perspective (Corbett et al. 2021, 128). Thus, where females fail to meet “male-typical” descriptions of autism, they are likely to miss out on diagnosis (Hull et al. 2017, 2520).

Female autism may be misdiagnosed or underdiagnosed due to gendered stereotypes and behavioural expectations (M. C. Lai, Lin, and Ameis 2022, 90; Russell et al. 2010). Self-soothing techniques tend to be less obvious in females, involving “hair twirling, teeth clenching, lip biting, or subtle hand movements” (Suckle 2021, 755). Camouflaging behaviours further complicate diagnosis of female autism (MacFayden et al 2023, p85), as autistic females may adeptly conceal difficulties in public, only to experience meltdowns once they return home (Mae Simcoe et al. 2022, 3634). Compared to autistic males, autistic females tend to demonstrate higher levels of social motivation (M. C. Lai, Lin, and Ameis 2022) through a desire for friendship and peer interaction (Hull, Petrides, and Mandy 2020, 308).

SEND students in mainstream schools

According to the National Autistic Society, “more than 70% of autistic children and young people are educated in mainstream schools” (2021, 5). However, the mainstream school environment has been persistently described by autistic students as stressful, with the “abundance of colourful materials over-stimulating”, the layout of the classroom “confusing” (Ashburner, Ziviani, and Rodger 2010, 20), and the school environment “contributing to stress and anxiety” (Adams, Simpson, and Keen 2018, 65).

The interaction of autistic traits with an over-stimulating and distracting school environment may cause distress which results in disruptive behaviour, such as “stereotypic utterances, physical mobility during quiet periods, off-task behaviour and task refusal” (Roberts and Simpson 2016, 1090). SEND (Special educational Needs and Disabilities) students are more likely to be sent home to “cool off”, put in isolation rooms, and to experience fixed or permanent exclusion from school (Porter and Ingram 2021). As disruptive behaviours may occur as a reaction to the school environment, punishing autistic students for misbehaviour in the same manner as their neurotypical peers is “unlikely to promote equity” (Emerson 2016, 105), and gives only a surface level appearance of school equality.

Masking

Autistic students may feel pressure to suppress natural behaviours in order to “pass” as neurotypical and “blend in” with their peers (McFayden et al. 2023, 83). This is known as masking, defined as when autistic individuals learn to perform “certain behaviours and suppress others in order to be more like the people around them” (Autism Service 2021, para 5). Whilst masking may help initiate friendships and avoid bullying (Ochoa-Lubinoff, Makol, and Dillon 2023, 385), it can delay the diagnostic process by hiding evidence of autism from the external eye, preventing parents and teachers from noticing neurodivergence (Mae Simcoe et al. 2022). Autistic females have been found to be highly adept at masking (M. C. Lai, Lin, and Ameis 2022), as they “observe and analyse the social and play behaviour of others and imitate this through their own play” (Kirkovski, Enticott, and Fitzgerald 2013, 2587), and mask social struggles through adaptive behaviours (Corbett et al. 2021, 128).

These camouflaging behaviours may be “explicitly learned or implicitly developed” (Hull, Petrides, and Mandy 2020, 309) and are informed by observation of neurotypical social interactions (M.-C. Lai et al. 2019, 1211), whether in day-to-day social reality or virtually through various media forms, such as television or the radio. This research study proposes that fiction books act as an additional key source of social and behavioural information for autistic girls.

Gendered reading habits

The international literacy study PIRLS has routinely found that girls outperform boys in every participating country (Department of Education 2017, 15); in England, girls were reported to have “higher engagement in reading lessons, confidence in reading, and liking of reading than boys” (Department of Education 2017, 16). The high intrinsic reading motivation of girls is contrasted by boys who tend to describe reading as “boring” (Ozturk, Hill, and Yates 2016, 704), demonstrating an absence of the self-motivating enjoyment that drives up female reading skill. The genre consumption of boys and girls also differs by gender; OECD (2010) reports that whilst girls tend to read fiction addressing “romance/relationships, poetry or school” (McGeown et al. 2016, 112) with popular themes including family and friendships (Mohr 2006), boys prefer non-fiction texts.

There is some evidence to suggest that autistic students struggle with reading comprehension (Grindle et al. 2020); whilst successfully decoding words, difficulties may arise when interpreting the meaning of the text (McIntyre et al. 2017, 1087; Ricketts et al. 2013). However, McIntyre et.al’s study into reading comprehension amongst autistic children reflected the male dominance within much autism research (Corbett et al. 2021), as 66 of the 81 participants were male. It is thus unclear to what extent this reflects reading comprehension in specifically autistic females. Furthermore, literacy ability in autistic individuals is heterogenous, with “no condition-specific profile of strengths and difficulties” (Grindle et al. 2020, 1–2).

This study aims to better understand how reading fiction can inform the camouflaging behaviours of autistic girls. Reading fictional literature may support understanding of social rules and behavioural expectations by providing access to a character’s thought processes, emotions, and behavioural motivations. Autistic girls may use this to inform masking, reproducing the behaviours observed and explained by neurotypical characters in narrative stories. To address this aim, the study conducted semi-structured interviews with four autistic girls, employing a qualitative methodology to give voice to their lived experiences.

Research methods

Research design

The study used a narrative research design. This qualitative interpretive methodology focuses on “understanding and accounting for the meaning of human experiences and actions” (Fossey et al. 2002, 720), ensuring participants’ unique experiences are at the centre of data collection. Narrative research is based on the understanding that people live in storied forms, and that “these stories are played out in the context of other stories” (Josselson 2011, 224). It is thus a fitting design to explore whether the lived experiences of autistic girls are affected by the fictional stories they read, and how their involvement in contexts structured by neurotypical social demands may necessitate behavioural adaptations to blend in with this dominant social narrative.

In narrative inquiry, the story of the participant is the “raw data” (Butina 2015, 190), and thus this research gives a voice to participants. The interview format is accessible, inviting participants to discuss their experiences, with the interviewer “assuming an open listening stance” (Polkinghorne 2007, 482) whilst richly descriptive verbal data is collected (Merriam 2002, 5).

Participants & recruitment

Participants were four female university students in their early twenties who had received, or were in the process of receiving, an autism diagnosis from a healthcare professional. In order to demarcate an area of focus within the autism spectrum, participants had to have attended a mainstream school.

A summary of the research, its intentions, and a pre-interview questionnaire was shared on a university student Facebook page. A voluntary opt-in approach was essential to the ethics of this research, and resulted in four volunteers. Of the four participants, two had received an autism diagnosis at the age of 14, one at the age of 16, and the fourth, aged 20, was on the waiting list for a diagnosis.

It is important to highlight that all the participants were of university level education; the data and experiences presented are thus of autistic females with high academic ability. As protecting participant anonymity was an ethical priority, there is an absence of information regarding the participants’ university experience and subject specialism. Future research undertaken in a less intimate setting and with a greater variety of educational backgrounds would, however, benefit from collecting more detailed personal data.

Materials and procedure

The researcher developed a pre-interview questionnaire by deconstructing the key components of the research aim: reading habits, autism diagnosis, and masking. The subsequent questions formulated for the questionnaire aimed to gather relevant information, concomitantly giving the participant time and space to consider the questions that would arise in the interview. This provided participants with a form of script to follow, taking into consideration that an interview may be a daunting prospect, particularly for an autistic individual.

Having given their informed consent, the volunteers received a questionnaire to complete prior to the interview. An explanation of the researcher’s use of the term “masking” was provided in the questionnaire, to ensure understanding of the questions.

The Questionnaire:

1. What kind of books do you like to read?

2. What is your favourite book/books?

3. How often do you read?

4. When were you diagnosed with autism?

5. How long did it take you to get an autism diagnosis?

6. What did it mean to you to get an autism diagnosis?

7. Is anyone else in your family autistic?

8. Did you/do you attend a mainstream school?

9. What 3 words would you use to describe your school experience?

10. Did you/do you feel the need to mask at school?

11. If you did/do mask, how did/does it make you feel?

12. Do you think reading informed your masking?

Once the questionnaire was completed, a meeting over Zoom was arranged with the participant; due to coronavirus, it was not possible to meet in person to conduct the interviews. In the semi-structured interview, the researcher orally repeated the questionnaire, and asked open-ended follow-up questions after the participant gave their answer. This supported the data collection by encouraging participants to expand their responses; for example, following the participant’s response to the question “what kind of books do you like to read?”, the researcher engaged with their answer to further discuss the reasons behind their literature preferences. The interview was recorded and transcribed; during this process, the names of participants were changed to ensure anonymity. The randomly generated pseudonyms aligned with the participants’ names, which were common names for individuals identifying as female and of British heritage.

Data analysis

The data was analysed using thematic data analysis. In narrative research, this involves a process of deconstruction, “linking seemingly unrelated facets of experience together” (Josselson 2011, 227). Interview transcripts were annotated with key themes, and recurring words were noted. The researcher then cross-referenced the transcripts against each other to highlight where participant experiences were shared and where they diverged. The participants’ responses were then organised and analysed according to the themes that had emerged, with the researcher using existing academic research to support the findings of the research project. It was necessary to have a high level of researcher reflexivity, interpreting the data objectively rather than drawing out themes and conclusions according to what the researcher expected to find (Fossey et al. 2002, 728).

Research findings

Reading habits

Participants expressed a strong preference for narrative fiction over expository texts; the primary favoured genre was fantasy, followed by traditional children’s classics such as “Heidi” or “The Secret Garden”. All participants read a considerable variety of genres; whilst Chloe named “Game of Thrones” as her favourite book, she also enjoyed “Harry Potter”, romance novels and “children’s books about families”. Amy listed an assortment of science fiction, fantasy, Greek myths, and detective stories amongst her favourite genres.

Reading at a considerably high frequency, participants demonstrated high intrinsic reading motivation, suggesting a “positive relationship” (Clark and Teravainen-Goff 2020, 12) between reading enjoyment, reading ability, and reading frequency (Bostock and Boon 2012); Amy commented “I learnt to read at the age of three … and then just kept going and never stopped!”, recalling “when I was about 11 … I read 56 books in a summer”. Participants demonstrated notable pride in their reading prowess:

“There was one section of gold books which were ‘really hard’ and no-one read them, and you bet your ass I was on them by year 6!” (Sammy)

This reflects the results of PIRLS 2016, which found that confident readers performed significantly higher than low confidence readers (DfE 2017, 15). The participants’ high reading frequency and enjoyment was in part motivated by the power of narrative fiction to act as a form of “escapism”, defined as a “mental diversion that allows individuals to ignore their everyday life” (Bitton 2021, para 2). Three of the participants used the word “escapism” to describe their enjoyment of fiction, and in particular, fantasy:

“Why though it’s fiction books in particular, I’ve always sort of liked escapism”. (Amy)

“I’ve always loved fantasy … . I love the escapism of it”. (Chloe)

“I enjoyed reading … I guess it was like escapism”. (Zoe)

Narrative fiction can act as a relief from day-to-day experiences, as stories provide an escape into an alternative existence (Hammet Price, van Kleeck, and Huberty 2009). Participants demonstrated reading for “literary experience”, to be “immersed into worlds with characters, atmospheres, feelings and ideas” (DfE 2017, 19). The escapism of reading additionally occurred through producing reliably foreseeable events in contrast to the unpredictability of day-to-day life, as stories were a stable and accessible source of comfort for some participants:

‘I like reading the same story over and over again because I don’t like looking at new things, I like knowing what is going to happen at any time. (Amy)

High reading frequency and reduced socialisation

It was suggested by several participants that high reading frequency and intrinsic reading motivation may have reduced practical opportunities to develop their social experiences and social skills. Sammy described how she would choose to read over interacting with peers:

‘Primary school was kind of like the start of the unhealthy obsession with reading, through primary school I kind of like spent most evenings, mornings, during breaks, literally whenever I could, I’d be reading. (Sammy)

Break-time is an opportunity for children to practice “essential social skills” (Locke et al. 2016, 2), and Pellegrini argued that it is in “free-play” that students develop social competence (Walker 2014, para 12). Chloe suggested that her “love of reading” not only reduced the time she spent socialising with her peers, but simultaneously may have hindered her social skills, as she was missing out on opportunities to gain practical social experience:

“I think that my love of reading made me stand out a bit just because I did it all the time, and it made me less equipped to like deal with social situations because my ability to learn social skills was hindered by the fact that I wasn’t spending as much time around children as I should be”. (Chloe)

Whilst Sammy and Chloe suggest that their high reading frequency was largely volitional and motivated by intrinsic enjoyment, Zoe recounted reading frequently as an alternative to unaccommodating playground activities. Despite her desire to be involved in the activities of her peers, she described these as ultimately inaccessible:

I guess it [reading] was like escapism in that sense … as much as I enjoy sports and I love running around and stuff, I never felt like I could get involved with it, because it just seemed to be very cliquey, and I didn’t understand how to get on with those people. Like it just felt a lot safer to spend time reading in the library than it did actually going outside and trying to get involved with the sports (Zoe).

I ended up just kind of like being bullied a fair amount because of the reading. (Zoe)

Zoe’s experiences are consistent with research showing that autistic children find the school environment to be an “isolating experience” (Eckart 2019, para 2). Zoes attributes this to the challenges in “getting involved” with her peers. This reflects Kang et al’s study which suggests that an “unaccommodating school environment” (Kang et al. 2023, 1) can hinder the social participation of autistic children. In this setting, Zoe found reading to be the “safer” option to interacting with her peers.

Locke et al argue that “there is a clear difference between solitary and solitude” (Locke et al. 2016, 9), and highlight the importance of asking students how they wish to spend recess. Whilst Sammy and Chloe suggest that reading was their activity of preference, choosing their solitary state, Zoe expressed a desire for participation in peer activities and yet found these inaccessible, experiencing forced solitude.

Social difficulties

Participants highlighted difficulties with neurotypical social demands, reflecting research suggesting that many autistic girls experience “feelings of ‘otherness’” during adolescence (Hull, Petrides, and Mandy 2020; Ochoa-Lubinoff, Makol, and Dillon 2023, 385). Chloe recalled her friends’ feedback of her social skills:

“They said, ‘you say thank you too much, you smile too much, you laugh too loudly’, and they said it in a much nicer way, but that was the overall gist”. (Chloe)

The phrase “too much” highlights how there is a socially accepted level at which to smile, laugh and communicate, and Chloe’s peers did not perceive her to be following this normative expectation. Failing to conform to social norms resulted in what participants understood to be “wrong” behaviour, highlighting the challenges some autistic adolescents experience in responding to “nuanced social cues across varying situations” (Dean et al. 2023, 2):

“Wherever I speak to people it always sounds like I’m saying the wrong thing, like am I coming across too strong or am I coming across too overexcited”. (Chloe)

“9 times out of 10 I’d misjudge the social situation and end up getting juice poured all over my head”. (Sammy)

“I’ve always quite liked the labels and knowing specifically where I sit … because I’ve always assumed that I’m a bad person … not that I’m a bad person, but that I’m bad at being a person, I’ve always been very specific in that difference, I’m not bad, I’m just bad at being a person”. (Sammy)

As Sammy explains, peers can police behaviour in the school environment, enforcing their own punishments for “wrong” social behaviour. Neurotypical behavioural expectations are a regulatory force, part of the “normalising society” we inhabit (Foucault 1976, 1449), with those who do not conform feeling as if they are “too much”, or even “bad at being a person”. Failure to follow socially dominant neurotypical behavioural demands can take an emotional toll on the autistic individual (Ochoa-Lubinoff, Makol, and Dillon 2023), and result in negative self-perception.

Masking aiding social situations

Behavioural masking can be used by autistic females to present “socially acceptable behaviour” (Mae Simcoe et al. 2022, 3634) to “fit in with others” or “avoid bullying” (Hull, Petrides, and Mandy 2020, 310). Three of the participants reported masking heavily during primary and early secondary school prior to diagnosis. This process was often not consciously done:

“I wasn’t aware of [what] masking was until I was about 13, and I wasn’t aware that I was doing it until I was about 17”. (Amy)

“I think when I was younger, I didn’t consciously realise I was doing it”. (Zoe)

“The entire concept is very new to me, I’m still trying to pick apart what I have been masking, and what isn’t masking”. (Sammy)

Masking may have been motivated – whether consciously or subconsciously – by a desire for social inclusion, involving “compensation techniques” (Hull et al. 2017, 1211), suppressing “socially unacceptable” behaviours (Corbett et al. 2021, 129), and imitating “body language and facial expressions” (Ochoa-Lubinoff, Makol, and Dillon 2023, 385). Amy described how her experiences of masking became less challenging as she gained a stronger understanding of social demands:

“The biggest one is controlling stimming, it’s the biggest noticeable thing that I do … as I’ve got older, I understand the social contract more … so it doesn’t feel like masking … it’s more like yeah this is the script, this is what you say, this is how you behave”. (Amy)

Chloe and Zoe described how they would imitate socially successful people in order to aid their camouflaging behaviour, observing neurotypical behaviours to similarly create a “script” for social scenarios:

“The two people who I copy social science off is my parents … I just tried to act like them”. (Chloe)

“You know that someone in your form is always very friendly and speaks in this way, so if you respond to them in the same way, you’ll get on with your form better”. (Zoe)

Zoe and Amy noted how masking helped with social outcomes by increasing positive interactions with peers, and also by avoiding actively negative interactions such as bullying:

“Better hide the fact that I don’t understand this because I don’t fancy getting picked on”. (Amy)

Corbett et alia suggest that masking is an artificial performance of “social behaviours considered more acceptable” (Corbett et al. 2021, 129), and several participants described masking as a role that they adopted, highlighting a disjunction between their neurodivergent selves and the neurotypical persona they presented:

“[masking] is very much a Zoe persona”. (Zoe)

‘I’ll very much play a role of like “I’m socially confident, I’m on it, I’m calm”. (Amy)

Masking can thus function as a coping strategy (Rødgaard et al. 2021, 482) to help autistic females to “better fit in” (Hull, Petrides, and Mandy 2020, 310) and avoid rejection (Ochoa-Lubinoff, Makol, and Dillon 2023, 385).

Masking informed by reading

Reading fiction was shown to facilitate participant understanding of behavioural norms, interpersonal relationships, and emotions of others. Oatley (2016, 619) comments: “if fiction is the simulation of social worlds, then, similar to people who improve their flying skills in a flight simulator, those who read fiction might improve their social skills”. As narrative fiction often involves comprehending “a story by assuming the perspective of a character” (Mar et al. 2006, 697), reading has the potential to aid social understanding by exploring the behavioural motivations, emotional experiences, and language choices of others. Amy described how reading fiction helped her to understand other people’s points of view:

“I can’t just look at people and know immediately what they are thinking, but when you are introduced to a character, because you are that character for a bit, then it does give you those other points of view”. (Amy)

“Most people don’t find sarcasm easier to read on a page, but I found it easier to learn on a page than I did in real life”. (Amy)

By transporting readers “imaginatively in a story” (Oatley 2016, 621), readers gain access to an alternative perspective. As Amy notes, narrative fiction explains social dynamics in an accessible way by reducing the complexities of social interactions, describing only that which is relevant to the progression of the narrative. “Literary fiction provides simulations of social complexes as they unfold” (Mar and Oatley 2008, 174); and by simulating “social complexes”, fiction provides an imitation of reality which can be instructive for individuals who have difficulties understanding social situations, as Zoe and Amy explain:

You’ve got kind of models for situations for instance, and kind of fictionalised expected realities, and feel like ok if people react, they might react like this because of this.

(Zoe)

“When I was younger … I was very good at analysing texts to try and be like ‘ok so she’s feeling this, and she says this. And he’s feeling like this, and he says this in reply’, and try to analyse, because you have that extra bit of emotional information, it was the only time I could put together how people were responding and how people were saying”. (Amy)

The participants described how stories functioned as a template for potential social scenarios or dynamics that could arise, allowing them to observe what the outcome of certain behaviours may be. Amy and Zoe described how they would imitate select characters to try and give a particular impression, as they could anticipate the outcome:

“Elements from Sarah from ‘The Little Princess’ I always tried to – that element of kindness – kindness was always very important to me because I don’t intend to be rude … so if I behave like the stupidly kind character maybe I will stop being … accidentally rude to people”. (Amy)

Characters embodying core values, such as kindness, can provide a model to imitate in order to aid inter-personal relationships, as by behaving like the “kind” character, the individual can ensure they too are being kind. Oatley’s comment “we need not lead just one life: by means of fiction, we can lead many lives” (Oatley 2016, 625) is thus powerfully dramatised by autistic girls using fiction to aid social behaviour; not only are they transported into the story’s alternative world when reading the book, but through their imagination and recall, they may transport elements of the book – such as particular characters – back into their own day to day reality. The phrase “leading many lives” thus gains significant weight. Amy “lived” the character when reading the book, her authentic self when not masking, an alternative personality when masking, and furthermore “lived” an additional life when masking through the framework of a particular character. It is thus little wonder that many autistic individuals struggle with a sense of identity (Miller, Rees, and Pearson 2021).

The vernacular of stories additionally influenced the participants, as Zoe commented how she would “pattern conversations” according to the literature she was reading:

‘[mum] was like “oh yeah, you fully spoke like someone from a school storybook for a long while”. (Zoe)

Reading thus once played a role in the participants’ communicative lives, imitating characters and behaviours to control their self-presentation and social skills. However, re-using elements of texts in day-to-day life was not always successful. Sammy recalled an incident when she quoted a line from a book, expecting a reaction in accordance with the narrative text, however the dialogue of the text did not align with the realities of contemporary life:

“There’s a scene where someone calls a character a bitch, and they reply ‘I’ll be an even bigger bitch by the time I’m through with you!’ and I was like ‘that’s such a cool line!’ … . it was year 8, I was on my bike … and the kid, he calls me a bitch, and I say it … . And I just got laughed off the street … and I was so confused, like oh it worked so well in the book!” (Sammy)

This highlights how literature is ultimately only a “simulation” (Oatley 2016, 619) of the world, and interactions will not always play out the same way as they do in texts, resulting in social “failures” where the tone of the text does not sufficiently align with the social reality.

Discussion

The aim of this research was to better understand how reading fiction can inform the camouflaging behaviours of autistic girls. Researcher reflexivity was shown to be essential for the validity of this data and its interpretation, as participants raised unanticipated views regarding the relationship between reading, masking, and social experiences.

Several participants suggested that high reading frequency may have hindered the development of their social skills and inter-personal relationships. Whilst for some participants reading was their activity of preference, other participants found that the inaccessibility of peer interactions led them to resort to reading during free time. The participants reported spending break-times reading instead of interacting with other students, reducing their opportunity to develop practical social skills. This finding highlights the importance of teacher support during free periods to facilitate the inclusion of autistic students with their peers (Kang et al. 2023; Locke et al. 2016) so they may develop their social skills (Dean et al. 2023, 1) and be involved with peers when they wish to.

Participants suggested that reading may have supported the development of their social understanding, with fictional literature assuming a didactic role in many of the participants’ childhood reading experiences. There was a high level of internal agreement amongst participants concerning how narrative fiction created a window into the psychology and emotional experiences of a character. This aided their understanding of social situations as they could observe the behavioural motivations and emotions of characters, which translated into their own understanding of social behaviour.

Narrative fiction was additionally shown to inform the masking behaviours of the participants by providing a model for specific desirable behaviours, such as confidence or kindness, and offering scripts for particular interactions or situations. The participants described how they could transport elements from the stories they read into their own lived reality to facilitate social interactions and carefully mould how they presented themselves in social situations; as summarised by Amy, “if I behave like the stupidly kind character maybe I will stop being accidentally rude to people”. Genres focusing on child-based inter-personal relationships, such as the traditional “school book” stories, were thus particularly valuable for providing insight into neurotypical psychology, by producing relevant scenarios for participants to learn what is normatively considered “appropriate” social behaviour. The research findings highlighted that where the participants reproduced phrases or behaviours from fictional texts in which the inter-personal interactions did not sufficiently align with the social context of the autistic individual, the result was what they considered to be a social “failure” as their behaviour diverged from the social “norm” and highlighted them as “different”.

Weaknesses and limitations

This research is not without its limitations. To truly be research “with” rather than “on” autistic individuals, it is necessary to involve autistic input throughout the development of the research design and the data analysis (Chown et al. 2017). Engaging with a specifically autistic demographic whilst developing the research design would ensure that this was a participatory research project, with increased “meaningful input from autistic people in autism research” (Fletcher-Watson et al. 2019, 943). This extends to the data analysis process, which should use member checking to ensure autistic input, from either the participants themselves or from autistic researchers or consultants (Milton and Bracher 2013). This would reduce the occurrence of misinterpretation, improving the validity and quality of the research.

It is additionally important to acknowledge the breadth of factors that may have played into the participants’ literary habits, experiences, and social skills. Whilst it was important to protect the anonymity of the participants, future research undertaken in a less intimate environment would benefit from acquiring more contextual information to strengthen the data by “accounting for other factors” (Jorgenson et al. 2020, 453). All the participants were university students attending the same institution; a greater variety of participants would have provided a broader perspective and greater understanding of the literary habits and experiences of autistic females.

Recommendations for future research

Future research would benefit from a more rigorous data collection process. Collecting data through a series of in-person interviews would facilitate the establishment of a neutral power dynamic between the participant and the researcher, building a relationship of trust to facilitate information disclosure and participant ease. This would further the richness of the data and provide autistic participants with more opportunity to discuss their experiences with the researcher.

It would be valuable for future research to explore how reading fiction can inform the camouflaging behaviours of both males and females from a variety of demographic backgrounds. This research study has suggested that the masking behaviours of autistic females are partly informed by the literature they consume, and that reading habits may interplay with the development of social skills. It would be useful to ascertain whether this is similar in males, as well as in autistic individuals from a greater variety of backgrounds. This may further elucidate some of the differences between autistic individuals identifying as male or female, and provide greater understanding regarding how female autism diverges from male autism. It was not within the scope of this research to consider gender self-identification beyond the male-female binary; however, it is recommended that future research would benefit from doing so.

Acknowledgments

I would like to express my great appreciation to all the participants who volunteered their time to participate in this research. Their willingness to discuss their personal experiences provided valuable information about female autism and made the research project possible.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Jemima Hill

Jemima Hill Having graduated in English and Education from Durham University, Jemima Hill is now studying for a Masters degree in International Child Rights and Development at Kings College London.