Autistic Traits, Empathizing-Systemizing, and Gender Variance Olivia Hendriks, MSc1, Yimeng Wei, MSc1, Varun Warrier2, Gareth Richards, PhD1,2* 1 School of Psychology, Faculty of Medical Sciences, Newcastle University, UK 2 Autism Research Centre, Department of Psychiatry, University of Cambridge, UK * Corresponding author address: School of Psychology, Newcastle University, 2.27 Ridley Building 1, Queen Victoria Road, Newcastle upon Tyne, UK; email: [email protected] Abstract Previous research indicates a link between autism and gender variance, though the basis for this association is not fully understood. The current study examined autistic traits (as measured by the Autism Spectrum Quotient [AQ]) and empathizing and systemizing (as measured by the Empathizing Quotient-Short [EQ-S] and Systemizing Quotient-Short [SQ- S]) in a sample of n=89 UK adults representing a broad range of gender identities and sexual orientations. Compared with cisgender individuals (i.e. those who identify as the same gender as that assigned at birth), gender variant participants had significantly higher AQ and SQ-S scores, and stronger systemizing relative to empathizing (D-score). Further analysis revealed that there were significant differences between cisgender females and those assigned female at birth who do not identify as female (transgender AFAB), but not between cisgender males and those assigned male at birth who do not identify as male (transgender AMAB). These findings are broadly in line with the extreme male brain theory of autism, and may be relevant for developing effective support for gender variant and/or autistic individuals. Keywords: Autism; Autistic Traits; Empathizing-Systemizing; Gender Variance; Transgender 2 Introduction Gender identity is distinct from the sex assigned at birth (typically male or female), and represents a person’s sense of their own gender. There is a diverse range of gender identities including transgender, non-binary, gender fluid, bigender (and many others). A person who is cisgender has a gender identity that is the same as the sex assigned at birth. In the current paper we refer to individuals whose gender does not match their sex assigned at birth ‘gender variant’ individuals, and define gender variance as “Gender identity or gender expression which does not conform to masculine or feminine gender norms, for example having a gender identity different to one’s biological sex, or not defining oneself within the gender binary of ‘male/female’” (Cooper et al. 2018, p. 3996). Gender dysphoria on the other hand, is a strong and persistent dissonance between the sex assigned at birth and one’s gender identity (American Psychiatric Association 2013). Gender variance is no longer viewed as a psychiatric diagnosis but as a reflection of individual differences, and so the diagnosis of gender dysphoria now focuses on associated distress rather than the gender identity itself. Although rigorous epidemiological research is lacking, gender dysphoria is uncommon and known to affect relatively more natal males than natal females (Zucker and Lawrence 2009). An emerging literature has examined the co-occurrence of autism/autistic traits and gender variance/gender dysphoria (for recent reviews, see Glidden et al. 2016; Øien et al. 2018; Van Der Miesen et al. 2016). This research generally points to there being an overrepresentation of autism diagnoses (de Vries et al. 2010) and elevated autistic traits (Skagerberg et al. 2015) in children and adolescents presenting at gender clinics, and elevated likelihood of diagnosed gender dysphoria (Hisle-Gorman et al. 2019) or parent-reported gender variant identity (Strang et al. 2014) in autistic children and adolescents. Although the literature relating to adults is more limited (see Glidden et al. 2016), it does suggest that autistic adults are at 3 increased likelihood of reporting gender variance (Cooper et al. 2018; George and Stokes 2018), and that there is an increased prevalence of autism diagnosis (Heylens et al. 2018) and elevated level of autistic traits in adults attending gender clinics (Heylens et al. 2018; Jones et al. 2012; Pasterski et al. 2014) or reporting self-identified gender variance (Kristensen and Broome 2015). The association between autism (and autistic traits) and gender variance is gaining clinical and research interest (Glidden et al. 2016; Strang et al. 2018a; Van Der Miesen et al. 2016), and attention to the specific needs of autistic individuals with gender variance (or gender dysphoria) appear to be warranted. Furthermore, the presentation of autism in girls and women has been shown to vary subtly from that of boys and men (Bargiela et al. 2016; Dworzynski et al. 2012; Lai et al. 2017), and people’s gender variance is sometimes questioned because of their autism diagnosis (Strang et al. 2018b). Efficacy of care for autistic individuals seeking treatment for gender dysphoria may also be affected by some of the core features of autism. For instance, it could be that autistic individuals attending gender clinics experience increased difficulty in communicating their current needs to medical practitioners (Strang, et al. 2018a). Such ideas suggest that a better understanding of the association between autism and gender variance is required in order to provide more effective support. Empathizing and systemizing may be important when considering the link between autism and gender variance (e.g. Stagg and Vincent 2019). Empathizing is the ability to identify and respond appropriately to emotions and thoughts in others, whereas systemizing is the drive to analyse and build systems based on input-operation-output rules. Males, on average, tend to score higher on measures of systemizing, whereas females tend to score higher on average on measures of empathizing, and autistic people tend to score higher in systemizing and lower in 4 empathizing than neurotypical males (Baron-Cohen et al. 2014) (we use the term ‘neurotypical’ in the current context as referring to individuals who are not autistic). The difference in standardised empathizing and systemizing scores (D-scores) is also higher in males than in females (Baron-Cohen et al. 2005; Greenberg et al. 2018), and higher in autistic people than neurotypical people. Preliminary findings suggest that adolescent (Di Ceglie et al. 2014) and adult transgender males (Smith 2014; Stagg and Vincent 2019) may show a masculinised (higher) systemizing and (lower) empathizing profile. So far, few studies have examined autism/autistic traits in relation to gender identities other than those that fit a binary classification (i.e. transgender males, transgender females, cisgender males, and cisgender females). However, Kristensen and Broome (2015) examined a range of gender identities (e.g. androgyne, trans, transsexual, third gender), and reported that autistic traits were particularly elevated in individuals identifying as gender-queer (i.e. a person who does not subscribe to the conventional gender distinctions of male or female, but identifies as neither, both, or a combination of male and female genders). Furthermore, autistic individuals are more likely than neurotypical individuals to report their gender identity as atypical (i.e., not congruent with their birth sex) (Bejerot and Eriksson 2014; George and Stokes 2018). Self-identified non-binary individuals also score higher in measures of autistic traits and systemizing, and lower in empathizing (Stagg and Vincent 2019). The present study aims to address gaps in the literature regarding how traits related to autism are associated with gender variance by sampling a broad range of gender identities. Instead of comparing only transgender males and transgender females (or non-binary) participants with cisgender males and females, we used an online survey to investigate the relationship between multiple gender variant identities and autistic traits, empathizing and systemizing. 5 Based on previous reports indicating an elevated autism prevalence in adults recruited from gender clinics (Heylens et al. 2018) or self-identifying as gender variant (Kristensen and Broome 2015), and that the association between autism (or autistic traits) and gender variance (or gender dysphoria) may be stronger in those assigned female at birth (George and Stokes 2018; Jones et al. 2012), we proceeded with the following hypotheses: (1) compared to cisgender individuals, gender variant individuals would have more male-typical scores on traits associated with autism (i.e. higher AQ, SQ-S, and D-scores; lower EQ-S scores), and (2) any observed differences would be larger for comparisons of cisgender females and transgender AFAB individuals than for comparisons of cisgender males and transgender AMAB individuals. Methods Participants To ensure a broad range of gender identities, we contacted >200 lesbian, gay, bisexual, transgender, queer+ (LGBTQ+) support groups across the UK, and also advertised our study on social media and online survey exchanges (e.g. SurveyCircle, PollPool). We aimed to recruit similar numbers of cisgender and gender variant participants, and to ensure that each of the four groups (i.e. cisgender females, cisgender males, transgender AFAB, transgender AMAB) examined in this study were represented by a broad range of sexual orientations. No exclusion criteria were specified other than that participants should be at least 18 years of age. Materials & Design We used the Autism-Spectrum Quotient (AQ; Baron-Cohen et al. 2001) to quantify autistic traits.