Speaking the Same Language? a Comparison of the Language and Communication Profiles of Girls and Boys with High Functioning Autism

Background: Young females with Autism Spectrum Disorder (FwASD) present with more subtle impairments in social interaction (Lai et al 2011) and restricted interests (Mandy et al 2012) than male peers (MwASD). This may impact negatively on correct diagnosis of ASD in females (Kopp and Gillberg, 2011) which in turn reduces access to suitable services. Less is known about gender differences in communication for young people with ASD, especially high-level language/discourse. There is some evidence that gender differences are found in both functional communication (Park et al 2012) and social impact of communication deficits (Sedgewick et al, 2016). Detailed investigation into subtle gender differences in language/discourse skills could support understanding of the female phenotype of autism and indicate areas of particular interest for this group; causal factors of poor well-being, appropriate diagnosis and bespoke therapy provision.

Objectives: To provide a thorough analysis of language and discourse skills for FwASD. To investigate all language domains: receptive/expressive skills, at word/sentence and above sentence level, and across pragmatic/semantic and language of emotion ability. To create a profile of skills for FwASD and compare skills sets with MwASD and typically developing females (FwTD) and males (MwTD).

Methods: Measures of structural language and functional communication were compared between 13 female and 13 male children (aged 8.11-11.06) with High-Functioning Autism Spectrum Disorder (HFASD); performance IQ normal range. 26 typically developing children (TD) were matched for age and gender. Subtle differences in PIQ were controlled during analysis. Participants were recruited through autism charities, NHS trusts and participant data bases. Assessments included: Standardised and experimental measures of vocabulary, sentence level and discourse level processing; Experimental pragmatic/semantic measures of non-linguistic communication, non-literal interpretation of spoken language and complexity of narratives; Novel experimental tasks of receptive and expressive emotion vocabulary. All assessments were conducted by clinical specialists, in school or at home. Analysis was undertaken using a 2 (Gender) x 2 (Group) analysis of variance (ANOVA), controlled for PIQ.

Results: Overall trends indicate that females with HFASD will outperform males with HFASD on a range of pragmatic and semantic tasks. However, they perform worse on these measures than TD females. TDs had consistently better scores than HFASD in above sentence level tasks (although not in measures of vocabulary or sentence level grammar where all groups showed matched results). Female HFASD perform similarly to TD females on some language of emotion measures (receptive and semantic category naming) and better than male TD or HFASD. These may represent relatively spared skills compared to gender norms.

Conclusions: Outcomes indicate a specific profile of language and communication strengths/weaknesses for female HFASD. They support the theory of a distinct female phenotype of ASD already identified in social skills and repetitive behaviours. They may indicate why FwASD fail to meet criteria for current diagnostic schedules, but also why they struggle to match communicative expectations of their female TD peer group. Results from this study have clinical implications for diagnostic services and speech and language intervention. As well as explaining difficulties maintaining friendships and managing well-being.