Naturalistic Social Communication in Autistic Females: Identification of Speech Prosody Markers

Univ Of North Carolina Chapel Hill, Chapel Hill NC

Investigators

Linked publications & trials

Abstract

PROJECT SUMMARY Autistic females are diagnosed at a significantly lower rate and at later ages than males. As a result, much less is known about the female autism phenotype. Studies have indicated that diagnostic tools may be biased towards males and that findings characterizing the autism phenotype may fail to capture females due to little or no female representation in their sample. Emerging research has identified unique sex differences in the autistic population across several modalities, including language. The handful of studies that have characterized language use in autistic females have done so through examining word use, identifying nuanced ways that autistic females use discourse markers, pronouns, and emotional words. These differences may contribute to autistic females’ heightened social motivation, using language to compensate for social difficulties, although no study has directly explored this possibility. We propose to bridge this gap through a focus on speech prosody, an understudied area of pragmatic language that allows individuals to convey meaning and emotion through speech. Prosody plays an important role in social-communication and in the early detection of autism, yet few studies have included females in their sample and no study has closely examined potential sex differences. In Aim 1, we will characterize differences in key prosodic markers of pitch and tempo in young children (4-8 years) by sex, diagnosis, and conversational content using a novel adaptation of natural language sampling methodology. In Aim 2, we will use multi-modal data collected through an ongoing study (R01-HD103895-01A1) to explore the relationship between speech prosody, social traits, and formal language. We hypothesize that the autistic (ASD) group will significantly differ from the non-autistic (NA) group in multiple measures of pitch and tempo, and that these differences will be exacerbated during conversation around a “special interest” topic. We predict that while ASD females will display more “typical” prosody in common prosodic measurements of pitch, they will be differentiated from ASD males and NA females in more socially complex measurements of intonation, speech rate, and pause. We predict overall negative relationships between social measurements and speech prosody markers associated with ASD: These relationships will be stronger for ASD females than males, while relationships between speech prosody and language scores will be stronger for ASD males. Training aims include advancing Ms. Putnam’s skills in multi-modal methodologies, eye tracking, language processing, and statistical analyses. This project will enhance current understanding of speech prosody and how it relates to other important aspects of the autism phenotype. By further characterizing the female autism phenotype, this work will contribute to more timely diagnoses and comprehensive care of this understudied population.