RT Journal Article SR Electronic T1 Is voice a biomarker for autism spectrum disorder? A systematic review and meta-analysis JF bioRxiv FD Cold Spring Harbor Laboratory SP 046565 DO 10.1101/046565 A1 Riccardo Fusaroli A1 Anna Lambrechts A1 Dan Bang A1 Dermot Bowler A1 Sebastian Gaigg YR 2016 UL http://biorxiv.org/content/early/2016/04/03/046565.abstract AB Lay Abstract Individuals with Autism Spectrum Disorder (ASD) are reported to speak in distinctive ways. Distinctive vocal production should be better understood as it can affect social interactions and social development and could represent a non-invasive biomarker for ASD. We systematically review the existing scientific literature reporting quantitative acoustic analysis of vocal production in ASD. We identify repeated and consistent findings of higher pitch mean and variability but not of other differences in acoustic features. We identify a recent approach relying on multiple aspects of vocal production and machine learning algorithms to automatically identify ASD from voice only. This latter approach is very promising, but requires more systematic replication and comparison across languages and contexts. We outline three recommendations to further develop the field: open data, open methods, and theory-driven research.Scientific Abstract Individuals with Autism Spectrum Disorder (ASD) tend to show distinctive, atypical acoustic patterns of speech. These behaviours affect social interactions and social development and could represent a non-invasive biomarker for ASD. We systematically reviewed the literature quantifying acoustic patterns in ASD. Search terms were: (prosody OR intonation OR inflection OR intensity OR pitch OR fundamental frequency OR speech rate OR voice quality OR acoustic) AND (autis* OR Asperger). Results were filtered to include only: empirical studies quantifying acoustic features of vocal production in ASD, with a sample size > 2, and the inclusion of a neurotypical comparison group and/or correlations between acoustic measures and severity of clinical features. We identified 32 articles, including 27 univariate studies and 15 multivariate machine-learning studies. We performed meta-analyses of the univariate studies, identifying significant differences in mean pitch and pitch range between individuals with ASD and controls (Cohen’s d of about 0.4 and discriminatory accuracy of about 61%). The multivariate studies reported higher accuracies than the univariate studies (63-96%). However, the methods used and the acoustic features investigated were too diverse for performing meta-analysis. We conclude that multivariate studies of acoustic patterns are a promising but yet unsystematic avenue for establishing ASD biomarkers. We outline three recommendations for future studies: open data, open methods, and theory-driven research.