MacFarlane, Heather Salem, Alexandra C. Adams, Joel R. Lawley, Grace O. Bedrick, Steven Dolata, Jill K. Fombonne, Eric

autism automated language measures expressive language sampling neurodivergent disfluency Natural Language Processing Autistic Disorder Language 2021 2021 Structural and pragmatic language deficits are core symptoms of Autism Spectrum Disorder (ASD) and predict long-term outcomes. Clinical measurement of language proficiency is cumbersome and costly; however, Automated Language Measures (ALMs) can be automatically calculated from language samples. Objectives: 1. examine language differences between three clinical groups (ASD, Attention Deficit Hyperactivity Disorder (ADHD), and Typically Developing (TD)); 2. analyze the convergent validity of these measures by calculating correlations between the ALMs and standardized language measures; 3. investigate the accuracy of each individual ALM in predicting ASD status; and 4. investigate any gains in accuracy obtained by combining all ALMs together to predict ASD status. Oregon Health and Science University Abstract 10.6083/0v838127z https://digitalcollections.ohsu.edu/record/9229/files/MacFarlane-Heather-OHSU-ResearchWeek-2021.pdf