Seminar | Assessment of Anomia Using Computer Adaptive Testing and Algorithmic Error Classification

Wednesday, October 23, 2019 - 11:15am to 12:15pm
Department of Communication Disorders Conference Room | 358 N. Pleasant St | ground floor

In this talk, Gerasimos Fergadiotis (Portland State University) will discuss his work in automating the assessment of confrontation naming tests, including both the development of algorithms for automatically classifying paraphasias and computer adaptive testing.

This event is supported by the ISSR Scholars Program, promoting successful research and grants development for social science faculty across UMass Amherst. Dr. Fergadiotis visits UMass to collaborate with ISSR Scholar Jacquie Kurland. His public talk is co-sponsored with the Department of Communication Disorders.

Dr. Gerasimos Fergadiotis Ph.D., CCC-SLP  is an Associate Professor in the Department of Speech and Hearing Sciences at Portland State University. Dr. Fergadiotis’ research focuses on developing and evaluating psychometric applications to quantify clinically relevant aspects of language processing and relate observed behaviors to their underlying cognitive substrates. His training and research interests integrate perspectives from cognitive psychology, psychometrics, statistics, and clinical aphasiology. In recent years, Dr. Fergadiotis has been involved in several projects including the development of a computer adaptive measurement tool for assessing anomia (1R03DC014556; PI: Fergadiotis); building algorithms for automatic detection and classification of paraphasias (1R56DC015999-01, 1R01DC015999-01; PIs: Bedrick, Fergadiotis); measuring clinically important constructs at the discourse level; and assessing the efficacy of anomia treatment. Common threads across these projects include latent variable modeling using item response theory and structural equations, Bayesian approaches to inference, and test validity as a central psychometric property for clinical test and scale development.