19267
Web-Based Toolkit for Multimodal Data Analysis in ASD Research

Friday, May 15, 2015: 10:00 AM-1:30 PM
Imperial Ballroom (Grand America Hotel)
O. O. Wilder-Smith and M. S. Goodwin, Northeastern University, Boston, MA
Background: Understanding human behavior and development through analysis of internal psychophysiological signals can provide objective insights into subtle behavioral, social, cognitive, and affective characteristics of individuals. Internal measures can afford caregivers, clinicians, and researchers with information that can both enhance and complement their own observations. For example, physiological signals, like electrodermal Activity (EDA), heart rate (HR), and heart rate variability (HRV), provide a window into an individual's internal affective and regulatory state, which may be otherwise unavailable or unobservable through audio-visual cues, especially in individuals with limited communication abilities and/or who have impoverished non-verbal displays (e.g., reduced facial expressivity). Recent advances in wearable sensing technology have made it possible to record ambulatory psychophysiological data from children with Autism Spectrum Disorder (ASD) together with time-synchronized video data in a variety of naturalistic settings. However, qualitative analysis and review of multimodal physiological data synchronized with video remains a cumbersome process, typically requiring multiple specialized software packages and complex configurations of data.

Objectives: To provide a suite of free, open, and easy-to-use tools to support qualitative analysis of physiological and behavioral data together with time-synchronized video to support ASD research.

Methods: We have developed the Computational Behavioral Science Toolkit (CBST), a free, web-based application for viewing and annotation of multimodal psychophysiological and behavioral data together with time-synchronized video. CBST runs in any modern web browser without the need for installing specialized software, and data loading and interaction is accomplished via a simple drag and drop interface.

Results:  Members of our lab and collaborators from both technical and non-technical fields have used CBST to review, present, and perform qualitative analyses on data collected in several on-going multi-site studies. Our demonstration will include opportunities for interacting with the application, present examples of physiology and video data from both assessment (ADOS-2) and intervention studies we are engaged in, and provide information on how other researchers can access and use the toolkit in their own research.

Conclusions: CBST provides a simple, free, web-based platform for viewing, analyzing, and presenting multimodal psychophysiological and behavioral data.