Validation of a Signal-Processing Pipeline for Electrodermal Activity Responses in Multimodal Paradigms

Author(s):

Thomas Rohaly, Leah Enders, Alex Stauff, and Michael Nonte

Abstract:

Situational awareness (SA) is considered a foundational skill for Soldiers, and maintaining SA is critical to reduce errors and casualties during combat. Soldiers are exposed to high levels of stress not only at work but also in their daily lives, negatively impacting their ability to maintain SA and overall job performance and safety. Advances in wearable technology and mixed reality have allowed us to increase our understanding of how stress conditions impact job performance for in-lab scenarios. Specifically, including multimodal sensors to jointly capture physiological responses (e.g., electrodermal activity and heart rate) and kinematics (e.g., joint position and movements) provides a comprehensive view of SA maintenance and job performance. However, many hurdles remain before we can feasibly use these types of sensors to capture stress responses in mobile, in-field operating paradigms. The purpose of this study was to develop a signal-processing pipeline for electrodermal activity signals to be used as an in-house standard. In this study, participants navigated through a desktop-based virtual environment and completed a visual search task. During this task, surprising flashbang events occurred intermittently and the team collected isolated electrodermal activity responses. Data collected from the task was then utilized to develop a processing pipeline for cleaning and formatting data for analysis, such as using standardized signal decomposition models. Results from this study conclude that the developed pipeline is a feasible option for cleaning and preparing electrodermal activity signal for analysis and encourage its use in future paradigms to further test and refine the pipeline.