Autonomic nervous system approach to measure physiological arousal and assess user experience in simulation-based emergency procedure training environment

Bui, Sinh (2018) Autonomic nervous system approach to measure physiological arousal and assess user experience in simulation-based emergency procedure training environment. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (3MB)

Abstract

Given of the impossibility of exposing trainees to hazardous scenarios for ethical, financial and logistical reasons, virtual-environment (VE) based simulation training has been adopted in various safety-critical industries. Through simulation, participants can be exposed to a variety of training scenarios to assess their performance under different conditions. Along with performance measures, physiological signals may provide useful information about trainees’ experience. The objective of this research is to investigate the ability of physiological measurement to provide information on trainees’ experiences by assessing their physiological arousals in a simulation-based training environment. In this study, 38 participants used a VE-based program called AVERT (All-hands Virtual Emergency Response Trainer). This program was developed for training emergency response procedures for the offshore petroleum industry. Signals of the autonomic nervous system (ANS), specifically electrocardiography (ECG), electrodermal activities (EDA), and respiration (RSP), were used to assess physiological arousal levels for 8 different conditions of an emergency evacuation task. On average, neutral and training conditions could be distinguished with an 82.4% average accuracy by a subject-specific machine learning classifier. Most importantly, arousal levels in different training scenarios provide useful information that performance measures alone do not reveal.

Actions (login required)

View Item