Toward a passive brain computer interface for simultaneous detection of mental workload and stress

Baheri, Mahsa (2022) Toward a passive brain computer interface for simultaneous detection of mental workload and stress. Doctoral (PhD) 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

A passive brain-computer interface (pBCI) is a system that continuously adapts a human-computer interaction to the user’s mental state. An example would be a system that aims to prevent traffic accidents by sending alerts to a truck driver when a state of drowsiness is detected. Key to the efficacy of such a system is the reliable estimation of the user’s state via neural signals, acquired through non-invasive methods like electroencephalography (EEG). Typically, in pBCI studies, the state being explored (e.g., fatigue, frustration, boredom, attention) is considered in isolation, and no other aspect of the user’s state is taken into account. In real-life scenarios, however, different aspects of the user’s state are likely to be changing simultaneously - for example, their cognitive (e.g., level of mental workload) and affective (e.g., level of stress/anxiety) states. This inevitable confounding of different states needs to be accounted for in the development of state detection algorithms in order for them to remain effective when taken outside the lab. In this work, simultaneous classification of two mental states via EEG is investigated for the first time. Specifically, mental workload and stress are explored since detection of both of these states would be useful in a variety of applications, including for improving safety in high risk work environments. Individually, both mental workload and stress have been studied extensively in the passive BCI literature, however in real-life scenarios they often vary concurrently within an individual. First, the effect of varying each state on classification of the other state was investigated to indicate if/how mental workload and stress confound one another. Then, different classification algorithms were proposed and evaluated to mitigate the confounding effects of variation in mental workload on the detection of stress and vice versa. Finally, a processing pipeline suitable for realizing an online BCI for simultaneous detection of mental workload and affective state was investigated. This work represents a step toward the ultimate goal of realizing a functional, reliable, and robust passive BCI capable of detecting both mental workload and stress.

Actions (login required)

View Item