Abstract
Measuring and assessing the cognitive load associated with different tasks is crucial for many applications, from the design of instructional materials to monitoring the mental well-being of aircraft pilots. The goal of this paper is to utilize EEG to infer the cognitive workload of subjects during intelligence tests. We chose the well established advanced progressive matrices test, an ideal framework because it presents problems at increasing levels of difficulty and has been rigorously validated in past experiments. We train classic machine learning models using basic EEG measures as well as measures of network connectivity and signal complexity. Our findings demonstrate that cognitive load can be well predicted using these features, even for a low number of channels. We show that by creating an individually tuned neural network for each subject, we can improve prediction compared to a generic model and that such models are robust to decreasing the number of available channels as well.
Highlights
The performance of complex tasks requires the integration of various mental resources, such as task-related knowledge, working memory, attention and decision making
We focused on Lempel-Ziv (Tononi and Edelman, 1998) complexity, Multi Scale Entropy (MSE) (Abásolo et al, 2006) and Detrended Fluctuation Analysis (DFA) (Rubin et al, 2013)
Neural complexity metrics- We focused on three measures of complexity, Lempel-Ziv complexity (LZC) (Zhang et al, 2001), Multi Scale Entropy (MSE) (Abásolo et al, 2006)
Summary
The performance of complex tasks requires the integration of various mental resources, such as task-related knowledge, working memory, attention and decision making. Stevens et al (2006) assessed subjects as they were learning to diagnose disorders of organ systems and Mak et al (2013) focused on performance improvement in a visual-motor task. Both studies showed a decrease in cognitive load metrics, with an increase in task familiarity. We chose a setting in which problem difficulty was rigorously validated and is commonly used in the psychological literature (see below) Another limitation of previous studies is that cognitive workload was assessed using discrete levels, often only two or three levels (Aricò et al, 2016a,b). We use a continuous scale for workload
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
