Shedding light on decision making: Detecting cognitive stages in EEG data using Hidden Multivariate Pattern Models

Muller, S.K. (2024) Shedding light on decision making: Detecting cognitive stages in EEG data using Hidden Multivariate Pattern Models. Master's Thesis / Essay, Artificial Intelligence.

Preview	Text Shedding_light_on_decision_making.pdf Download (5MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (136kB)

Abstract

Cognitive models are among the major tools used to study human behavior. In order to develop these models we require knowledge of the cognitive stages people go through when performing their tasks. The development of cognitive models used to be done mainly with behavioral results, causing us to only use the reaction time per trial, making it difficult to determine the onset of each cognitive stage. Instead, we want to work with EEG data as this provides us with much more information. Significant cognitive events result in EEG peaks, and locating these peaks would allow us to pinpoint when any processing stages would occur. However, even nowadays the identification of processing stages in EEG data remains challenging. In this project, we test a new method for detecting cognitive stages in EEG data by comparing our results with known results found in previous experiments. To this aim, we collected EEG data from a decision making experiment based on contrast manipulation with 22 participants, and used a pipeline applying a novel approach of Hidden Multivariate Pattern Models (HMP). Our results confirm and build upon previous findings in a contrast based decision making experiment. Furthermore, we show that HMP is capable of identifying the onset and durations of the expected stages within a perceptual decision making task. All in all, HMP shows promise when it comes to identifying EEG peaks, which should aid us in the development of more accurate cognitive models in the future.

Item Type:	Thesis (Master's Thesis / Essay)
Supervisor name:	Borst, J.P. and Weindel, G.M.
Degree programme:	Artificial Intelligence
Thesis type:	Master's Thesis / Essay
Language:	English
Date Deposited:	02 Feb 2024 15:39
Last Modified:	02 Feb 2024 15:39
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/31904

Actions (login required)

View Item