Developing Deep Learning Models for Classifying Cognitive Load States by ECG and Pupil Dilation Data Integration to Study Human Behavior

Cognitive load is a widely studied subject, especially in contexts like cognitive ergonomics, training and education, and UX research. While there exist subjective methods to measure cognitive load, researchers are looking into ways to objectively and, in particular, automatically classify different cognitive load states. Such automatic detection would be beneficial in various contexts ranging from real-time operator monitoring to adaptive training (~flow). In this thesis project, we aim to delve into the development of automated methods for classifying cognitive load states. Our goal is to create a robust framework that can objectively assess and differentiate between varying levels of cognitive load without relying on subjective human assessments. The project mainly involves the exploration of various data sources, sensor technologies, and machine learning and Deep Learningtechniques to develop a reliable and practical system for automatic cognitive load classification. Successful completion of this project will contribute to the advancement of human-centered technology and research and provide valuable insights into optimizing cognitive ergonomics across a wide spectrum of applications –think about adaptive training systems, improved worker productivity and wellbeing, UX research, and many more.