Automated boredom detection using multimodal physiological signals
Project ID: OER 14/22 YR
Subject area: Digital literacies
Principal Investigator: Dr Yuvaraj Rajamanickam
Email: yuvaraj.rajamanickam@nie.edu.sg
About the project
Emotions can influence decision making, learning, and other aspects of human behaviour. Therefore, computer-driven detection of human emotions has become an essential field of research. Recognizing and detecting emotions in the field of education is in one concrete example in action (Picard, 2000; Sottilare & Goldberg, 2012). Emotion detection using computers devices have been useful within education since the 1980s (Bereiter, 2002; Bruce & Rubin, 1993; Scardamalia & Bereiter, 1994). Thus, if computers used in educational services are equipped with suitable sensors for the detection of learner’s emotions, then their potential as an educational tool may be extended further. Numerous studies have been constructed to detect and act upon various emotions in the field of education. However, there is one emotion, boredom, which has gained little attention as a target emotion from educational researchers. For instance, in education or learning contexts, boredom can disrupt learning by impeding focused attention, resulting in lower engagement and motivation to learn (Cui, Yao, & Zhang, 2017).
Previous studies indicate that emotions should be associated with changes in central nervous system (CNS) or autonomic nervous system (ANS) activity, which can be measured using physiological signals such as electroencephalogram (EEG), electrocardiogram (ECG), galvanic skin resistance (GSR), and eye gaze (Nummenmaa, Glerean, Hari, & Hietanen, 2014; Purves, Augustine, & Fitzpatrick, 2001). Yet, the neurophysiological correlates of boredom are underexplored. Given these observations, the proposed research aims to systematically explore the comprehensive range of central and autonomic signals that may characterize boredom, to understand that emotional state and the patterns of physiologcial responses (measured using EEG, ECG, GSR, and eye gaze) associated with boredom. We will design and conduct an experiment, which will use video stimulus to evoke boredom. The proposed work will also develop an automatic boredom detection system using multimodal physiological signals and machine learning, which can support methods to reduce boredom and increase engagement, a prerequisite for reaching flow states during learning.
Looking for
|
What will be expected of you
|