Statistical Representation of Emotions for Puzzle Workload using Electroencephalogram and Heart Rate Variability

Abstract

We attempted to express the psychological quantity of executing workload through statistical analysis of indices extracted from an electroencephalogram (EEG) and a heart rate variability (HRV) score of subjects when they were asked to solve jigsaw puzzles. First, we conducted a regression analysis of the emotional score of the mood evaluation questionnaire after the completion of the workload and the indices of the EEG and HRV before and after the start and completion of the workload, and thereafter confirmed the strongest correlation before the completion of the workload. Next, we conducted a principal component analysis of the indices of the EEG and HRV before the completion of the workload, and thereafter confirmed that three principal components were obtained that correlated with “friendship,” “fatigue-inertia,” and “vigor-activity” in the mood evaluation questionnaire. Therefore, we demonstrated that the physiological quantities of the EEG and HRV indices could statistically express the psychological quantities of positive/negative emotions, even with small data.