Stress Pattern Recognition Using Spatio Temporal Features Based on EEG Sport Simulation

Several emotional factors that can induce stress affect participants performance during the competition, such as choking under pressure. This study aims to evaluate the performance of golf players under psychological pressure by creating a scenario in a golf putting simulation test. We forced each participant to compete by making a monetary reward scenario based on their winning score to increase each participant's stress level. Their winning score is classified as poor or good by measuring the distance to the target achievement. The objective is to detect the stress pattern by observing the prominent features and window segments in the alpha band. Our approach is processing the EEG data by wavelet transform and extracting the data into spatio temporal features in the central and occipital lobe. Our investigation found that the maximum amplitude and band power feature values were higher when the participants performed well, and this pattern is reversed when the participant's performance is poor. Meanwhile, based on brain area activity in a good performance, we found that central area activity is high from the first to the fourth segment. In contrast, occipital area activity is high only until the third segment. Following these parts, brain activity decreases compared to its previous state. This study also discovered that the significant segment window is located in the tenth segment, with a P-value < 0.05 between good and poor performance. Thus, this analysis's results can be used to recommend neurofeedback training for athletes to prevent choking under pressure.