The Statistical Characteristics of P3a and P3b Subcomponents in Electroencephalography Signals

Abstract

The P300 waveform is a common event-related potential (ERP) component in electroencephalography (EEG) signals used in clinical neurophysiology, brain-computer interface (BCI), and cognitive neuroscience research. Comprehensive documentation of P300 is needed to support research growth in those areas, especially for P300 subcomponents, P3a and P3b, which commonly used to quantify EEG characteristic. Therefore, this study aims to explore the quantitative characteristics of P3a and P3b subcomponents during the flanker test and Stroop task experiment. ERP waveforms were obtained from 16 healthy subjects to analyze the statistical features of the P3a and P3b subcomponents. This study also used decision trees for measuring the importance index. The result indicates that P3a can be defined as a prominent peak in the time window of 250–350 ms at the frontal area. Positive peak amplitude, area amplitude, mean amplitude, and the prominence of the positive peak have different distributions between group with P3a subcomponent and group without. Whereas P3b is characterized as a positive sloping peak in 300–700 ms at the parietal area. All statistical features influence P3a and P3b identification. However, features related to statistical characteristics of positive peaks have a greater importance index compared with others.