Analysis of computer game player stress level using EEG data

The goal of our research is to analyze the stress level of a human player during a game session. In this paper, we propose a system to classify certain human states recorded by EEG analysis during playing computer games activity. At this moment, the classification procedure is still off-line, thus the recording will not interfere the game. From that recording, we can differentiate certain player states. The games we used for the experiment are different challenges in racing games, chess, and first person shooter with different types of difficulty levels. The ultimate goal from this research is to have an adaptive game that will adjust its difficulty levels to the appropriate brain activities. We preprocessed the data using, Independent Component Analysis to remove mostly eye movement artifacts. Then, we extracted several features mostly related to the frequency domain of signals acquired. Finally using Waikato Environment for Knowledge Analysis (WEKA), we tried several classifiers method to know which one give a better result. In our research, we conducted three experiments in which three stress levels were compared; no-stress, average and hiah-stress level. We were able to classify player state with an average of 79.089% in accuracy level using, Decision Tree classifier. We also performed a comparison between classifying 3 user states and pair-wise classification (only two states). On average, we achieved 78.7864% for distinguishing three classes of states. While, classifying two-states achieved an average of over than 80% in accuracy level.