HE ASSESSMENT OF LEARNING EMOTIONAL STATE USING EEG HEADSETS

• Published in: eLearning and Software for Education 2015, Vol.11 (1), p.587-593
• Main Authors: LUPU, Robert Gabriel, Ungureanu, Florina
• Format: Article
• Language: English
• Subjects: Education
• ISSN: 2066-026X; 2066-8821

In the last decade, the interest in the development of tools and devices for recognizing human emotions in learning process has increased continuously. It was proved that the electrical brain activity using electroencephalography (EEG) represents a useful methodological tool in understanding cortical processes that underlie performance and students' engagement in learning activities. Specific emotional states, like mental stress, concentration, relaxation, fatigue, and cognitive increase activation in Delta (0.5-4 Hz), Theta (4-7 Hz), Alpha 1 and 2 (8-12 Hz), Beta 1 and 2 (12-30 Hz), Gamma (30-70 Hz) frequencies. For example, the increase of frontal Beta-1 spectral power is associated with cognitive tasks demands and the decrease of Beta-1 power values reflects relaxation. Alpha is the dominant frequency in the human EEG and is generated in widespread reas of the cortex through cortico-cortical and thalamo-cortical interactions reflecting emotions. The EEG-based Brain Computer Interface (BCI) systems have been widely studied in different medical labs with high quality EEG recording devices which are much too expensive and need special employment. As an alternative to these professional equipment, several low-cost EEG devices were developed for out of the lab applications e.g. schools, universities, sports medicine, psychology or even neurophysiology. Two pertinent and detailed studies developed at Princeton University (http://compmem.princeton.edu/experimenter) and University of Mons (www.biomedical-engineeringonline.com/content/12/1/56) reveal that the Emotiv Epoc headset could be taken into account. This low-cost EEG device has higher relative operational and maintenance costs than its medical-grade competitor, it "is able to record EEG data in a satisfying manner" but it should only be chosen for non critical applications such as games, communication or feedback evaluation in a well-known scenarios. Our study aims to use the Emotiv headset for evaluating the students' emotional state in learning process. For this goal, some steps were completed: EEG data acquisition and analysis. The headset connects wirelessly to any PC via a USB dongle allowing freedom of movement, has 14 channels/sensors and the data sample rate is 128Hz. Due to the problem of eye/head movements automatic artefact detection must be completed and only artefact-free segments must be used for analysis. A good option is to use the Matlab toolbox EEGLAB, freely available open source resea