Toward Exploiting EEG Input in a Reading Tutor

摘要

A new type of sensor for students' mental states is a single-channel EEG headset simple enough to use in schools. Using its signal from adults and children reading text and isolated words, both aloud and silently, we train and test classifiers to tell easy from hard sentences, and to distinguish among easy words, hard words, pseudo-words, and unpronounceable strings. We also identify which EEG components appear sensitive to which lexical features. Better-than-chance performance shows promise for tutors to use EEG at school.