The effects of EEG analysis on computer-assisted instruction

Abstract

The primary research objective presented in this dissertation is a demonstration of the feasibility of incorporating electroencephalographic monitoring and feedback into a conventional computer-assisted instruction environment. The underlying hypothesis is that the knowledge and appreciation of primitive physiological characteristics observed during certain forms of mental processing can form the basis for programming efforts designed to significantly enhance the effectiveness of automated educational procedures. As an initial effort in this direction, a straightforward anagram solving exercise was organized to detect and encourage a mental posture characterized by brief but concerted concentrations. The difficulty level of each anagram was under strict program control. The subject's electroencephalogram was continually monitored and analyzed, and an indication of his mental posture calculated as a series of parameters. The relative proximity of the subject's mental state parameters to a program-projected parameter set determined the difficulty level of subsequent anagrams. In addition, an indication of mental progress based on this proximity statistic was fed back to the subject as a quiet tone in his headphone set. The subject was encouraged to solve the anagram as quickly as possible while at the same time attempting to keep the feedback tone as low as possible..