Assessing cognitive load in adaptive hypermedia systems: Physiological and behavioral methods

Abstract

It could be advantageous in many situations for an adaptive hypermedia system to have information about the cognitive load that the user is currently experiencing. A literature review of the methods proposed to assess cognitive load reveals: (1) that pupil size seems to be one of the most promising indicators of cognitive load in applied contexts and (2) that its suitability for use as an on-line index in everyday situations has not yet been tested adequately. Therefore, the aim of the present study was to evaluate the usefulness of the pupil size index in such situations. To this end, pupil diameter and event-related brain potentials were measured while subjects read texts of different levels of difficulty. As had been hypothesized, more difficult texts led to lower reading speed, higher subjective load ratings, and a reduced P300 amplitude. But text difficulty, surprisingly, had no effect on pupil size. These results indicate that pupil size may not be suitable as an index of cognitive load for adaptive hypermedia systems. Instead, behavioral indicators such as reading speed may be more suitable. © Springer-Verlag 2004.