Measuring (meta)cognitive processes in multimedia learning: Matching eye tracking metrics and think‐aloud protocols in case of seductive details

Abstract Background In recent years, eye tracking has become a prominent method for learning research as it is assumed to indicate (meta)cognitive processes. However, there is little empirical evidence for hypothesized relations between eye tracking indicators and specific (meta)cognitive processes so that construct validity of used metrics can be questioned. Objectives The main goal was to provide validity hints in order to create an empirical basis for interpreting specific eye tracking indicators in terms of respective (meta)cognitive processes of multimedia learning. Methods N = 60 students learned with multimedia instructional material. Referring to a process model of multimedia learning, correlations between theoretically deduced eye tracking indicators with verbalized (meta)cognitive processes of multimedia learning captured by think‐aloud protocols were examined. In addition, the sensitivity of both process measures of (meta)cognitive processes was regarded considering the well‐investigated seductive details effect of an established multimedia instruction in a two‐group design. Finally, serial mediations were calculated in order to investigate whether both process measures complement one another in a joint explanation of the seductive details effect. Results and Conclusions Eye tracking indicators and verbalized (meta)cognitive processes did only partly correspond as it was shown by correlation and serial mediation analyses. However, both measures were sensitive to indicate the seductive details effect. Thus, even though the study provided insights in how validation could be possible, further systematic research will be needed for validating eye tracking indicators of specific (meta)cognitive processes in multimedia learning.