Abstract: The main objective of this study was to investigate the potential of combining subjective and objective measures of learning process to uncover the mechanisms underlying the spatial contiguity effect in multimedia learning. The subjective measures of learning process were self-reported cognitive load ratings and the objective measures were eye-tracking and EEG measures. Learning outcome was measured by scores on retention and transfer posttests. A sample of 78 university students participated in a between-subjects design in which a multimedia slideshow lesson on how lightning storms develop was presented either with printed text as a caption at the bottom of each illustration (separated presentation) or with printed text placed next to the corresponding part of each illustration (integrated presentation). Regarding spatial contiguity, the integrated group spent significantly more time looking at the text (d = 0.64), but significantly less time looking at irrelevant illustrations (d = 1.10), and reported a significantly lower level of extraneous load (d = 0.57), compared to the separated group. As expected, they also scored significantly higher on the transfer test (d = 0.49). Students who performed best on posttests reported a lower level of extraneous load (d = 0.56). Furthermore, EEG based alpha band activity was predictive of intrinsic cognitive load but not predictive of extraneous cognitive load, and EEG based theta activity was not predictive of intrinsic or extraneous load. The results suggest that subjective and objective measures of cognitive load can provide different information to test the theoretical mechanisms involved in multimedia learning.