Abstract
Retest effects refer to performance improvements in a final test by completing previous tests with the same or similar testing materials. Improvements in test-related skills and/or increasing familiarity with the stimulus materials are considered sources of the retest effect. The present study analyzes retest effects in the context of spatial thinking, considering complementing perspectives (behavioral performance, cognitive processing, and cognitive workload). N = 141 participants completed a recently developed ability test for the visualization factor of spatial thinking (R-Cube-Vis Test). This test provides the opportunity to monitor the progression of changes in solving behavior from item to item within each of the six distinct difficulty levels. Items of one difficulty level all require the same spatial solving strategy but vary in visual appearance. Multi-level models were estimated, with items on level 1 and participants on level 2. Results demonstrated retest effects as changes from the beginning to the end of a set of items within each difficulty level by increasing accuracy. Gaze patterns showed the development of solving strategies by participants through, e.g., shifting the focus on relevant item parts. Increasing familiarity with the stimulus materials was indicated in reduced reaction times and increased confidence ratings, but also by the results of a pupillary-based cognitive workload measure. Furthermore, differences between participants with overall high vs. low spatial ability were considered. In addition to a deeper understanding of the underlying mechanisms of the retest effect, the complementing perspectives provide more detailed information about individual ability profiles for diagnostic purposes.