Measuring mental rotation and perspective taking in children. A validation study

Abstract

Mental rotation (MR) and perspective taking (PT) are key spatial abilities that have been linked to various cognitive and academically relevant skills (e.g., Baumeler & Frick, 2017; Frick, 2019; Laski et al., 2013; Mix & Cheng, 2012). Despite their importance, there is a lack of validated instruments for measuring these skills in young children. In this study, the psychometric properties of four MR tasks were examined by presenting them to 96 children aged 6 to 9 years. Among these tasks, two were specifically designed for children, whereas the other two were established tasks that were originally created for adults, with a computer-based task and a paper-pencil task in each category. The study also included a computerized assessment of spatial PT, an ability that is conceptually related to but clearly distinct from MR, thus serving as a stringent test for discriminant validity. The new MR tasks for children showed good to excellent reliabilities, exceeding those of the reference tasks for adults. Almost all MR tasks indicated significant improvement in performance with increasing age, highlighting their sensitivity to developmental changes across the age range studied. The only exception was the adult computerized task, which showed only a non-significant age trend, pointing to its inadequacy for tracking developmental progress in MR. The PT task also exhibited good reliability and a steep developmental trajectory. Whereas the largest gains in MR were observed between ages 7 and 8, PT performance improved almost linearly, with the largest gains from 6 to 7 and 8 to 9 years. This asynchronous developmental progression in MR and PT performance speaks to the tasks’ discriminant validity. Factor analyses revealed that all MR tasks converged on a single factor, with PT showing a weak association with this factor, indicating high construct validity. The adult computerized task loaded moderately on this factor but formed a separate (yet related) factor when a two-factor solution was enforced, again attesting to its limited suitability for measuring MR in children. Overall, the newly developed tasks provide reliable and valid instruments for measuring individual differences and developmental progress across ages 6 to 9 in single and group settings.