Measuring Spatial Abilities in Children: A Comparison of Mental-Rotation and Perspective-Taking Tasks

Abstract

Mental rotation (MR) and perspective taking (PT) are important spatial abilities and predictive of performance in other cognitive domains. Yet, age-appropriate measures to assess these spatial abilities in children are still rare. This study examined psychometric properties of four MR tasks in 6- to 9-year-olds (N = 96). Two were developed specifically for children and two were based on established assessments for adults; one of each was a computerized and one was a paper–pencil task. Furthermore, spatial perspective taking (PT)—a different but closely related ability—was assessed to determine discriminant validity. Factor analyses showed that all MR tasks loaded on one single factor, with PT only loading weakly on the same factor, suggesting high construct validity. The computerized task for adults showed moderate factor loadings, constituted its own (but correlated) factor when a two-factor solution was forced, and showed the lowest reliabilities, suggesting that it was very difficult for children. On average, the new MR tasks had good to excellent reliabilities, differentiated well between age groups, and proved to be well-suited to assess MR in this age range. The PT task also showed good reliability and a steep developmental progression. Relations to verbal skills, gaming experience, and TV consumption are discussed.