Fig. 4
From: When it all falls down: the relationship between intuitive physics and spatial cognition
Reliable variance in the Unstable Towers task after regressing out the contribution of spatial abilities and working memory. We ran a linear regression to find the degree to which a combination of the spatial and working memory measures could account for individual differences in performance on the towers task (see “Methods”). We then found the residual (unexplained) variance in the towers task and plotted independent split halves of the data (each data point represents one participant). We identified robust positive correlation in this split-half analysis (r (98) = 0.48, p < .001), indicating that reliable individual differences in physical prediction cannot be fully accounted for by spatial abilities or working memory capacity