Skip to main content
Fig. 2 | Cognitive Research: Principles and Implications

Fig. 2

From: When it all falls down: the relationship between intuitive physics and spatial cognition

Fig. 2

Spatial reasoning and working memory tests. a Example item from the Paper Folding Test (Ekstrom, Dermen, & Harman, 1976). The three items on the left indicate how a square piece of paper is folded and hole-punched. Participants must determine where the holes would be located if the paper was unfolded back to its original form. b Example question from the Mental Rotation Test (Peters et al., 1995; Shepard & Metzler, 1971; Vandenberg & Kuse, 1978). Participants must determine which two items on the right are rotated versions of the item on the left. The remaining two incorrect choices are structurally different to the target item. c General layout of the spatial working memory task. Participants were required to remember a series of filled in grid positions while performing a secondary symmetry judgement (see “Methods”). The symmetry judgement required participants to determine whether the shape was perfectly symmetrical when split vertically down the middle. The number of grid positions that participants were required to hold in memory varied between three and eight on different trials. At the end of each trial, participants reported the remembered grid positions by clicking within a blank grid (far right panel). d General layout of the verbal working memory task. Participants were required to remember a series of letters while performing a secondary lexical decision task (see “Methods”). In the lexical decision task, participants reported whether a word was a real or fake English word. The number of letters that participants were required to hold in memory varied between three and eight on different trials. At the end of each trial, participants reported the remembered series of letters by clicking within a grid (far right panel). e Split-half reliability for each task. The trials from each task were split into two sets and performance was computed for each independent half (each data point represents the performance of one participant). All split-half correlations were statistically significant (p < .001)

Back to article page