Enhanced Mental Rotation Abilities in Autistic Individuals

Background: Autistic strengths in low-level perception and visuospatial skills are being increasingly recognized. For example, it is well established that autistics excel at the Block Design task. Visual imagery, the formation and manipulation of mental images, represents a key ability for successfully solving visuospatial problems like the Block Design task.  Mental rotation consists in deciding whether two shapes are identical or different (mirror-reversed) after mentally rotating one shape to the position of the other shape. Autistic individuals are reported to be faster than nonautistic individuals at mental rotation involving three-dimensional geometric shapes (Falter et al., 2008), and to exhibit less task-related activity in frontal cortex (Silk et al., 2006).

Objectives: This study first explored mental rotation with different task material types to test the robustness and scope of the reported group differences in visual imagery. Second, it aimed to characterize the mechanisms responsible for these differences by asking: a) Is the difference linked to visual imagery or more generally to the processing speed of perceptual information?, and b) Does the difference lie in the capacity to form a mental representation of the objects or to “spatially manipulate” that representation?

Methods: Participants were 14 autistic and 14 nonautistic individuals, matched on age (mean 21 years), sex, manual preference and Wechsler IQ (mean 104). We used four mental rotation same-different tasks, including images of two-dimensional geometric figures, three-dimensional geometric figures, drawings of hands and letters. We investigated the second aim by comparing performance in mental rotation with: a) speed of perceptual processing and b) visual imagery without manipulation of the mental representation. Speed of perceptual processing was measured using a classic inspection time task, in which participants indicated which of two very briefly presented vertical lines was longer. A staircase procedure was used to estimate the minimal exposure time for each participant. For the visual imagery task, participants imagined a specified letter in a circle and then decided which of two highlighted portions of the circle would contain the greater proportion of the letter.

Results: Mixed effects linear models were used for the four mental rotation tasks. The analysis showed a significant between-group difference in accuracy intercept for hand stimuli, and in both accuracy and response time for letters. In both cases, autistic participants had faster or more accurate performance. There were no significant between-group differences in the mental rotation of 2D and 3D figures, although trends were seen in the same direction. The visual imagery and inspection time tasks yielded no significant between-group differences.

Conclusions: Subtle but consistent increased performance in mental rotation with various materials was found in autistic individuals. As the processing speed of perceptual information and ability to form mental representations did not differ between groups, the different performance in mental rotation is therefore likely to arise from a differing ability to spatially manipulate the mental representations of objects. This ability could be beneficial in various complex cognitive tasks such as matrix reasoning or drawing and interpreting plans, and might be exploited in educative methods with autistic individuals.