The "Tetris effect": autistic and non-autistic people share an implicit drive for perceptual cohesion.

Abstract

Background: When working on jigsaw puzzles, we mentally "combine" two pieces to form a composite image even before physically fitting them together. This happens when the separate pieces could logically create a cohesive picture and not when they are mismatched or incoherent. The capacity of the brain to combine individual elements to form possible wholes serves as the basis of perceptual organisation. This drive for perceptual cohesion-the "Tetris effect"-can be seen in the famous game, where people automatically perceive logical combinations from separate pieces. However, it is unclear how this presents in populations known to have perceptual differences, such as autistic people. Theories on the inclination to process local over global details in autism and autistic strengths in pattern recognition lead to conflicting predictions regarding the drive for perceptual cohesion in autistic compared to non-autistic people.

Methods: In this large-scale (n = 470) pre-registered online behavioural study, we aimed to replicate previous research conducted on neurotypical participants and to extend this work to autistic participants. We used two tasks with Tetris-style stimuli to examine how autistic (n = 196) and non-autistic (n = 274) adults implicitly perceive possible wholes from individual parts. Data were analysed using logistic mixed-effects regression models and hierarchical Signal Detection Theory modelling.

Results: Overall, we replicated the results from the original study in finding participants are more likely to perceive parts as wholes when there is the potential to form a whole, compared to when there is not. However, we found no differences between autistic and non-autistic participants across both tasks.

Limitations: Although power calculations were carried out to assess sample sizes needed to detect a group difference, given the small effect size (Cohen's d = 0.37) in the original study, it may be that any existing group differences are still undetectable with the current sample size.

Conclusions: We conclude that the "Tetris effect" is ubiquitous and seen in both neurotypical and neurodiverse populations. Our findings challenge the deficit-focussed narrative often seen in the autism literature and highlight the similarities in task performance between autistic and non-autistic participants.