Visual layout is known to affect computational performance, yet its relationship to testing format and numerical proficiency remains unclear. In this preregistered study, we examined how spatial presentation format influences arithmetic reasoning. We conceptually replicated Experiment 3 of Landy and Goldstone (2010) using the Precedence and Proximity Task in a forced-choice format and additionally assessed arithmetic fluency using a speeded test. A total of 201 adults solved 200 mixed addition–multiplication expressions. Spatial congruency was manipulated by varying the spacing of addition and multiplication operations so that grouping cues either aligned with or conflicted with the order-of-operations rules; operation order was also manipulated. Participants were faster and more accurate in congruent spacing conditions, whereas incongruent spacing increased operation-order errors, particularly when expressions began with addition. A subgroup of outliers exhibited very high accuracy in congruent conditions but extremely high error rates in incongruent ones, suggesting heightened susceptibility to perceptual grouping in some individuals. The spacing effect emerged despite differences in task format and response procedure compared to the original study, supporting its robustness. Higher arithmetic fluency was associated with greater accuracy and shorter reaction times. Overall, these findings corroborate that visuospatial layout systematically biases symbolic computation and that individuals differ in their susceptibility to such effects.
