“Test-retest reliability of the implicit and explicit SNARC effect”

The Spatial–Numerical Association of Response Codes (SNARC) effect reflects a systematic association between numerical magnitude and spatial response codes, such that small numbers are typically associated with the left side of space and large numbers with the right. This phenomenon is robust and highly replicable at the group level; however, its measurement at the individual level often shows limited temporal stability, an issue that has been discussed within the framework of the reliability paradox in cognitive psychology. The present study investigated the test–retest reliability and the inter- and intra-individual variability of the SNARC effect by comparing two widely used task versions: an implicit parity-judgment task, in which magnitude processing is not required, and an explicit magnitude-classification task, in which numerical magnitude is directly relevant to the response. The experiment consisted of two separate sessions with a high number of trials. One group completed the explicit task in both sessions, whereas the other group completed the implicit task in both sessions, and individual SNARC effects were estimated using regression-based indices. By directly comparing implicit and explicit paradigms under repeated-measures conditions, this study aimed to clarify the temporal stability of spatial–numerical associations and to evaluate the suitability of different SNARC tasks for research on individual differences. Results showed overall low test–retest reliability for the SNARC effect. Reliability anyway was higher in the explicit task than in the implicit one, likely because the explicit task creates more differences between people. These findings highlight the challenges of measuring stable individual differences in spatial–numerical associations and suggest that explicit tasks may be somewhat more suitable for this purpose.