Prevalence of Spatial-Numerical Associations: Psychometric Approach In numerical cognition, one of the most replicated group-level phenomena is the SNARC effect. It reflects the observation that Western participants respond faster to small/large numbers with the left/right side respectively. When the regression slope between number magnitude and right-left RT difference is taken as an index of the effect, c.a. 75% of participants exhibit negative slopes and are therefore considered as revealing the effect. Individual slopes are, however, only estimates of true slope parameters with measurement error, and only if the task was perfectly reliable, they would reflect true effects. In psychometrics, confidence intervals (CIs) are built around observed scores to account for non-perfect reliability. In case of the SNARC effect, individuals could be considered as revealing the effect, if the CI around the observed negative slope does not contain zero. We calculated CIs around individual SNARC slopes from 18 different and uniformly-analyzed datasets using the traditional psychometric approach: measurement reliability and sample variance (SD) were used to determine the standard error of measurement (SEM), based upon which CIs were determined. When 90% CIs were considered, only 37% of participants revealed the consistent effect. Consequently, with only 3% revealing a consistent reverse mapping, 60% of individuals did not feature a consistent SNARC effect. This suggests that despite its group-level robustness, the SNARC effect only occurs in a minority of people. The SNARC effect might thus not be a good proxy of universal spatial-numerical associations (SNAs). Alternatively, SNAs might be present in only a part of the population. In this case, theories of numerical cognition should be reconsidered. To disentangle these alternatives, future studies should determine the prevalence of other SNAs by applying the current method. Carrie MAZZI-GEORGES, PhD UNIVERSITY OF LUXEMBOURG Faculty of Language and Literature, Humanities, Arts and Education