# Domain-general and domain-specific cognitive factors mediating the relationship between math anxiety and mathematical performance in primary school children

## TL;DR

This study explores how math anxiety affects math performance in children through general and specific cognitive factors, identifying working memory and symbolic skills as key mediators.

## Contribution

The study introduces a multiple mediation model to distinguish the roles of domain-general and domain-specific cognitive factors in math anxiety's impact on performance.

## Key findings

- Working memory mediates the effects of math anxiety on word problem-solving and calculation.
- Symbolic skills mediate the link between math anxiety and calculation and math fluency.
- Non-symbolic skills do not mediate any mathematical outcomes related to math anxiety.

## Abstract

Math anxiety (MA) has been widely recognized as a barrier to mathematical achievement, yet the cognitive mechanisms underlying its effects on different mathematical outcomes remain poorly understood. This study applies a multiple mediation model to examine how MA affects three key domains of mathematical performance – word problem-solving, calculation, and math fluency – through both domain-general (working memory, inhibition, cognitive flexibility, selective attention) and domain-specific (transcoding, ordering, symbolic and non-symbolic comparison) cognitive factors. A sample of 472 French third-graders (aged 7.7–9.4 years) was analyzed using structural equation modeling (SEM) to identify cognitive mediators of the relationship between MA and mathematical outcomes, while controlling for sex, age, nonverbal IQ, socioeconomic status, and general anxiety. Results revealed that among domain-general factors, only working memory mediated the effects of MA on both word problem-solving and calculation. Among domain-specific factors, symbolic skills significantly mediated the link between MA and both calculation and math fluency abilities, whereas non-symbolic skills did not play a mediating role for any mathematical outcomes. Together, these findings provide novel insights into the interplay between domain-general and domain-specific processes in contributing to the effects of MA on mathematical performance, highlighting potential targets for educational interventions during the early years of formal schooling.

The online version contains supplementary material available at 10.1038/s41598-025-30898-2.