Poster #27 - Development of a Negative Priming effect in a non-symbolic numerical comparison task

• In Event: 3-159 - Poster Session 12
  In Poster Session: PS 12 Section - Cognitive Process

Integrative Statement

Negative priming paradigms consist in investigating the deleterious impact of a prime item on the subsequent probe item, when the latter requires to activate a perceptual feature which was interfering with the correct response in the former. Negative priming effect thus provides evidence for the role of inhibitory control processes in the task at hand. Recent research on numerical estimation points to a crucial role of inhibitory control processes, in particular to extract the numerical aspects of non-symbolic representations of number from other continuous non-numerical dimensions of magnitude, such as item size or total surface area. So far, studies have investigated the implication of inhibitory control processes in the classic non-symbolic numerical task by contrasting performance in incongruent trials (i.e., non-numerical dimensions of magnitude interfere with numerosity) to performance in congruent (i.e., non-numerical dimensions of magnitude are congruent with numerosity) ones. However, a few studies failed to observe such congruency effect. The negative priming paradigm could thus provide a new method for investigating the role of inhibitory control processes in numerical estimation. Forty-seven primary school children (23 girls and 24 boys, mean age = 7.92 years, SD = .87) and thirty-two adults (21 females and 11 males, mean age = 27.86 years, SD = 6.13) performed a non-symbolic comparison task adapted to a negative priming paradigm. Probe items all consisted in number/Size congruent pairs of arrays of dots, while prime items could either be number/size incongruent (test condition) or with equal dot size in both arrays (control condition). We used the Size dimension as defined by (De Wind et al., 2015), which allowed us to control for the Spacing dimension across all the experimental trials. Primes were in a 1:2 numerical ratio, so as to ensure high performance (i.e. successful handling of the interference), whereas probes were in 4:5 and 5:6 numerical ratios (12 v. 15 and 10 v. 12 dots). We observed a negative priming effect both in school-age children, t(42) = 4.91, p < .0001, d = 1.04, and in adults, t(29) = 3.4, p = .002, d = 0.75, as demonstrated by lower performances on number/size congruent items when preceded by incongruent items than when preceded by neutral items (controlling for size). Additionally, the amplitude of the negative priming effect was larger in children than in adults, t(71) = 3.63, p = .001, d = 0.873. Finally, the amplitude of the negative priming effect was correlated with the congruency effect in a symbolic numerical Stroop task in the adult group, r(26) = .42, p = .03, confirming that negative priming effects reflect inhibitory control processes. Our results show that the inhibition of the non-numerical dimensions of magnitude impacts performance in numerical estimation throughout development. These results provide converging evidence for the role of inhibition in numerical estimation. Our finding has educational implications for the investigation of the predictive values of basic numerical representations and executive functions for general math achievement.

Authors

• Arnaud Viarouge, Paris Descartes University
  Presenting Author

• Olivier Houdé, University Paris Descartes
  Non-Presenting Author

• Grégoire Borst, Paris Descartes University
  Non-Presenting Author