Differences in Cognitive Performance on Arithmetic Principles and Their Relationship with Mathematical Ability

Description

This study tested three main hypotheses: (1) elementary students would show differential performance across four arithmetic principles—inversion, commutativity, associativity, and distributivity—with inversion being the easiest; (2) after controlling for executive functions, domain‑specific abilities (arithmetic fluency, whole‑number arithmetic, number sense) would uniquely predict arithmetic principle understanding and mathematics achievement; and (3) students would cluster into distinct mathematical ability profiles. To examine these, 77 fourth‑graders completed a unified sentence‑verification task for the four principles, along with measures of executive functions (working memory, cognitive flexibility, inhibitory control), number sense (symbolic magnitude comparison), arithmetic fluency, whole‑number arithmetic, and a standardized mathematics achievement test. Key findings Performance across principles was not uniform: inversion yielded the highest accuracy and fastest responses; commutativity, associativity, and distributivity showed similar difficulty, all exceeding baseline control performance. After statistically controlling for executive functions, arithmetic fluency, whole‑number arithmetic, and arithmetic principle understanding each remained significantly associated with mathematics achievement. In contrast, number sense (numerical distance effect) was not significantly related to either principle understanding or overall achievement. Latent profile analysis revealed three distinct ability groups—high, moderate, and low—indicating substantial heterogeneity in students’ integrated mathematical competence. Interpretation The graded difficulty suggests inversion is cognitively most accessible, likely because its cancellation structure imposes low processing demands, whereas other principles require more complex relational reasoning. The persistent link between arithmetic fluency and principle understanding supports the view that procedural efficiency enables conceptual abstraction. The lack of a number‑sense correlation may reflect that basic magnitude comparison tasks lose predictive power by fourth grade, when higher‑order conceptual skills become more critical. The three identified profiles underscore that arithmetic principle understanding is embedded in broader mathematical competence, calling for differentiated instruction tailored to students’ specific ability patterns. Overall, the findings highlight arithmetic principle understanding as a key bridge between procedural fluency and broader mathematical achievement, with clear implications for sequenced, strategy‑oriented mathematics teaching.

Steps to reproduce

Participants: 77 fourth‑graders (Shenzhen, China; mean age 9.19 years). Written consent from parents and teachers. Procedure (fixed order): Executive functions (computerised) Number sense (computerised) Arithmetic fluency (computerised) Whole‑number arithmetic (paper‑pencil) Arithmetic principles (computerised) Mathematics achievement (standardised test) Tasks & scoring: Working memory – adaptive n‑back‑like: reproduce movement sequence on 4×4 grid. Score = longest correct sequence length. Cognitive flexibility – computerised Wisconsin Card Sorting Test (colour/shape/number rules). Score = completed categories. Inhibitory control – Stroop colour‑word (Chinese). Score = RT difference (incongruent–congruent). Number sense – symbolic magnitude comparison (1‑9 digits, large/small distance). Score = distance effect slope. Arithmetic fluency – 64 equations (+,−,×,÷), 60s verification. Score = correct answers. Whole‑number arithmetic – 30 curriculum‑based items, 40 min. Score = total correct. Arithmetic principles – sentence verification (40 equations: inversion, commutativity, associativity, distributivity, control). 10s limit, true/false response. Scores = accuracy & RT for correct trials. Mathematics achievement – 28‑item standardised exam, 90 min. Software & reproducibility: Power analysis: G*Power (N=65 minimum). Analysis: repeated‑measures ANOVA, partial correlations (controlling for executive functions), latent profile analysis. Key controls: fixed task order (principles last), no feedback on principles task, timeout handling, counterbalanced true/false and left/right.

Institutions

• Capital Normal University

  Beijing, Beijing

Categories

Licence