When Math Word Problems Stall — Where Working Memory and Language Comprehension Collide

5 min read日本語版あり
Audience
Parents of children in grades 2–5 (roughly 7–11 years old)
Target length
~1,500 words
Status
Draft v1 (translated from Japanese v1)
Original
../164_math_word_problems.md

Lead

Arithmetic drills: eight out of ten correct. Word problems: total freeze. "How can you do the calculation but not solve the problem?" — that question is usually directed at the child. But word problems genuinely require more than arithmetic. They are not simply harder arithmetic. They are a different kind of task.

"Can calculate but can't solve word problems" is not a skills deficit. It is a bottleneck that emerges when multiple cognitive processes are demanded simultaneously. At the center of that bottleneck is the interaction between working memory and language comprehension.

Background

Explaining this stumbling point as "poor reading comprehension" or "poor concentration" flattens a more structured problem.

Kintsch and Greeno (1985) described word problem solving as a process of transforming a surface-level understanding of a text (the ) into a mental representation of the situation it describes (the ) [4]. To solve "Taro has 3 apples and receives 2 more from Hanako — how many does he have in total?" a child needs more than vocabulary. They need to mentally construct a scene in which apples are being added, and then translate that changing situation into an arithmetic expression. Working memory is the key to that transformation.

is the cognitive function that holds information briefly in mind while simultaneously operating on it. Reading a word problem while tracking numerical relationships and then executing a calculation requires keeping several pieces of information active at once. That simultaneous demand is where processing breaks down.

Working Memory and Word Problem Performance

Caviola et al. (2023) conducted a meta-analysis of 55 samples comprising 11,224 elementary school children and found a significant association between verbal working memory and arithmetic performance [1]. The relationship was especially pronounced in tasks — like word problems — that cross linguistic and numerical demands.

A three-level meta-analysis by Ji and Guo (2023) covering 130 studies and 43,938 participants reported a moderate correlation of r = 0.28 between working memory and mathematical problem solving [2]. This is not a deterministic claim that "low working memory equals poor word problems." It means that working memory capacity is one contributor to word problem performance, and limits on that capacity matter.

Fuchs et al. (2005) showed that calculation ability, reading comprehension, and working memory each contributed independently to math achievement [5]. When a child can compute but not solve word problems, either working memory, language comprehension, or both are likely acting as a constraint — not the arithmetic itself.

Practical Approaches

Peltier and Vannest (2017) conducted a meta-analysis of in elementary school students and reported an effect size of g = 0.61 [3]. Schema instruction teaches children to categorize word problems by type — combining, comparing, changing — and to represent the structure visually. The focus is on recognizing the underlying structure rather than drilling a particular problem format.

Approaches that work at home draw on the same logic.

Three Ways to Reduce the Cognitive Load

1. Read the problem aloud. Silent reading can concentrate more cognitive demand in working memory; externalizing the language processing — reading out loud — can reduce what needs to be held mentally and free up capacity for the mathematical reasoning.

2. Draw a picture or diagram. Visualizing the quantities and their relationships supports construction of the situation model. Having a child sketch what is happening in the problem — the apples being moved, the two groups being compared — converts words into images and makes the question clearer.

3. Identify what the question is asking before reading the whole problem. Reading the final sentence first, or asking "What does this problem want you to find?" before starting, establishes a target. When the destination is known, the reading that follows has direction.

Practical Takeaways

Treating a stumble as "needs more practice" leads to assigning more problems of the same type. From a cognitive standpoint, supporting the transformation process should come before increasing practice volume.

  1. Make "I understand the situation" the first goal. If a child can describe the problem in their own words, the arithmetic can then be addressed separately. The habit of "draw a picture, then write the equation" is worth building explicitly.

  2. Distinguish error types. A calculation error ("12 − 3 = 8") and a situation comprehension error ("misread 'more than' as 'less than'") call for different responses.

  3. Acknowledge the difficulty. Working memory load is genuinely experienced as difficulty by the child. "Where did it get confusing?" is a more useful question than "Why don't you get it?" — it invites the child to put their cognitive process into words, which itself supports learning.

Summary

Stumbling over word problems is often not a math problem or a concentration problem. It is a difficulty that arises when language comprehension and working memory are demanded simultaneously — a specific cognitive challenge, not a general weakness. Having that model in mind shifts the starting question from "try harder" to "where specifically is this getting stuck?"

The supports that help are small and concrete. Drawing a picture. Reading aloud. Naming the question before reading the whole problem. Those adjustments can change how cognitive resources are allocated — and that change is sometimes enough.

References

  1. Caviola S, Colling LJ, Mammarella IC, Szucs D. The relationship between working memory and arithmetic in primary school children: a meta-analysis. Brain Sci. 2023;13(1):22. doi:10.3390/brainsci13010022. PMID: 36672006.
  2. Ji C, Guo W. The association between working memory and mathematical problem solving: A three-level meta-analysis. Front Psychol. 2023;14:1091126. doi:10.3389/fpsyg.2023.1091126. PMID: 37057173.
  3. Peltier C, Vannest KJ. A meta-analysis of schema instruction on the mathematical problem solving of elementary school students. Rev Educ Res. 2017;87(5):899–920. doi:10.3102/0034654317720163
  4. Kintsch W, Greeno JG. Understanding and solving word arithmetic problems. Psychol Rev. 1985;92(1):109–129. doi:10.1037/0033-295X.92.1.109
  5. Fuchs LS, Fuchs D, Compton DL, et al. The prevention, identification, and cognitive determinants of math difficulty. J Educ Psychol. 2005;97(3):493–513. doi:10.1037/0022-0663.97.3.493
  6. Swanson HL, Jerman O. Math disabilities: a selective meta-analysis of the literature. Rev Educ Res. 2006;76(2):249–274. doi:10.3102/00346543076002249
  7. Fuchs LS, Geary DC, Compton DL, et al. Do different types of school mathematics development depend on different constellations of numerical versus general cognitive abilities? Dev Psychol. 2010;46(6):1731–1746. doi:10.1037/a0020662. PMID: 20873937.

Related