
Our Working Memory and Math
Our Working Memory and Math
October 26th, 2026
By Tomasz Arkusz, Instructional Technology Analyst
Engineering requires us to reason with mathematical ideas that often don’t sound easy. Consider, for example, the epsilon-delta definition of continuity:
Given a function and an element of the domain , is said to be continuous at the point when the following holds: For any positive real number , there exists some positive real number such that for all in the domain of with , the value of satisfies .
This is a fundamental concept in mathematical analysis. Did you understand it? Could you recreate the definition in your own words? If you use analysis in your daily work, you will have no difficulty. If you are a bit rusty, however, it may take some thinking.
I remember a professor telling us in college, many years ago, that to understand this definition we had to memorize it. I was incensed. What was she talking about? I could understand it perfectly well. Why should I memorize it?
Today I look at her advice differently. Memorization can reduce the mental effort required to simultaneously read and think. If you know the definition by heart, you do not have to devote attention to converting letters into words and words into sentences. Nor do you have to struggle to remember the first part of the definition while reading the next. The whole definition is already in your head, leaving you free to think about what it means.
Working memory — our ability to temporarily hold and manipulate information — has limited capacity (Baddeley, 2007). We should therefore look for ways of presenting new ideas that require as little incidental cognitive effort as possible, leaving more working memory available for thinking about the ideas themselves.
Memorizing mathematical text is one way of making an idea easier to comprehend. Unfortunately, you can memorize only so much. Another way is to make the process of reading automatic. Most fluent readers not only decode text, but also simultaneously comprehend what they are reading (Kuhn et al., 2010).
While students with less-than-fully-automated reading skills have always been around, the question today is how much the overall balance of skills students bring with them has changed. Reading books for leisure now competes with computer games and interactive media. Students arriving at engineering school may have more practice with dynamic interactive environments than with sustained reading. If so, by relying too heavily on text-based delivery of instruction – books, slides, the web – we may be depending on cognitive skills students have practiced less while underusing skills that students have well developed already.
So, what do we do?
One possible direction is related to “game-based learning”, although the term may be misleading here. The objective is not to build another educational game. It is to borrow from games forms of visual interaction that draw on skills students have well developed.
Tools such as MATLAB and Mathematica already allow students to manipulate mathematical objects and plot interactively. But an interactive tool is not the same as an interactive learning environment. The next step is to design the interaction around the mathematical concepts we want the student to understand. A student learning continuity, for example, could manipulate an -band around for various functions and explore how the allowable neighborhood around changes. Once students have developed an intuition for the idea, precise mathematical language can follow, formalizing and refining that intuition.
Developing such an approach to engineering mathematics would be a major undertaking. But the potential payoff is substantial: students could spend more of their cognitive effort building understanding of mathematical ideas and less on deciphering presentation of those ideas.
References
Baddeley, A. D. (2007). Working memory, thought, and action. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198528012.001.0001
Kuhn, M. R., Schwanenflugel, P. J., & Meisinger, E. B. (2010). Aligning theory and assessment of reading fluency: Automaticity, prosody, and definitions of fluency. Reading Research Quarterly, 45(2), 230–251. https://doi.org/10.1598/RRQ.45.2.4
