Glossary
Working memory load
What is working memory load?
Working memory load is the demand a task places on holding and manipulating information. How much can be handled depends on the task, how information is organized, and how capacity is measured.
The four-chunk estimate
Cowan's 2001 account argues for capacity averaging about four chunks under conditions that limit rehearsal and grouping through long-term knowledge. A chunk can contain more than one familiar item. Four is not a universal ceiling for every task.
Example: making task status visible
Imagine a clinical team trying to remember each patient's current condition, which medication information still needs checking, which tests are pending, and who will review each result. The load comes from retaining and updating those details, not simply from the number of patients.
A hypothetical dashboard could display a row for each pending test, its status, and the person responsible for follow-up. Staff could consult that record instead of remembering every open task unaided. The design still needs testing: an outdated status or unclear label could create a different problem. Making information visible does not by itself demonstrate fewer errors.
Chunking versus external memory aids
As an illustration, a reader familiar with the organizations might group the letters “F B I C I A” into “FBI” and “CIA.” Prior knowledge supplies two familiar units where another reader may see six unrelated letters. This illustrates why an item count is not automatically a chunk count.
Chunking organizes information within memory. An external aid instead leaves information available to look up—for example, a visible checklist of completed and pending tasks. Both address memory demands in different ways, without establishing a universal four-item design rule.
Why it matters
Look for information people must keep mentally available and ask whether the task needs that demand. Reducing unnecessary memory work can be a useful design aim.