Working Memory Capacity — Why People Differ and What Affects It

Written by TOMOY · Published: April 8, 2026 · Updated: April 8, 2026

1. What Is Working Memory Capacity?

Working memory capacity refers to the amount of information a person can actively hold and process at a given moment. It is not simply about how much can be stored — passive storage is the domain of long-term memory. What working memory capacity captures is the active workspace: how many items can be kept accessible, manipulable, and protected from interference simultaneously.

Capacity is not fixed in the way that, say, the size of a hard drive is fixed. It varies across individuals, fluctuates with state (fatigue, stress, arousal), and is shaped by experience and strategy. Understanding what determines this capacity is one of the central questions of working memory research.

2. Miller's 7 ± 2 and What It Actually Means

The most well-known claim about working memory capacity comes from a paper by George Miller, who proposed that immediate memory holds approximately 7 items, plus or minus 2. This figure, often called "the magical number seven," became widely cited as a universal limit on how much the mind can hold.

Miller's insight, however, was about chunks rather than raw items. A chunk is any unit that the brain treats as a single, familiar entity. A single digit is a chunk. But so is a familiar word, a well-known acronym, or a pattern that has been deeply learned. Crucially, the 7 ± 2 figure applies to chunks regardless of their complexity — which means that people who can form larger chunks from the same material can effectively hold more information with no change in the underlying limit.

This is why an expert chess player can glance at a board position and recall it far more accurately than a novice: the expert perceives the board in terms of meaningful configurations (chunks), while the novice sees a collection of individual pieces. The capacity limit is the same; the size of the units filling it is not.

3. Cowan's Revision: Around Four Chunks

Later research, particularly by Nelson Cowan, proposed a downward revision of the capacity estimate. Cowan argued that when chunking is controlled for — when the items presented are genuinely unfamiliar and cannot be grouped by prior knowledge — the capacity of the focus of attention in working memory is closer to four chunks on average.

This does not contradict Miller so much as refine the picture. Miller's 7 ± 2 was largely based on tasks where participants could use chunking strategies. Cowan's 4 reflects a more controlled estimate of the attentional bottleneck itself — the core capacity of the system before strategies are applied.

Both figures describe a genuine constraint. Working memory is a severely limited resource, and most people reach its ceiling quickly when material is unfamiliar, complex, or arrives faster than it can be organized.

4. Individual Differences in Capacity

Working memory capacity varies substantially across individuals, and these differences are considered stable enough to be measured reliably with standardized tasks. Several sources of variation have been identified in the research literature:

• Age: Capacity tends to be lower in young children and declines again in older adults. Developmental trajectories suggest a peak in early adulthood.
• Attentional control: The ability to suppress irrelevant information and maintain focus on task-relevant items is thought to be a significant contributor to individual differences in measured capacity.
• Processing speed: Faster information processing may allow more items to be refreshed before they decay, supporting higher effective capacity under time pressure.
• Domain expertise: Familiarity with a domain allows richer chunking, which translates into higher apparent capacity on domain-specific tasks.

5. Factors That Temporarily Reduce Functional Capacity

Beyond stable individual differences, a number of transient factors are thought to reduce the effective capacity of working memory in the moment:

• Sleep deprivation: Research suggests that insufficient sleep is associated with reduced performance on working memory tasks. The mechanisms are thought to involve impaired attentional control and slower information processing.
• Stress and anxiety: High cognitive load from emotional content — worry, rumination, intrusive thoughts — is thought to occupy working memory resources that would otherwise be available for the current task.
• Dual-task demands: When attention must be divided between two ongoing tasks, the effective capacity available to each task decreases. This is a structural property of working memory, not a failure of will.
• Cognitive fatigue: Sustained cognitive effort over time is associated with reduced working memory performance, though the mechanisms underlying this effect remain an active area of research.

6. Capacity, Span, and What Tasks Measure

Working memory capacity is commonly measured through span tasks — procedures in which a person is asked to recall a sequence of items in correct order after a brief presentation. Simple span tasks (forward digit span) primarily tax the phonological loop. Complex span tasks, which interleave processing demands with memory items, are thought to provide a better measure of the central executive component — the system responsible for maintaining information under distraction.

What span scores reflect is therefore not a single, simple quantity, but the product of multiple interacting factors: the capacity of the relevant subsystem, the efficiency of rehearsal processes, the ability to resist interference, and the application of chunking and other strategies. Two people with the same digit span score may be drawing on quite different cognitive resources to achieve it.