6.2 Capacity and Limitations

Working Memory Capacity

Working memory can only hold and manipulate a small amount of information at any given time. Understanding these capacity limits is central to cognitive psychology because they shape how we think, learn, solve problems, and make decisions.

Limitations of Working Memory

George Miller's classic 1956 paper proposed that short-term memory holds about 7 ± 2 items, a finding often called "Miller's Magic Number." More recent research, particularly by Nelson Cowan, has revised that estimate downward. Most cognitive psychologists now place the true capacity of working memory at roughly 3 to 5 chunks of information. That's a meaningful difference, and it underscores just how narrow the bottleneck really is.

This bottleneck restricts how many cognitive operations you can run at the same time. When you try to multitask, you're bumping up against this limit. It also constrains complex decision-making and problem-solving, because you can only hold so much relevant information in mind at once.

Cognitive load theory breaks down the demands on working memory into three types:

1. Intrinsic load comes from the complexity of the material itself. Solving a calculus problem has higher intrinsic load than basic addition.
2. Extraneous load comes from how the material is presented. Poorly designed instructions or cluttered visuals add unnecessary burden.
3. Germane load is the productive effort you spend building schemas and integrating new knowledge. This is the "good" load that actually supports learning.

Two common strategies help you work within these limits:

• Rehearsal: Repeating information to keep it active in working memory. Think of silently repeating a phone number while you walk across the room to write it down.
• Elaboration: Connecting new information to something you already know. If you're learning the Spanish word biblioteca (library), linking it to the English word Bible (a book) gives the new word a meaningful hook.

Factors Affecting Memory Capacity

Working memory capacity isn't fixed for everyone or across your entire lifespan. Several factors shift it up or down.

Age plays a clear role. During childhood, capacity and processing speed improve steadily as the brain develops. In older adulthood, working memory function gradually declines, partly due to slower processing and reduced attentional control.

Expertise can dramatically expand capacity within a specific domain. Chess masters, for example, can glance at a mid-game board and recall the positions of most pieces, not because they have "better" working memory overall, but because they recognize meaningful patterns and chunk the positions into familiar configurations. Musicians show a similar advantage for remembering longer musical sequences.

Automaticity also matters. When a skill becomes automatic through practice (like reading words or driving a familiar route), it demands less working memory, freeing up resources for other tasks.

Individual differences account for a lot of variation:

• Genetic factors contribute to your baseline capacity
• Neural efficiency varies from person to person
• Attention control abilities affect how well you manage and filter information

Environmental factors modulate performance too:

• Stress and anxiety consume working memory resources, reducing effective capacity
• Poor sleep impairs both memory consolidation and next-day recall
• Nutrition and physical exercise support cognitive function. Research links omega-3 fatty acids and regular aerobic activity to better working memory performance.

Chunking for Memory Extension

Chunking is the process of grouping individual pieces of information into larger, meaningful units. It's one of the most effective ways to stretch working memory capacity without actually increasing the number of slots available.

Consider a 10-digit phone number: 1234567890. That's 10 separate items, well beyond the 3-to-5 chunk limit. But formatted as 123-456-7890, it becomes three chunks, which fits comfortably. You haven't changed the information, just how it's organized.

Chunking works because it relies on pattern recognition and meaningful associations. The more knowledge you have in a domain, the larger and more sophisticated your chunks can be.

Everyday examples:

• Grouping grocery list items by category (produce, dairy, meats) instead of memorizing a random sequence
• Acronyms like ROY G BIV compress seven rainbow colors into a single memorable unit
• Hierarchical organization structures complex information, like how biological taxonomy nests species within genera within families

Memory Capacity and Cognitive Performance

Working memory capacity is one of the strongest predictors of higher-order cognitive abilities. Research consistently shows correlations between working memory capacity and:

• Fluid intelligence: the ability to reason through novel problems
• Reading comprehension: processing and integrating meaning across sentences and paragraphs
• Mathematical problem-solving: holding intermediate results in mind during mental calculations

These correlations extend to real-world outcomes. Working memory capacity predicts performance on standardized tests (SAT, GRE) and success in learning new complex skills like programming languages or musical instruments.

Working memory also underpins key executive functions:

• Task switching: shifting between different activities without losing your place
• Inhibition: filtering out irrelevant information to maintain focus
• Updating: revising mental representations as new information comes in

Strategies to optimize performance:

• Break complex tasks into smaller components to reduce cognitive load
• Use external memory aids like note-taking or digital reminders to offload information
• Minimize distractions and interruptions to preserve working memory resources

A note on working memory training: Programs using tasks like the dual n-back (where you track sequences of stimuli along two dimensions simultaneously) have shown some promise for improving working memory performance on the trained task. However, whether those gains transfer to other cognitive domains remains an active and unresolved debate in the field.