Working Memory

In short

Working memory is the cognitive system that holds information active for a few seconds while manipulating it. Its capacity is severely limited — to roughly four units — making it the bottleneck of every learning process.

What is working memory?

Working memory is the part of memory in which information is held briefly and, crucially, worked on at the same time — as when you do mental arithmetic, parse a sentence, or weigh up an argument. It differs from plain short-term memory in that it does not merely store content but manipulates it. The most influential account comes from Alan Baddeley and Graham Hitch (1974).

What are its components?

In the Baddeley–Hitch model, a central executive directs attention and controls two subordinate systems: the phonological loop for verbal and acoustic information, and the visuospatial sketchpad for images and spatial layout. Baddeley later added the episodic buffer, which binds information from the subsystems and from long-term memory into a single coherent episode.

How much fits in it?

In 1956 George Miller proposed the famous magical number seven, plus or minus two. More recent work, notably by Nelson Cowan (2001), points to a smaller capacity of about four independent units (chunks) once inner rehearsal is prevented. What matters is not the number of raw stimuli but how well they are grouped into larger, meaningful units — which is why chunking expands effective capacity.

Why is it so important for learning?

Because new knowledge must pass through the working-memory bottleneck before it can be stored durably in long-term memory, its limited capacity caps how much you can take in at once. Overload it (see cognitive load) and comprehension collapses. Strategies such as chunking, activating prior knowledge, and removing distractions deliberately ease the burden. As prior knowledge grows, load falls, because long-term memory supplies ready-made schemas that compress many details into a single unit.

Can working memory be trained?

Commercial training programmes promise to enlarge working-memory capacity for good. Research is cautious here: people do become measurably better at the trained task, but that gain transfers poorly to other abilities such as general intelligence or school performance. More effective than training raw capacity is to offload the limited system cleverly — through solid prior knowledge, clear structuring of the material, and external aids such as notes or sketches that hand off part of the processing.