EDUF3040 · Psychological Perspectives in Education
Working Memory and Executive Functions
Week 7 contrasts working memory - sharply limited in capacity and duration - with the effectively unlimited long-term memory, and explains expert-novice differences through chunking. It covers Mayer's Select-Organise-Integrate model of multimedia learning and the development of executive functions (working memory, inhibitory control, cognitive flexibility) from childhood into adulthood. In University of Sydney EDUF3040 this is examined as multiple-choice items on WM limits, chunking and the three core executive functions, and short-answer questions on why working memory is the bottleneck for learning.
What this chapter covers
- 01Working memory as the 'bottleneck' for learning; definition (hold and mentally manipulate information briefly)
- 02Multi-store architecture: sensory memory → working memory (central executive + phonological loop + visuospatial sketchpad) ↔ long-term memory
- 03WM limits: ~7 ± 2 chunks for rehearsal, ~2-4 for manipulation, ~20 s without rehearsal; LTM effectively unlimited
- 04Expert-novice differences via chunking; Ericsson's S.F. (digit span ~7 → ~79); Nuthall's 3-4 exposures rule
- 05Mayer's Cognitive Theory of Multimedia Learning: dual channels; Select-Organise-Integrate (SOI)
- 06Three multimedia goals: reduce extraneous, manage essential (segment/pre-train), foster generative processing
- 07Brain training and the transfer problem (near vs far transfer)
- 08Three core executive functions in developmental order: working memory → inhibitory control → cognitive flexibility; PFC development
Why the expert recalls the board but the novice cannot
- +1State the working-memory limit. Working memory holds only about 7 ± 2 chunks for simple rehearsal (fewer, ~2-4, when information must be manipulated), so 25 independent pieces far exceed a novice's capacity.
- +1Introduce chunking. A chunk is a group of elements bound into a single unit by a schema in long-term memory, so 'one chunk' can carry many elements without adding to the working-memory count.
- +1Apply it to the realistic board. The expert recognises meaningful configurations (openings, common structures) built from thousands of hours of practice, so 25 pieces collapse into a handful of familiar chunks that fit within working memory.
- +1Explain the random board. A random arrangement matches no stored schema, so there is nothing to chunk; the expert is thrown back on raw working-memory capacity and performs no better than the novice.
- +1Instructional implication. Expertise bypasses the working-memory limit only within a domain and only for meaningful material (cf. Ericsson's S.F., who raised his digit span from ~7 to ~79 by recoding digits as running times). Teaching should build and automate schemas so learners can chunk, freeing working memory for higher-order work.
Key terms
- Working memory
- The system that holds and mentally manipulates information over short periods, comprising a central executive plus a phonological loop ('mind's ear') and a visuospatial sketchpad ('mind's eye'). It is sharply limited - about 7 ± 2 chunks for simple rehearsal, ~2-4 for complex operations, and about 20 seconds without rehearsal - making it the bottleneck for learning.
- Chunking
- Binding several elements into a single unit via a schema in long-term memory, so working memory treats many elements as 'one chunk'. Expertise lets experts chunk meaningful domain material and so effectively bypass the working-memory limit (Ericsson's S.F. reached a digit span of ~79 by recoding digits) - but only for material that matches a stored schema.
- Cognitive Theory of Multimedia Learning (Mayer)
- Learning uses two limited-capacity channels (verbal/auditory and pictorial/visual). Meaningful learning requires Selecting relevant words and images, Organising them into verbal and pictorial models, and Integrating those with each other and with prior knowledge (the SOI model). Instruction should reduce extraneous processing, manage essential processing (segmenting, pre-training) and foster generative processing.
- Executive functions
- Neurocognitive skills for the conscious, top-down control of thought, action and emotion (the brain's 'air-traffic control'). The three core skills develop in order: working memory (holding and using information), then inhibitory control (thoughtful rather than automatic responses; managing distraction), then cognitive flexibility (switching perspectives and refocusing). They underpin self-regulated learning.
- Prefrontal-cortex development
- The neural basis of executive-function growth. Childhood brings rapid synaptogenesis and slower, unrefined processing; adolescence brings refinement and increased myelination (with a 'developmental mismatch' between reward and still-maturing control systems); early adulthood brings strengthened PFC connectivity supporting planning, abstract reasoning and emotion regulation.
- Transfer problem (brain training)
- Working-memory training reliably improves the trained tasks, less so closely related tasks, and little for distant tasks or everyday performance (near vs far transfer). Reviews (e.g. Simons et al.) find limited durable academic gains, though training may help specific populations such as those with ADHD - a caution against 'brain-training' claims.
Working Memory and Executive Functions FAQ
Why is working memory called the bottleneck for learning?
Because everything new must pass through a system that holds only a handful of chunks for a few seconds. If a task requires holding more interacting elements than working memory can manage, the learner overloads and no schema is built. Long-term memory, by contrast, is effectively unlimited, so the goal of instruction is to move knowledge into long-term memory as automated schemas that can then be chunked - which is exactly why cognitive load theory (Week 8) matters.
How do experts seem to beat the working-memory limit?
Through chunking. An expert's long-term memory holds rich schemas, so many elements of a familiar, meaningful pattern collapse into a single chunk that occupies little working-memory space. This only works within their domain and only for material that matches a schema - shown by the classic result that a chess expert's advantage vanishes on a randomly arranged board, and by Ericsson's S.F., who raised his digit span dramatically by recoding digits as running times.
How is Week 7 assessed?
Expect multiple-choice items on working-memory capacity and duration, the multi-store model, chunking, the SOI model, and the three core executive functions and their developmental order, plus short-answer questions on why working memory limits learning or how multimedia should be designed. Confirm coverage on Canvas.
Can AI help me with working memory and executive functions?
Yes. Sia can quiz you on WM limits and the multi-store model, walk through Mayer's SOI process, and explain the three executive functions in developmental order, checking each step. It mirrors how EDUF3040 teaches this material and does not do graded work for you; University of Sydney academic-integrity rules apply.
Exam move
Memorise the working-memory limits (about 7 ± 2 chunks, ~2-4 for manipulation, ~20 seconds without rehearsal) and the multi-store diagram, because these are frequent MCQs. Practise explaining the expert-novice chunking result in a way that makes the random-board detail do the work, since it proves chunking is schema-dependent. Learn Mayer's SOI model and the three multimedia design goals as a set. For executive functions, lock in the developmental order - working memory, then inhibitory control, then cognitive flexibility - with the prefrontal-cortex timeline. Connect the week forward: expertise lowers the intrinsic load you meet in Week 8. When capacity numbers or the SOI steps slip, ask Sia to test you and set fresh scenarios.
Working through Working Memory and Executive Functions in EDUF3040? Sia is AskSia’s AI Educational Psychology tutor — ask any EDUF3040 Working Memory and Executive Functions question and get a clear, step-by-step explanation grounded in how EDUF3040 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
