The working memory model (Baddeley and Hitch, 1974)

The components

• Central executive: the controller. It directs attention, decides what the other parts do and handles tasks such as mental arithmetic and problem solving. Limited capacity; no store of its own.
• Phonological loop: holds speech-based information. It has a phonological store (the 'inner ear', holding sounds for about 1 to 2 seconds) and an articulatory control process (the 'inner voice', which rehearses and turns written words into sounds).
• Visuo-spatial sketchpad: holds visual and spatial information (the 'inner eye'). Logie (1995) split it into a visual cache (form and colour) and an inner scribe (spatial and movement information).
• Episodic buffer: added by Baddeley in 2000. A limited-capacity store (about four chunks) that combines information from the other parts and from LTM into single episodes.

Evidence: dual-task studies

If two tasks use the same component they interfere and performance drops; if they use different components they can be done together. Repeating 'the the the' aloud (articulatory suppression) disrupts reading but not a visual tracking task.

• Word length effect (Baddeley, Thomson and Buchanan, 1975): lists of short words are recalled better than lists of long words, because the loop holds about as much as can be said in around 2 seconds; articulatory suppression removed the effect for written lists.
• KF (Shallice and Warrington, 1970): poor verbal STM but better visual STM, suggesting separate verbal and visual stores.

Evaluation

Strengths: explains doing two tasks at once, which the multi-store model cannot; supported by dual-task experiments, case studies (KF) and brain scans; useful for understanding learning difficulties.

Weaknesses:

• The central executive is vague and its capacity has never been measured.
• Lieberman (1980) argued that the sketchpad wrongly assumes spatial information is visual: blind people have good spatial awareness.
• Most evidence comes from artificial lab tasks.
• It only covers STM, so it says little about LTM.

Worked example

Explain why it is hard to listen to a podcast while reading, but easy to listen to a podcast while drawing a map.

1. Listening and reading both need the phonological loop (reading is converted to sound by the articulatory control process), so they compete and interfere.
2. Drawing a map uses the visuo-spatial sketchpad, a different component, so both tasks can be done at once.

Answer: Podcast and reading compete for the phonological loop; drawing uses the visuo-spatial sketchpad, so it does not interfere.

Check yourself

1. Which component of working memory directs attention and controls the other components?
   • Phonological store
   • Visuo-spatial sketchpad
   • Central executive
   • Episodic buffer
   Show answer

   Central executive. The central executive is the controller.

2. Which part of the phonological loop acts as the 'inner voice'?
   • Articulatory control process
   • Phonological store
   • Visual cache
   • Inner scribe
   Show answer

   Articulatory control process. It silently rehearses words.

3. Why did Baddeley add the episodic buffer in 2000?
   • To replace the central executive
   • To store only visual information
   • To explain how information from different components and LTM is combined
   • To explain the sensory register
   Show answer

   To explain how information from different components and LTM is combined. The original model could not explain linking with LTM and across codes.

4. A participant can track a moving dot while repeating 'the the the', but struggles to read while repeating it. What does this show?
   • STM is a single store
   • Reading and repeating use the same component, but tracking uses a different one
   • The central executive is not needed
   • Long-term memory is acoustic
   Show answer

   Reading and repeating use the same component, but tracking uses a different one. Interference only happens when tasks share a component.

5. What was Lieberman's (1980) criticism of the model?
   • The central executive has too large a capacity
   • The model ignores the sensory register
   • The phonological loop does not exist
   • The sketchpad assumes spatial information is visual, but blind people have good spatial awareness
   Show answer

   The sketchpad assumes spatial information is visual, but blind people have good spatial awareness. Spatial information need not be visual.

6. Give one weakness of the working memory model.
   Show answer

   The central executive is vague: it is described by what it does but its capacity and workings have never been clearly measured.. The central executive is poorly specified.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification (issue January 2026)
• Pearson International A Level Psychology component guide: WPS01
• Simply Psychology: Working memory model
• IAL Psychology revision site: The working memory model

Last updated 8 November 2026. Spotted a mistake? Tell us.