- Central executive
- The attentional control system at the heart of working memory. It allocates resources between the slave systems, switches and divides attention between tasks, inhibits irrelevant information, and connects working memory to long-term memory. Miyake et al. (2000) identified three separable but correlated executive functions — inhibition, updating, and shifting — that help clarify some of the processes often attributed to the central executive.
A major criticism of the model is that the central executive is defined largely by the functions it performs rather than by a clearly specified mechanism, making it difficult to generate precise predictions about how it operates. It is sometimes described as a "ragbag" of executive functions.
- Phonological loop
- The "inner ear and inner voice" — a slave system for holding and rehearsing verbal and acoustic information. It consists of two sub-components: (1) the phonological store, which holds a relatively brief phonological trace that decays unless refreshed through rehearsal; and (2) the articulatory rehearsal process, which refreshes traces by sub-vocal repetition. Evidence: the phonological similarity effect (acoustically similar items harder to recall), the word length effect (more short than long words recalled), and the irrelevant speech effect (background speech disrupts recall even when semantically irrelevant).
- Phonological store
- The passive component of the phonological loop — holds a relatively brief phonological trace that is vulnerable to decay unless refreshed through rehearsal. The commonly cited duration is around 1.5–2 seconds without rehearsal. Auditory verbal information has relatively direct access to the phonological system, whereas visually presented verbal material can be recoded phonologically through subvocal rehearsal. The phonological store is disrupted by phonologically similar stimuli (the phonological similarity effect) and by the irrelevant speech effect.
- Articulatory rehearsal process
- The active "inner voice" component of the phonological loop, which refreshes decaying phonological traces through sub-vocal rehearsal. It recodes visual verbal material into phonological form for storage. The word length effect arises because longer words take longer to articulate, reducing the number of items that can be rehearsed per cycle before decay. Articulatory suppression (saying "the the the" aloud) prevents this recoding and abolishes the word length effect.
- Visuospatial sketchpad
- The "inner eye" — a slave system for maintaining and manipulating visual and spatial information. Supports mental imagery, spatial navigation, and tasks requiring the maintenance of spatial relationships (e.g., mental rotation, map reading). Dual-task studies show that spatial tracking tasks selectively disrupt visual memory tasks, demonstrating independence from the phonological loop. The sketchpad may itself consist of separate visual and spatial components.
- Episodic buffer
- Added by Baddeley (2000) to account for the integration of information from different sources. A limited-capacity multimodal store that binds together information from the phonological loop, visuospatial sketchpad, and long-term memory into coherent, integrated episodic representations. Baddeley has described its capacity as approximately 4 chunks. It links working memory to long-term memory and is accessible to conscious awareness. Explains why people can recall more words when presented in a sentence (benefiting from LTM structure) than in a random list.
Working memory does not operate independently of long-term memory. Stored knowledge can help organise and chunk incoming material, allowing people to handle more than immediate capacity limits would otherwise permit.
- Word length effect
- The finding that people recall more short words than long words from immediate verbal memory (Baddeley et al., 1975). Memory span correlates with how many words can be articulated within one rehearsal cycle before traces decay. The word length effect is consistent with the articulatory rehearsal account and is abolished by articulatory suppression, confirming it depends on the active rehearsal mechanism. However, later research suggests that factors such as articulation rate and phonological complexity also contribute to the effect.
- Dual-task paradigm
- A foundational experimental method for studying the structure of working memory. Participants perform two tasks simultaneously, and the pattern of interference reveals which components they share. If two tasks interfere strongly, they draw on the same sub-system. Baddeley and Hitch's (1974) original dual-task studies showed that holding digit strings interfered only modestly with reasoning — implying separate systems for verbal storage and reasoning, unlike a single-gateway STM model would predict.