The Working Memory Model

The Working Memory Model (WMM) was proposed by Baddeley & Hitch (1974) as an alternative to the MSM's concept of a single, unitary STM store. Rather than a simple storage system, working memory is an active processor that manipulates information while it is being used for complex cognitive tasks such as reasoning, comprehension, and learning.

Key Definition: Working memory is the part of memory that is active when we are temporarily storing and manipulating information, for example during problem-solving, mental arithmetic, or reading.

---

Components of the Working Memory Model

graph TD
    CE["Central Executive<br/>(Supervisory attention system)"]
    PL["Phonological Loop"]
    VSS["Visuo-Spatial Sketchpad"]
    EB["Episodic Buffer<br/>(Added by Baddeley, 2000)"]
    LTM["Long-Term Memory"]

    CE --> PL
    CE --> VSS
    CE --> EB
    EB --> LTM

    PL --> PS["Phonological Store<br/>(Inner ear)"]
    PL --> AP["Articulatory Process<br/>(Inner voice)"]

    VSS --> VC["Visual Cache<br/>(Inner eye)"]
    VSS --> IS["Inner Scribe<br/>(Spatial/movement)"]

1. Central Executive (CE)

Feature	Detail
Function	Directs attention and allocates processing resources to the other components (the "slave systems")
Capacity	Very limited — it can only attend to a small number of tasks at once
Modality	Modality-free — it can process any type of information
Role	Decides which information to attend to, switches attention between tasks, and coordinates the other components

The central executive is the most important but least understood component. It does not store information itself but rather acts as a supervisory attention system that controls the other components.

2. Phonological Loop (PL)

The phonological loop processes auditory and verbal information (speech-based). It has two sub-components:

Sub-component	Function
Phonological store ("inner ear")	Holds speech-based information for 1–2 seconds. Information decays unless refreshed by the articulatory process.
Articulatory process ("inner voice")	Rehearses information by silently repeating it (like a loop of inner speech). Also converts written words into a phonological (speech-based) code.

Evidence — the word length effect:

• Baddeley, Thomson & Buchanan (1975) found that people could recall more short words (e.g., sum, wit) than long words (e.g., opportunity, university) in immediate serial recall.
• This is because longer words take more time to rehearse in the articulatory process, so fewer can be maintained in the phonological loop before they decay.
• This supports the existence of a time-limited articulatory rehearsal loop.

3. Visuo-Spatial Sketchpad (VSS)

The visuo-spatial sketchpad processes visual and spatial information — what things look like and where they are in relation to each other.

Sub-component	Function
Visual cache ("inner eye")	Stores visual information about the appearance of objects (e.g., shape, colour)
Inner scribe	Records the arrangement of objects in the visual field and rehearses/manipulates spatial information (e.g., planning a route)

• The VSS has limited capacity — it is difficult to perform two visual tasks simultaneously (e.g., imagining a route while also visualising a picture).
• Logie (1995) proposed the distinction between the visual cache and the inner scribe.

4. Episodic Buffer (Baddeley, 2000)

Baddeley added the episodic buffer in 2000 because the original WMM could not explain how information from different sources (visual, auditory, LTM) was combined into a coherent experience.

Feature	Detail
Function	Acts as a temporary store that integrates information from the phonological loop, visuo-spatial sketchpad, and LTM into a single, coherent episode
Capacity	Limited — approximately 4 chunks of information
Control	Controlled by the central executive
Link to LTM	Provides a bridge between working memory and long-term memory, allowing information to flow between them

Key Definition: The episodic buffer is a temporary store that integrates information from the other components of working memory and from long-term memory into coherent episodes or scenes.

---

Evidence for the Working Memory Model (AO3)

Dual-Task Studies

The strongest evidence for the WMM comes from dual-task experiments, which show that:

• Performing two tasks that use the same component simultaneously leads to interference and poor performance (e.g., two visual tasks or two verbal tasks).
• Performing two tasks that use different components simultaneously causes no interference (e.g., a verbal task and a visual task at the same time).