Working Memory Model (WMM) (AQA A-Level Psychology): Revision Notes
Working Memory Model (WMM)
Overview of the working memory model
The Working Memory Model (WMM) represents a significant advancement from the simple multi-store model of memory. Developed by Baddeley and Hitch in 1974, this model explains short-term memory as an active system that can simultaneously hold and process multiple pieces of information rather than being just a passive temporary store.
Unlike Atkinson and Shiffrin's multi-store model, which viewed STM as a single, unified store, the WMM proposes that working memory consists of several specialised components that work together. This model emerged because researchers recognised that STM was far more complex than simply transferring information to long-term memory - it actively manipulates information whilst being worked upon, hence the term 'working' memory.
The key innovation of the WMM was recognising that short-term memory is not just a passive store, but an active workspace where information is manipulated and processed simultaneously.
The model initially comprised three main components, with a fourth added later in 2000 to address certain limitations:
The central executive
The Central Executive (CE) functions as the supervisory system of working memory, acting as the 'boss' that coordinates and controls the other slave systems. This component plays a crucial role in determining which information deserves attention and which should be ignored.
The CE processes information from all sensory modalities and directs information flow to the appropriate slave systems. It possesses extremely limited storage capacity and can only effectively handle one stream of information at a time. This limitation explains why we find it challenging to perform multiple demanding tasks simultaneously, such as having a complex conversation while navigating through heavy traffic.
One of the CE's key functions involves attentional control - it allows us to switch our focus between different sources of information and helps maintain concentration on relevant tasks while filtering out distractions. The CE also manages the balance between competing tasks when attention must be divided.
Research evidence for the central executive
Research Example: Baddeley's Dual Task Study (1996)
Participants were asked to generate lists of random numbers while simultaneously switching between pressing numbers and letters on a keyboard. The results showed that participants struggled with this combination because both tasks competed for the same CE resources, demonstrating the limited capacity of this component.
Further neurological evidence comes from D'Esposito et al. (1995), who used fMRI scans to observe brain activity during dual tasks. When participants performed verbal and spatial tasks simultaneously, the prefrontal cortex showed activation, but not when tasks were performed separately. This suggests a specific brain region associated with CE functioning.
Evaluation of the central executive
The CE remains the least understood component of the WMM. Its vagueness presents both strengths and weaknesses - while this flexibility allows the model to explain various experimental findings, it also means the CE can be used to account for almost any result.
When two tasks cannot be performed together, researchers can argue the processing components are conflicting, but when tasks can be completed simultaneously, they claim the tasks don't exceed available resources.
Most contemporary researchers view the CE as an attentional control system rather than a memory store, distinguishing it from the phonological loop and visuo-spatial sketchpad, which function as specialised storage systems.
Phonological loop
The Phonological Loop (PL) specialises in processing auditory information and maintaining the order of sounds and words. This component closely resembles the rehearsal system from the multi-store model, but with enhanced functionality and clearer structure.
The PL has limited capacity, typically allowing storage of information that can be spoken aloud within approximately two seconds. This explains why we can usually remember about as many words as we can say in two seconds, regardless of the actual number of words involved.
The "two-second rule" is a key characteristic of the phonological loop - its capacity is determined by how much information can be articulated in about two seconds, not by the number of items.
Baddeley further subdivided the PL into two components in 1986:
Primary acoustic store
The Primary Acoustic Store (PAS) functions like an 'inner ear', holding recently heard words for a brief period. This store maintains acoustic information in its original auditory form, allowing us to retain the sound of words and speech patterns.
Articulatory process
The Articulatory Process (AP) acts like an 'inner voice', enabling subvocal rehearsal of information within the phonological loop. This component links directly to speech production and allows us to repeat words silently to ourselves, maintaining information in the PL through rehearsal.
Research evidence for the phonological loop
Research Example: Word Length Effect Study (Baddeley et al., 1975)
Participants were presented with lists of either short words (like "cat", "pen", "day") or long words (like "university", "refrigerator", "elephant") in serial order. Results showed significantly better recall for short words, supporting the idea that PL capacity depends on temporal duration rather than the number of items.
Trojani and Grossi (1995) provided compelling evidence for the PL's independence through their case study of patient SC, who suffered brain damage affecting phonological loop functioning while leaving the visuo-spatial sketchpad intact. This demonstrated that these components operate as separate systems.
Evaluation of the phonological loop
Neuroimaging studies using PET scans have revealed that different brain areas activate during verbal versus visual tasks, providing biological evidence for separate PL and VSS systems. This neurological support strengthens the model's credibility.
The PL shows strong evolutionary connections to human vocal language development. The slave system appears crucial for developing our enhanced short-term memory for vocalisations, which subsequently facilitated learning complex language abilities including grammar and meaning expression.
Visuo-spatial sketchpad
The Visuo-spatial Sketchpad (VSS) handles non-auditory information, serving as a temporary storage system for visual and spatial items and their relationships. Often described as the 'inner eye', this component processes what objects look like and where they are located in space.
The VSS enables individuals to navigate their environment and interact with the physical world around them. Information undergoes coding and rehearsal through mental imagery, allowing us to manipulate visual information mentally.
Logie (1995) proposed dividing the VSS into two subcomponents:
Visual cache
The Visual Cache (VC) stores information about visual form and colour, maintaining the appearance characteristics of objects and scenes.
Inner scribe
The Inner Scribe (IS) handles spatial relationships and rehearses information, transferring it to the visual cache when needed. This component manages the spatial arrangement of objects and their movements through space.
Research evidence for the visuo-spatial sketchpad
Research Example: Dual Task Study (Gathercole and Baddeley, 1993)
Participants found it difficult to simultaneously track a moving light point and describe angles on a hollow letter F, since both tasks required VSS resources. However, participants managed combined verbal and tracking tasks more easily, as these utilised different slave systems (PL and VSS respectively), demonstrating separate processing systems.
Klauer and Zhao (2004) reported greater interference between two visual tasks compared to combined visual and spatial tasks, providing evidence for the separate visual cache and inner scribe components within the VSS.
Evaluation of the visuo-spatial sketchpad
PET scanning technology has revealed that the PL and VSS activate different brain regions - visual tasks show left hemisphere activation while spatial tasks activate the right hemisphere. This neurological evidence supports the division of VSS into separate components.
Many VSS studies employ artificial dual-task techniques requiring participants to perform simultaneous activities rarely encountered in daily life. This artificial nature raises questions about mundane realism and whether findings genuinely reflect how the VSS operates in real-world situations.
The episodic buffer
Baddeley added the Episodic Buffer (EB) in 2000 to address limitations in the original three-component model. The PL and VSS possess limited capacity and the CE lacks storage capability, yet the model needed to explain how integrated information from multiple sources could be temporarily maintained.
The EB serves as a general storage system capable of combining information from the CE, PL, VSS, and long-term memory. This component provides a workspace where different types of information can be integrated and manipulated together, addressing the model's previous inability to explain how complex, multifaceted information is processed.
The episodic buffer was added to solve the "binding problem" - how different types of information are integrated and held together in working memory when the original components had such limited capacity.
Research evidence for the episodic buffer
Prabhakaran et al. (2000) used fMRI scanning to identify greater right-frontal brain activation when processing combined verbal and spatial information, compared to greater posterior activation for non-combined information. This provides biological evidence for a buffer system that integrates different information types.
Research Example: Patient Study (Alkhalifa, 2009)
A patient with severely impaired long-term memory nevertheless demonstrated STM capacity for up to 25 prose items - far exceeding typical PL and VSS capacities combined. This suggested an additional component (the EB) that maintains integrated information in working memory until recall is required.
Contemporary research on the episodic buffer
Alkhalifa (2009) conducted controlled experiments presenting university students with number sequences either sequentially or simultaneously in different screen locations, with varying complexity levels designed to overwhelm both PL and VSS capacities. Participants receiving sequential presentation performed better on problem-solving questions than those receiving parallel presentation, suggesting that sequential processing utilises the EB more effectively and demonstrates that working memory's total capacity exceeds that of individual components.
Evaluation of the working memory model
Strengths:
- Provides a more realistic and detailed explanation of short-term memory than the multi-store model
- Supported by extensive research evidence from multiple methodologies including case studies, experimental research, and neuroimaging
- Successfully explains phenomena that the multi-store model could not account for
- Has practical applications, particularly in educational settings and understanding attention disorders
The WMM's strength lies in its ability to explain complex memory phenomena that simpler models cannot account for, such as why we can perform some tasks simultaneously while others interfere with each other.
Limitations:
- The central executive remains poorly understood and somewhat vague, making it difficult to test scientifically
- Many supporting studies use artificial laboratory tasks that may not reflect real-world memory use
- The model continues to evolve, suggesting it may still be incomplete
- Some researchers argue the components may not be as separate as the model suggests
The vagueness of the central executive is both a strength and weakness - while it allows flexibility in explaining results, it also makes the component difficult to test and potentially unfalsifiable.
Real-world applications
The WMM has significant practical implications, particularly for children with Attention Deficit Hyperactivity Disorder (ADHD) who often experience working memory impairments. Alloway (2006) recommends several educational strategies:
- Use brief and simple instructions to prevent information overload
- Break complex instructions into individual steps
- Frequently repeat important instructions
- Ask children to periodically repeat instructions back
Additionally, Klingberg et al. (2002) found that computerised working memory training using systematic exercises produces cognitive improvements, particularly benefiting individuals with poor working memory function.
Key Points to Remember:
- The Working Memory Model explains STM as an active system with multiple specialised components rather than a single passive store
- The Central Executive coordinates the slave systems but has very limited storage capacity and controls attentional focus
- The Phonological Loop processes auditory information through the Primary Acoustic Store and Articulatory Process, with capacity limited to about 2 seconds of speech
- The Visuo-spatial Sketchpad handles visual and spatial information via the Visual Cache and Inner Scribe components
- The Episodic Buffer integrates information from all components and LTM, addressing the model's original limitations