The Multi-Store Model of Memory (MSM) (AQA A-Level Psychology): Revision Notes
The Multi-Store Model of Memory (MSM)
Introduction to the model
Atkinson and Shiffrin (1968) proposed the Multi-Store Model (MSM) as one of the earliest explanations of human memory. This model suggests that memory consists of three separate components that work together in a linear sequence: the sensory register, short-term memory (STM), and long-term memory (LTM).
According to the MSM, information flows through these stores sequentially, with each store having distinct characteristics in terms of how information is processed, how much can be stored, and for how long it remains accessible.
The MSM proposes that memory is not a single unified system, but rather consists of three distinct stores that process information in a specific linear sequence. Understanding this sequential flow is essential for grasping how the model explains human memory processes.
The three memory stores
Sensory register
The sensory register acts as the initial gateway for all incoming information from our environment. This store receives data from all five senses simultaneously and holds it in its original, unprocessed form.
Key features of the Sensory Register:
- Capacity: Unknown but believed to be very large, potentially unlimited
- Duration: Extremely brief, lasting less than one second (approximately 250 milliseconds)
- Coding: Raw and unprocessed information that is modality-specific (visual information stays visual, auditory stays auditory, etc.)
Information in the sensory register will be lost through decay unless it receives attention, which allows it to transfer to short-term memory.
Short-term memory (STM)
Information that receives attention moves from the sensory register into short-term memory. This store acts as a temporary workspace where we consciously process information.
Key features of Short-Term Memory:
- Capacity: Limited to approximately 7±2 'chunks' of information (Miller, 1956)
- Duration: Limited to approximately 20 seconds without rehearsal (Peterson & Peterson, 1959)
- Coding: Primarily acoustic (sound-based), meaning we tend to store information based on how it sounds (Baddeley, 1966)
For example, when trying to remember a phone number, you might repeat it to yourself using its sounds rather than visualising how it looks written down. Information in STM can be maintained through rehearsal or transferred to long-term memory for permanent storage.
Long-term memory (LTM)
Long-term memory serves as our permanent storage system, holding information that has been successfully transferred from STM through rehearsal and processing.
Key features of Long-Term Memory:
- Capacity: Unlimited - there appears to be no upper limit to how much information can be stored
- Duration: Lifetime - information can potentially last forever (Bahrick, 1975)
- Coding: Semantic (meaning-based), focusing on the significance and connections between pieces of information (Baddeley, 1966)
Information stored in LTM can be retrieved back to STM when needed, allowing us to consciously access our memories.
Key research studies
Miller (1956): "The magical number seven, plus or minus two"
Research Study: Miller's Investigation of STM Capacity
Aim: To investigate the capacity limitations of short-term memory.
Method: Miller conducted a literature review examining existing research on perception and STM from the 1930s to 1950s.
Findings: The research consistently showed that people could successfully process approximately seven separate pieces of information before their performance declined significantly. However, this capacity could be extended through 'chunking' - organising individual items into meaningful groups.
Example: A phone number like 07673456789 contains 11 individual digits, but when chunked as 0767-345-6789, it becomes only four meaningful units, making it easier to remember.
Conclusion: STM has a limited capacity of 7±2 chunks, but this can be optimised through effective organisation strategies.
Evaluation:
- Strengths: Supported by subsequent research, including Jacobs (1887) who found similar digit span results using controlled laboratory conditions
- Weaknesses: Miller didn't specify the exact size of chunks, making it difficult to determine precise STM capacity. The research also didn't account for individual differences or age-related changes in memory capacity.
Peterson and Peterson (1959): Duration of STM
Research Study: Peterson and Peterson's STM Duration Investigation
Aim: To investigate how long information can be held in STM without rehearsal.
Method: 24 university students were presented with three-letter combinations (trigrams) such as 'JBW' or 'PDX'. Immediately after hearing the trigram, participants had to count backwards in 3s or 4s from a given number for varying time intervals (3, 6, 9, 12, 15, or 18 seconds). This prevented rehearsal of the trigram.
Findings: Memory accuracy decreased dramatically as the delay increased. At 3 seconds, participants recalled about 80% of trigrams correctly, but this fell to only 10% after 18 seconds.
Conclusion: Without rehearsal, information in STM decays rapidly, lasting approximately 18 seconds at most.
Evaluation:
- Strengths: Highly controlled laboratory conditions allowed for precise measurement and easy replication
- Weaknesses: Low ecological validity as memorising meaningless trigrams doesn't reflect real-world memory use. The sample consisted entirely of psychology students who may have had prior knowledge of memory research, potentially affecting their behaviour.
Bahrick (1975): Duration of LTM
Research Study: Bahrick's Investigation of LTM Duration
Aim: To investigate whether information stored in long-term memory can last a lifetime.
Method: 392 American university graduates were shown photographs from their high school yearbooks and asked to match names with faces.
Findings: Participants demonstrated remarkable long-term retention: 90% accuracy after 14 years and 60% accuracy even after 47 years since graduation.
Conclusion: Long-term memory can retain information for decades, supporting the MSM's claim that LTM has an extremely long duration.
Evaluation:
- Strengths: High ecological validity as it used real-life memories rather than artificial laboratory tasks
- Weaknesses: Limited population validity (only American graduates) and inability to determine whether the decline in accuracy was due to normal ageing or actual memory limitations. The study also couldn't control for how much exposure participants had to classmates during school.
Supporting evidence
The case of Clive Wearing
Clive Wearing provides compelling real-world evidence for the MSM's structure. After contracting a virus that caused severe brain damage and amnesia, Wearing could only retain new information for 20-30 seconds. However, he could still access some long-term memories, such as his wife's name, and retained procedural memories like playing piano.
Wearing's case demonstrates the separation between STM and LTM proposed by the MSM - damage to his memory system prevented the transfer of information from STM to LTM while leaving some long-term memories intact.
Neurological evidence
Brain imaging studies have identified different neural regions associated with STM and LTM processing:
Neurological Support for the MSM:
- STM tasks activate the hippocampus and subcortical areas
- LTM tasks primarily involve the motor cortex
- The hippocampus plays a crucial role in transferring information from STM to LTM
This neurological evidence supports the MSM's proposal that memory consists of distinct, separate stores.
Evaluation of the MSM
Strengths
Empirical support: Extensive research supports the model's key features. Miller (1956) confirmed STM's limited capacity, Peterson and Peterson (1959) demonstrated its brief duration, and Bahrick (1975) showed LTM's extended duration.
Clinical evidence: Case studies like Clive Wearing provide real-world validation of the model's structure, showing how damage to specific memory processes affects different stores in predictable ways.
Influential framework: The MSM established fundamental concepts in memory research and influenced the development of more sophisticated models.
Weaknesses
Oversimplification: The model presents a nomothetic (universal) approach that may not capture the complexity of individual memory experiences. An idiographic approach examining personal memory differences might provide more accurate insights.
Reductionist approach: By breaking memory into isolated components, the MSM may overlook the integrated nature of human memory processing. Memory likely involves more complex interactions than the simple linear flow the model suggests.
Limited explanatory power: The model cannot account for phenomena like parallel processing or the complexity of short-term memory, leading to the development of alternative models such as Baddeley and Hitch's (1974) Working Memory Model.
Research limitations: Many supporting studies used artificial laboratory tasks that lack ecological validity, making it unclear how well the findings apply to everyday memory use.
While the MSM has been highly influential and is supported by substantial research evidence, it's important to recognise that it represents an early attempt to understand memory. More recent models have addressed many of its limitations while building upon its foundational insights.
Summary
Key Points to Remember:
-
The Multi-Store Model proposes three separate memory stores: sensory register, short-term memory, and long-term memory, each with distinct characteristics
-
STM has a limited capacity of 7±2 chunks and duration of ~20 seconds, while LTM has unlimited capacity and lifetime duration
-
Key research by Miller (1956), Peterson and Peterson (1959), and Bahrick (1975) provides empirical support for the model's predictions
-
The case of Clive Wearing and neurological evidence demonstrate the separation between different memory stores
-
While influential and well-supported, the MSM is criticised for being overly simplistic and reductionist in its approach to understanding human memory