Models of Memory
Atkinson and Shiffrin (1968) created the Multi-Store Model:
The multi-store model proposes that memory consists of three stores- a sensory register, a short-term store and a long-term store, and information has to become a memory.
Information from our environment (e.g. visual or auditory) initially goes into the sensory register. You don’t really notice much of this stuff. However, the information will pass into short-term memory if you pay attention to it or think about it.
Short-term memory has a finite capacity and duration. But if the information is processed further (rehearsed) then it can be transferred to long-term memory. In theory, the information can then remain there forever. (Unless you really, really need to remember it, in which case it’ll probably stay there until something more interesting comes along, like a bee or a cloud.)
Studies support the multi-store model:
Milner et al (1957) carried out a case study on a patient called HM who had suffered from severe and frequent epilepsy. His seizures were based on a brain structure called the hippocampus. Doctors decided to remove part of the brain around this area surgically. The operation reduced his epilepsy but led to his suffering from memory loss. He could still form short-term memories but was unable to create new long-term memories. This case study supports the idea that different types of memory types of memory are separate systems in the brain.
Limitations of the Multi-Store Model:
In the model, information is transferred from the STM to LTM through rehearsal. But in real life, people don’t always spend time rehearsing, yet they still transfer information into LTM. Rehearsal is not always needed for information to be stored and some items can’t be rehearsed, e.g smells
The model is oversimplified. It assumes there is only one long-term store and one short-term store. This has been disproved by evidence from brain-damaged patients, suggesting several different short-term stores, and other evidence suggesting different long-term stores
Baddeley and Hitch (1974) developed the Working Memory Model:
Baddeley and Hitch developed a multi-store model of STM called the ‘Working Memory Model’. Their model proposed that STM, rather than being a single store, is an active processor which contains several different stores.
The central executive is a key component and can be described as attention. It has a limited capacity and control ‘slave’ systems that also have limited capacity:
The Phonological Loop holds speech-based information- it’s made up of a phonological store (the inner ear) and an articulatory process (the inner voice, which rehearses information by repeating it). It has a limited capacity (approx 2secs before decay) and is a temporary storage system for holding verbal information in acoustic form. (Its kind of like an "inner voice")
The visuospatial sketchpad deals with the temporary storage of visual and spatial information. It is a limited capacity (approx 3-4 items), a temporary memory system for holding immediate visual and spatial information e.g what it looks like, and where it is?
The episodic buffer (which was added to the model in 2000) briefly stores information from the other subsystems and integrates it together, along with information from LTM, to make complete scenes or ‘episodes’ He proposed that tasks such as remembering prose (a common form of writing that follows the natural flow of speech and doesn't have the repeating rhythm of poetry) requires quiet complex information processing.
Baddeley and Logie (1999) Phonological Loop:
They redefined this model
The phonological loop consists of a passive storage system (non-attention-based/automatic) called the phonological store, which is linked to an active (attention-based) rehearsal system called the articulatory loop- whereby words can be maintained by sub-vocal repetition (silently pronouncing words in your mind).
The working memory model came from experimental evidence:
Baddeley and Hitch based their model on results from studies that used ‘interference tasks’:
If participants are asked to perform two tasks simultaneously that use the same system, their performance will be affected- e.g. saying ‘the the the’ while silently reading something difficult.
According to the working memory model, both these tasks use the phonological loop. This has limited capacity, so it can’t cope with both tasks. Performance on one, or both tasks, will be affected.
However, if two tasks involve different systems, performance isn’t affected on either task (e.g saying ‘the the the’ whilst tracking a moving object)
As usual, the Model has strengths:
Shallice and Warrington (1974) found support for the working memory model through their case study of KF:
KF was a brain-damaged patient who had an impaired STM. His problem was with immediate recall of words presented verbally, but not with visual information. This suggested he had an impaired articulatory loop but an intact visuospatial sketchpad, therefore providing evidence for the working memory model view of STM. This finding could not have been explained using the multi-store model of memory, which proposed that short-term memory was just one system
And Weaknesses:
The working memory model is currently the main model of short-term memory, but some psychologists have criticised it. For examples:
The model only explains how information is dealt with in short-term memory. It doesn’t explain how information is transferred to long-term memory.
Much of the research which has supported the working memory model has been laboratory studies. This reduces the ecological validity of the evidence, as highly controlled studies might not be representative of what happens in the real world
fMRI scans:
Bunge et al (2000) found that when Ps were doing a dual task, fMRI scans showed significantly more activation
This shows that the increased attentional demands of the two simultaneous tasks increased brain activity
This provides support for the proposed role of the central executive as the aspect which directs attention and allocates resources to the slave systems to carry out their task
Clarification of the central executive?
It could be argued that the role of the CE is a little vague/simplistic
Is it possible that the CE, like the other aspects of the model, is also not unitary?
Eslinger and Damasio's (1985)’s patient EVR (cerebral tumour removed), performed well on reasoning tasks would involve the CE, this suggests that the CE is intact but not fully functioning- therefore must have sub-components involved for each task respectively (CE is more complex than WMM suggests, which challenges the integrity of WMM)
Logie (1995):
Suggests that the visual cache stores information about the visual form and colour and that the inner scribe processes spatial and movement information. Klaur and Zhao 2004- supported this idea by asking P’s to carry out one of two primary tasks
A visual task- A spatial task
At the same time doing either
A spatial interference task- A visual interference task- No secondary task
They found that the performance of spatial tasks was much poorer for those doing spatial interference tasks than people doing the visual interference task and vice versa
PET Scans:
Studies using PET scans have also provided evidence for separate spatial and visual systems
There appears to be more activity in the left half of the brain for visual tasks but more in the right half during spatial tasks
Bunge et al (2000):
Used fMRI to see which parts of the brain were more active when participants were doing tasks (reading a sentence and recalling the final word in each sentence)
The same brain areas were active in either dual or single-task conditions but there was significantly more activation in the dual-task condition single-task
Indicating that increased demands were reflected in brain activity
LH (Trojano and Grossi):
Another patient LH, who had been involved in a road accident, performed better on spatial tasks than those involving visual imagery
This suggests separate visual and spatial systems
KF- showed STM works independent of LTM as he had no problem with long-term learning but some aspects of his immediate memory were impaired
LH- Performed better on spatial tasks than those involving visual imagery
Strengths of Working Memory Model:
It is very influential
There is a considerable amount of research evidence to support the model
Especially more recent studies show direct links between certain tasks and activation of parts of the brain that represent components of the WM model
Makes sense of a range of tasks i.e Reading- phonological loop
Navigation- Visual-spatial processing
Limitations:
WMM does not offer a complete understanding of how memory works- no LTm or SM (although acknowledges these components proposed by MSM)
The exact role played by the central executive remains unclear. The capacity of CE is still unclear
Also, WMM fails to account for musical memory because we can listen to instrumental music without impairing performance on other acoustic tasks
Evidence for Central Executive:
The Central Executive has a critical role in attention, planning and coordination, and is the most important and coordination, and is the most important and most flexible component of working memory
Baddeley himself accepts that this complexity makes it considerably harder to investigate
Models of Memory
Atkinson and Shiffrin (1968) created the Multi-Store Model:
The multi-store model proposes that memory consists of three stores- a sensory register, a short-term store and a long-term store, and information has to become a memory.
Information from our environment (e.g. visual or auditory) initially goes into the sensory register. You don’t really notice much of this stuff. However, the information will pass into short-term memory if you pay attention to it or think about it.
Short-term memory has a finite capacity and duration. But if the information is processed further (rehearsed) then it can be transferred to long-term memory. In theory, the information can then remain there forever. (Unless you really, really need to remember it, in which case it’ll probably stay there until something more interesting comes along, like a bee or a cloud.)
Studies support the multi-store model:
Milner et al (1957) carried out a case study on a patient called HM who had suffered from severe and frequent epilepsy. His seizures were based on a brain structure called the hippocampus. Doctors decided to remove part of the brain around this area surgically. The operation reduced his epilepsy but led to his suffering from memory loss. He could still form short-term memories but was unable to create new long-term memories. This case study supports the idea that different types of memory types of memory are separate systems in the brain.
Limitations of the Multi-Store Model:
In the model, information is transferred from the STM to LTM through rehearsal. But in real life, people don’t always spend time rehearsing, yet they still transfer information into LTM. Rehearsal is not always needed for information to be stored and some items can’t be rehearsed, e.g smells
The model is oversimplified. It assumes there is only one long-term store and one short-term store. This has been disproved by evidence from brain-damaged patients, suggesting several different short-term stores, and other evidence suggesting different long-term stores
Baddeley and Hitch (1974) developed the Working Memory Model:
Baddeley and Hitch developed a multi-store model of STM called the ‘Working Memory Model’. Their model proposed that STM, rather than being a single store, is an active processor which contains several different stores.
The central executive is a key component and can be described as attention. It has a limited capacity and control ‘slave’ systems that also have limited capacity:
The Phonological Loop holds speech-based information- it’s made up of a phonological store (the inner ear) and an articulatory process (the inner voice, which rehearses information by repeating it). It has a limited capacity (approx 2secs before decay) and is a temporary storage system for holding verbal information in acoustic form. (Its kind of like an "inner voice")
The visuospatial sketchpad deals with the temporary storage of visual and spatial information. It is a limited capacity (approx 3-4 items), a temporary memory system for holding immediate visual and spatial information e.g what it looks like, and where it is?
The episodic buffer (which was added to the model in 2000) briefly stores information from the other subsystems and integrates it together, along with information from LTM, to make complete scenes or ‘episodes’ He proposed that tasks such as remembering prose (a common form of writing that follows the natural flow of speech and doesn't have the repeating rhythm of poetry) requires quiet complex information processing.
Baddeley and Logie (1999) Phonological Loop:
They redefined this model
The phonological loop consists of a passive storage system (non-attention-based/automatic) called the phonological store, which is linked to an active (attention-based) rehearsal system called the articulatory loop- whereby words can be maintained by sub-vocal repetition (silently pronouncing words in your mind).
The working memory model came from experimental evidence:
Baddeley and Hitch based their model on results from studies that used ‘interference tasks’:
If participants are asked to perform two tasks simultaneously that use the same system, their performance will be affected- e.g. saying ‘the the the’ while silently reading something difficult.
According to the working memory model, both these tasks use the phonological loop. This has limited capacity, so it can’t cope with both tasks. Performance on one, or both tasks, will be affected.
However, if two tasks involve different systems, performance isn’t affected on either task (e.g saying ‘the the the’ whilst tracking a moving object)
As usual, the Model has strengths:
Shallice and Warrington (1974) found support for the working memory model through their case study of KF:
KF was a brain-damaged patient who had an impaired STM. His problem was with immediate recall of words presented verbally, but not with visual information. This suggested he had an impaired articulatory loop but an intact visuospatial sketchpad, therefore providing evidence for the working memory model view of STM. This finding could not have been explained using the multi-store model of memory, which proposed that short-term memory was just one system
And Weaknesses:
The working memory model is currently the main model of short-term memory, but some psychologists have criticised it. For examples:
The model only explains how information is dealt with in short-term memory. It doesn’t explain how information is transferred to long-term memory.
Much of the research which has supported the working memory model has been laboratory studies. This reduces the ecological validity of the evidence, as highly controlled studies might not be representative of what happens in the real world
fMRI scans:
Bunge et al (2000) found that when Ps were doing a dual task, fMRI scans showed significantly more activation
This shows that the increased attentional demands of the two simultaneous tasks increased brain activity
This provides support for the proposed role of the central executive as the aspect which directs attention and allocates resources to the slave systems to carry out their task
Clarification of the central executive?
It could be argued that the role of the CE is a little vague/simplistic
Is it possible that the CE, like the other aspects of the model, is also not unitary?
Eslinger and Damasio's (1985)’s patient EVR (cerebral tumour removed), performed well on reasoning tasks would involve the CE, this suggests that the CE is intact but not fully functioning- therefore must have sub-components involved for each task respectively (CE is more complex than WMM suggests, which challenges the integrity of WMM)
Logie (1995):
Suggests that the visual cache stores information about the visual form and colour and that the inner scribe processes spatial and movement information. Klaur and Zhao 2004- supported this idea by asking P’s to carry out one of two primary tasks
A visual task- A spatial task
At the same time doing either
A spatial interference task- A visual interference task- No secondary task
They found that the performance of spatial tasks was much poorer for those doing spatial interference tasks than people doing the visual interference task and vice versa
PET Scans:
Studies using PET scans have also provided evidence for separate spatial and visual systems
There appears to be more activity in the left half of the brain for visual tasks but more in the right half during spatial tasks
Bunge et al (2000):
Used fMRI to see which parts of the brain were more active when participants were doing tasks (reading a sentence and recalling the final word in each sentence)
The same brain areas were active in either dual or single-task conditions but there was significantly more activation in the dual-task condition single-task
Indicating that increased demands were reflected in brain activity
LH (Trojano and Grossi):
Another patient LH, who had been involved in a road accident, performed better on spatial tasks than those involving visual imagery
This suggests separate visual and spatial systems
KF- showed STM works independent of LTM as he had no problem with long-term learning but some aspects of his immediate memory were impaired
LH- Performed better on spatial tasks than those involving visual imagery
Strengths of Working Memory Model:
It is very influential
There is a considerable amount of research evidence to support the model
Especially more recent studies show direct links between certain tasks and activation of parts of the brain that represent components of the WM model
Makes sense of a range of tasks i.e Reading- phonological loop
Navigation- Visual-spatial processing
Limitations:
WMM does not offer a complete understanding of how memory works- no LTm or SM (although acknowledges these components proposed by MSM)
The exact role played by the central executive remains unclear. The capacity of CE is still unclear
Also, WMM fails to account for musical memory because we can listen to instrumental music without impairing performance on other acoustic tasks
Evidence for Central Executive:
The Central Executive has a critical role in attention, planning and coordination, and is the most important and coordination, and is the most important and most flexible component of working memory
Baddeley himself accepts that this complexity makes it considerably harder to investigate
