MEMORY
Atkinson and Shiffrin’s multi store model of memory describes how information flows through the memory
system. →It suggests that memory is made up of three stores linked by processing.
Three stores of memory
Short-term memory (STM): The limited capacity memory store. In STM, coding is mainly acoustic (sounds),
capacity is between 5 and 9 and on average, duration is about 18 seconds.
Long-term memory (LTM): The permanent memory store. In LTM, coding is mainly semantic (meaning), it
has unlimited capacity and can store memories for up to a lifetime.
Sensory register: When the brain obtains information from the environment. 5 senses = 5 stores.
→ Iconic – visual
→ Echoic – sound
Prolonged rehearsal
Short term memory → Long term memory
Coding: The format in which
information is stored in the various
memory stores.
Capacity: The amount of
information that can be held in a
memory store.
Duration: The length of time
information can be held in memory.
Evaluation of multi-store model
✓ Gives a good understanding of the structure Oversimplified – STM & LTM don’t operate in a
and process of STM. single, uniform fashion → it’s far more complex
→ Allows for further research to expand on this as we may skip a process e.g., STM to get
model. through to the LTM.
✓ Credible and reliable – supported by case Too much emphasis on the role of rehearsal
studies such as Clive Wearing → it’s not the only way information moves
→ brain damage = no new memories between stores.
Research on coding: Alan Baddeley
Depending on the memory store, information in memory is stored differently → coding is
when information is converted between different forms.
o Conducted research by giving different lists of words to 4 groups of participants to
remember.
o 2 groups were acoustically (sound) similar & dissimilar, 2 groups were semantically
(meaning) similar & dissimilar.
o By recalling them from STM in the correct order, he found they did worse on acoustically similar words.
After 20 mins (LTM), semantically similar words did worse. → Shows how info is coded according to LTM
(semantically) & STM (acoustically).
(+) Provides clear evidence to support the different memory stores in the memory system – lab experiment
therefore can establish a cause and effect.
(+) Temporal validity – results haven’t changed over time, it only led to the multi-store model of memory.
(-) Artificial stimuli rather than meaningful material – this isn’t how info is coded in everyday tasks and life.
→ Limited application – meaningful material may mean semantic coding is used for STM as well as LTM.
Page | 39
, Research on capacity: Jacobs (1887)
→ Jacobs found how much information can be held by STM by measuring digital span.
o Study: Participants were given a set of words each time and had to recall these
aloud in correct order. Repeated by adding on more words until the participant
was recalling incorrectly → Shows the individual’s digital span.
o The mean span for digits was 9.3 items and letters were 7.3.
Evaluation
✓ High control over variables – can establish cause Lacks mundane realism & ecological validity –
and effect. conducted in a lab and not a realistic, daily task.
✓ Standardized procedures - easy to repeat study. Use of different ages reduces validity – memory
✓ Use of different ages can be used to generalize influenced by age not how many words.
to all members of society.
Span of memory and chunking: Miller (1956)
o Miller made observations of everyday practice.
o Found that things come in sevens e.g., 7 days of week, 7 deadly sins etc.
o The span (capacity) of STM is 7 ± 2.
o He also found people recall 5 words as easily as 5 letters.
o This happens through chunking – grouping sets of digits/letters into units/ chunks.
Evaluation
✓ Reliable - supported by Jacobs (1887) study Overestimated capacity of STM – research by
where participants recalled 7.3 letters and 9.3 Cowan shows it is 4 ± 1
words → Miller found span of STM is 7 ± 2 May be prone to bias – observation done and
✓ Controlled in a lab environment – use of only noticed factors supporting his research.
metronome to control EV of different pace of Jacobs also found age improves memory –
reading words/letters → high replicability Miller didn’t account for other factors like age.
Research on duration:
STM – Petersons (1959) LTM – Bahrick et al (1975)
o 24 students given cognitive test to recall a set of o 392 American PPTs varying in age undertook:
consonant letters + 3 digits to count back from. 1. Recognition test – 50 photos from high
o The 3 digits were given to stop the participants school yearbooks.
from mentally rehearsing the letters. 2. Free recall test – PPTs listed the names of
o The retention retrieval was varied e.g., some their graduating classes.
took 3 seconds and others took 18 seconds.
o They found after 3secs, the average recall was o After 15 years, recall for RT was 90% and
about 80%, after 18 secs, average recall was 3%. declined by 20% after 33 years.
o Their findings suggested that unless we have o Free recall test was less accurate as they
rehearsal (verbal or mental), our STM duration dropped to 30% after 48 years.
is about 18 seconds. o Shows some material remains in LTM for
lifetime.
Stores STM LTM Sensory Memory
Acoustic (sound) Semantic (meaning)
Coding Sense specific
Baddeley Baddeley
Limited
Capacity Unlimited Very large
Jacobs and Miller
Limited to 18 seconds Lifetime/years
Duration Up to 0.5 seconds.
Petersons Bahrick et al
Page | 40
Atkinson and Shiffrin’s multi store model of memory describes how information flows through the memory
system. →It suggests that memory is made up of three stores linked by processing.
Three stores of memory
Short-term memory (STM): The limited capacity memory store. In STM, coding is mainly acoustic (sounds),
capacity is between 5 and 9 and on average, duration is about 18 seconds.
Long-term memory (LTM): The permanent memory store. In LTM, coding is mainly semantic (meaning), it
has unlimited capacity and can store memories for up to a lifetime.
Sensory register: When the brain obtains information from the environment. 5 senses = 5 stores.
→ Iconic – visual
→ Echoic – sound
Prolonged rehearsal
Short term memory → Long term memory
Coding: The format in which
information is stored in the various
memory stores.
Capacity: The amount of
information that can be held in a
memory store.
Duration: The length of time
information can be held in memory.
Evaluation of multi-store model
✓ Gives a good understanding of the structure Oversimplified – STM & LTM don’t operate in a
and process of STM. single, uniform fashion → it’s far more complex
→ Allows for further research to expand on this as we may skip a process e.g., STM to get
model. through to the LTM.
✓ Credible and reliable – supported by case Too much emphasis on the role of rehearsal
studies such as Clive Wearing → it’s not the only way information moves
→ brain damage = no new memories between stores.
Research on coding: Alan Baddeley
Depending on the memory store, information in memory is stored differently → coding is
when information is converted between different forms.
o Conducted research by giving different lists of words to 4 groups of participants to
remember.
o 2 groups were acoustically (sound) similar & dissimilar, 2 groups were semantically
(meaning) similar & dissimilar.
o By recalling them from STM in the correct order, he found they did worse on acoustically similar words.
After 20 mins (LTM), semantically similar words did worse. → Shows how info is coded according to LTM
(semantically) & STM (acoustically).
(+) Provides clear evidence to support the different memory stores in the memory system – lab experiment
therefore can establish a cause and effect.
(+) Temporal validity – results haven’t changed over time, it only led to the multi-store model of memory.
(-) Artificial stimuli rather than meaningful material – this isn’t how info is coded in everyday tasks and life.
→ Limited application – meaningful material may mean semantic coding is used for STM as well as LTM.
Page | 39
, Research on capacity: Jacobs (1887)
→ Jacobs found how much information can be held by STM by measuring digital span.
o Study: Participants were given a set of words each time and had to recall these
aloud in correct order. Repeated by adding on more words until the participant
was recalling incorrectly → Shows the individual’s digital span.
o The mean span for digits was 9.3 items and letters were 7.3.
Evaluation
✓ High control over variables – can establish cause Lacks mundane realism & ecological validity –
and effect. conducted in a lab and not a realistic, daily task.
✓ Standardized procedures - easy to repeat study. Use of different ages reduces validity – memory
✓ Use of different ages can be used to generalize influenced by age not how many words.
to all members of society.
Span of memory and chunking: Miller (1956)
o Miller made observations of everyday practice.
o Found that things come in sevens e.g., 7 days of week, 7 deadly sins etc.
o The span (capacity) of STM is 7 ± 2.
o He also found people recall 5 words as easily as 5 letters.
o This happens through chunking – grouping sets of digits/letters into units/ chunks.
Evaluation
✓ Reliable - supported by Jacobs (1887) study Overestimated capacity of STM – research by
where participants recalled 7.3 letters and 9.3 Cowan shows it is 4 ± 1
words → Miller found span of STM is 7 ± 2 May be prone to bias – observation done and
✓ Controlled in a lab environment – use of only noticed factors supporting his research.
metronome to control EV of different pace of Jacobs also found age improves memory –
reading words/letters → high replicability Miller didn’t account for other factors like age.
Research on duration:
STM – Petersons (1959) LTM – Bahrick et al (1975)
o 24 students given cognitive test to recall a set of o 392 American PPTs varying in age undertook:
consonant letters + 3 digits to count back from. 1. Recognition test – 50 photos from high
o The 3 digits were given to stop the participants school yearbooks.
from mentally rehearsing the letters. 2. Free recall test – PPTs listed the names of
o The retention retrieval was varied e.g., some their graduating classes.
took 3 seconds and others took 18 seconds.
o They found after 3secs, the average recall was o After 15 years, recall for RT was 90% and
about 80%, after 18 secs, average recall was 3%. declined by 20% after 33 years.
o Their findings suggested that unless we have o Free recall test was less accurate as they
rehearsal (verbal or mental), our STM duration dropped to 30% after 48 years.
is about 18 seconds. o Shows some material remains in LTM for
lifetime.
Stores STM LTM Sensory Memory
Acoustic (sound) Semantic (meaning)
Coding Sense specific
Baddeley Baddeley
Limited
Capacity Unlimited Very large
Jacobs and Miller
Limited to 18 seconds Lifetime/years
Duration Up to 0.5 seconds.
Petersons Bahrick et al
Page | 40
