Multi-store Model of Memory (Atkinson and Shiffrin, 1968)

What does the Glanzer and Cunitz (1966) study support?

Existence of separate STM and LTM stores.

Three stores of the MSM

1.Sensory register
2. Short-term memory (STM)
3. Long-term memory (LTM)

Encoding

The way information is changed so it can be stored in memory.
-- Turning the info into something we can understand.

Three main ways in which information can be encoded (changed)

1. Acoustic (sound)
2. Visual (picture)
3. Semantic (meaning)

Encoding, capacity and duration in relation to short-term memory (STM)

Encoding: acoustic (sound)
Capacity: around 7 items (7 +/- 2)
Duration: around 18 minutes

Encoding, capacity and duration in relation to long-term memory (LTM)

Encoding: mainly semantic (meaning)
Capacity: potentially unlimited
Duration: few minutes to lifetime

Glanzer and Cunitz (1966): Conclusion

- When presented with word list ➡️ ppts. tend to remember the first and last words
- More likely to forget words in the middle of the list (serial position)
- Words early on in the list ➡️ put into LTM (primary effect) ➡️ person has time to rehearse the word
- Words at the end of the list ➡️ went into STM (recency effect)

Miller (1956)

Magic Number 7+/-2

Miller (1956): Chunking

- Miller didn't specify how much info can be held in each slot
- Chunking info = more info stored in STM

What does Miller (1956) support?

Capacity of STM

Peterson and Peterson (1959): Aim

- Investigate the duration of STM
- Provide empirical evidence for MSM

Peterson and Peterson (1959): Procedure

- Lab experiment ➡️ 24 psych students had to recall trigrams
- Asked to count backwards in 3s ➡️ until a red light appeared ➡️ known as Brown-Peterson technique ➡️ prevented rehearsal
- Ppts. asked to recall trigrams after intervals of 3, 6, 9, 12, 15, or 18 seconds

What does the Peterson and Peterson (1959) study support?

Duration of STM

Bahrick et al. (1975)

Duration of LTM

Bahrick et al. (1975): Aim

To investigate very long-term memory (VLTM)

Bahrick et al. (1975): Conclusion

- Accurate recall is possible for time periods up to 50 years

What does the Bahrick et al. (1975) study support?

Duration of LTM

Tulving (1972)

PSE in LTM (Weakness)

Tulving (1972): Distinctions of LTM

1. Procedural memory
2. Semantic memory
3. Episodic memory

Strength of MSM: Case study of Clive Wearing

- Contract virus that caused severe amnesia (memory loss)
- Could only remember info for 20-30 seconds ➡️ able to recall info from his past ➡️ e.g. his wife's name
- Wearing was unable to transfer info from STM to LTM ➡️ able to recall info successfully
- Supports idea that memories are formed by passing ➡️ one to another in linear fashion ➡️ damage to any part of MSM can cause impairment

Strength of MSM: Supporting psychological studies

- Glanzer and Cunitz (1966): existence of separate memory stores
- Miller (1959): limited capacity of 7+/-2 chunks of info in STM
- Peterson and Peterson (1959): limited duration in STM, approx. 20 seconds
- Bahrick (1975): lifetime duration in LTM
- Suggests that the model is an accurate representation of human memory

Strength of MSM: Evidence from brain scans

- Active areas of the brain when performing STM tasks: hippocampus + subiculum
- Hippocampus also involved in transferring ST memories into LT memories
- Active areas of the brain when performing LTM tasks: motor cortex
- Suggests that different brain regions are responsible for different MSM components
- Supports the idea that memory is made up of discrete stores

Weakness of MSM: Rehearsal

- Rehearsal ➡️ considered too simple of an explanation for transfer of info from STM to LTM
- Schmolck ➡️ rehearsal is not effective if memory loss is from brain damage

Weakness of MSM: Oversimplified

- Model is oversimplified ➡️ when MSM suggested both STM and LTM operate in a single, uniform fashion ➡️ we know this isn't the case
- Baddeley and Hitch (1974) ➡️ Working Memory Model ➡️ showed STM is more than just one simple unitary store ➡️ comprised of different components ➡️ e.g. central executive, visuo-spatial sketchpad

Multi-Store Model of Memory

- Proposed by Atkinson and Shiffrin (1968)
- Structural model
- Proposed that memory consisted of three stores: sensory register, short-term memory (STM) and long-term memory (LTM)
- Info passes from store to store in a linear way
- Described as an information processing model (like a computer, with an input, process and output)

Capacity

How much information can be stored.
-- How much your memory can hold.

Duration

The period of time information can last in the memory stores.
-- How long it can hold for (length of time).

Glanzer and Cunitz (1966)

Primary/Recency Effect

Glanzer and Cunitz (1966): Procedure

- Presented two groups of ppts. with identical word lists
- Group 1: recalled immediately after presentation; Group 2: recalled after waiting 30 seconds
- Group 2 had to count backwards in 3s (Brown-Peterson technique) ➡️ prevented rehearsal, caused recency effect to disappear
- Both groups could free recall words in any order

Miller (1956): Capacity of STM

- Provides evidence for capacity of STM
- Most adults can store 5-9 items in their STM
- Miller thought that memory can only hold 7+/-2 ➡️ STM only had a certain number of 'slots' where items can be stored

Bahrick et al. (1975): Results

- Within 15 years = 90% accurate in identifying names and faces
- Within 48 years = 80% verbal, 70% visual
- Free recall was worse
- Free recall ➡️ 15 years = 60%, 48 years = 30%

Glanzer and Cunitz (1966): Results

- Words at the end of the list ➡️ only remembered if first and tested immediately
- Delaying recall by 30 seconds ➡️ prevented recency effect

Peterson and Peterson (1959): Results

- The longer the interval delay = the less trigrams recalled
- After 3 seconds ➡️ ppts. able to recall 80% of trigrams
After 18 seconds ➡️ less than 10% of trigrams were recalled correctly

Peterson and Peterson (1959): Conclusion

- STM has a limited duration when rehearsal is prevented
- Info is lost from STM by trace decay
- STM is different from LTM in terms of duration ➡️ supports MSM

Bahrick et al. (1975): Procedure

- Sample consisting of nearly 400 ppts. ➡️ age 17-74
- Free recall test ➡️ ppts. tried to remember names of people in a graduate class
- Photo recognition test ➡️ consisting of 50 pictures
- Name recognition test ➡️ ex-school friends

Tulving (1972): Procedural memory

- Responsible for knowing how to do things ➡️ memory of motor skills
- E.g. driving, swimming, riding a bike

Tulving (1972): Semantic memory

- Responsible for storing info about the world ➡️ knowledge about word meanings + general knowledge
- E.g. a song that reminds you of an event, or a perfume

Tulving (1972): Episodic memory

- Responsible for storing info about events (i.e. episodes) that we have experienced in our lives
- E.g. birthdays, or first day of school

Weakness of MSM: Tulving (1972)

- Different types of LTM: procedural, semantic, and episodic memory
- MSM is too simple of an explanation for LTM

Weakness of MSM: Weaknesses in supporting studies

- Experiments that give evidence for the model ➡️ use artificial tasks ➡️ might not be valid