-
What is the influential model proposed by Atkinson (and who else?) on how memory might be organised?
- Multi-store model
- by Atkinson and Shiffrin (1968)
- Info detected by modality-specific sensory stores, and if attended, enters STM, and this is transferred to LTM only if that info is rehearsed. (First, rehearsal was described as a maintenance procedure, but later suggested it could be elaborative).
-
Who's theory influenced Atkinson & Shiffrin's multi-store model?
- William James (1890)
- Distinction between primary and secondary memory stores
- Primary: info remaining in consciousness after perception (the psychological present)
- Secondary: info about events that have left consciousness (psychological past)
-
List the 2 sensory stores for which there have been studies on and the researchers who studied them.
- Iconic store: (sensory store for visual info)
- Sperling (1960)
- Echoic store: (for audiotry modality)
- Treisman (1964)
-
Describe the iconic store experiment.
- Sperling presented participants with 12 letters in a grid array for 50ms
- They could only report 4 or 5 letters but claimed they saw more
- In another condition - partial report condition - they only report top, middle or bottom row - however, the cue for which line is given after the array dissapears
- Able to report most of the letters from the line, but only if delay between removal of array and presentation of the prompt was 1second or less
- This suggests that info in iconic storage decays in less than a second.
-
Describe the echoic store experiment - how long does info in this store last?
- Treisman - present participants with auditory message to one ear and asked them to repeat that message while ignoring message from other ear
- BUT if the message in both ears are same but started at different time - participants only noticed they were the same if they started within 2 seconds of each other.
- Suggests persistence of unattended info in echoic store is around 2 secs, otherwise info decays.
-
What is the capacity of short-term memory? Study?
- Atkinson & Shiffrin suggested ST store contains info currently held 'in mind' and is of limited capacity.
- George Miller (1956)
- Participants try to recall digit strings
- Typically - could recall strings correctly up to a length of 7±2 digits
- Similar result for letters and words
- Miller suggested STM holds around 7 integrated units of info (or 'chunks')
-
What was an additional study conducted by another person to qualify Miller's finding?
- Simon (1974)
- found that span is less with larger chunks (eg. 8-word phrases than smaller chunks - 1 syllable words)
-
How can info in ST store be retained, according to Atkinson & Shiffrin? What study investigated this claim?
- Info can be retained in STM by rehearsing it (either out loud or sub-vocally). The longer an item is held in ST store, the greater likelihood of long-term storage.
- Rundus (1971)
- Present list of 20 words and asked them to rehearse them out loud
- In subsequent recall test, the more frequently a word has been rehearsed, the more likely it was to be recalled.
- (Apart from the last words on the list which were consistently recalled because of recency effect)
-
Describe a study which tried to identify the duration of short-term memory.
- Peterson & Peterson (1959)
- Participants remember 3-letter stimulus for few seconds while counting backwards in threes
- Ability to recall diminished rapidly as the retention interval became longer
- Suggesting info decays from STM within a matter of seconds.
-
However, what was the criticism given for Peterson & Peterson's study? Who said this and what was their study?
- Waugh & Norman (1965)
- Counting backwards task was a source of interference, and thus P&P's study was due to inteference, not decay.
- They presented 16 digits in 2 conditions (one presented faster than the other)
- Decay hypothesis will predict that the recall will be better with the fast condition - because STm only lsats for a few seconds.
- However, results showed that recall was unaffected by speeds of presentation.
- Suggests STM forgetting is due to interference from exposure to additional info, not passage of time.
-
What can recency and primacy effects tell us about whether STM and LTM are in separate stores or not (don't mention HM etc for now).
- Atkinson & Shiffrin attributed recency effect to the last few items still being present in ST store at end of list presentation
- Glanzer & Cunitz (1966)
- showed that recency effect can be eliminated if people were asked to count backwards prior to recall, supporting the link with STM (limited capacity etc)
- Atkinson & Shiffrin also proposed primacy effect was due to the first few items being stored in LTM.
- Glanzer & Cunitz
- showed that primacy effect was unaffected by counting backward after list presentation, suggesting that the info at the beginning and end were stored in different stores and that STM and LTM are indeed in separate stores.
-
How have studies of patients with brain lesions further supported the separation of STM and LTM?
- HM: MTL damage and poor LTM but good STM - normal digit span
- preserved recency effect but much reduced primacy effect
- (reported by Scoville & Milner (1957)
- KF: parieto-occipital lobe damage and good LTM but bad STM
- primacy effect preserved, but no recency effect
- (reported by Shallice & Warrington (1970)
- NB - just check this with KATE - cos she said KF case doesn't support it...
-
What are the criticisms for Atkinson & Shiffrin's multi-store model?
- 1. According to model, processing in ST store (rehearsal) is required for encoding into LTM. However, KF's case with defective STM but preserved LTM (and primacy effect) provides evidence against this.
- 2. Model predicts the longer an item is held in ST store, the greater likelihood of LT storage. Although Rundus did show that longer rehearsal for words does correlate with LT retention, other factors such as depth of processing (Craik & Tulving) are far more important for determining whether info will subsequently be remembered after delay of few mins or more.
- 3. Assumption that ST and LT stores are unitary, operating in single, uniform way.
- However, KF - worse ST memory for auditory letters and digits than for visual sitmli (Warrington & Shallice) - suggests there may be distinct memory stores for different kinds of material.
- ALSO: Baddeley & Hitch (1974) - found that aduitory rehearsal of digits did not affect number of errors made in concurrent grammatical reasoning task.
- Suggests distinciton between auditory-verbal ST store and central info processing system involved in reasoning -> so the working memory model was formed.
-
In an attempt to address the limitations of the multi-store model, who proposed the new concept and what was it?
- Working Memory Model
- Baddeley & Hitch
- Comprises 4 primary components
- Visuospatial sketchpad: short-term storage of spatial and visual info
- Phonological loop: short-term storage of speech-based form
- Episodic buffer: limited capacity system for integrating phonological and visuospatial representations with info from LT memory
- Central exectuvie: modality-free processing system - coordinates operation of other systems for performing demanding cognitive tasks
-
What is the big overall difference in emphasis between short term memory and working memory?
- STM is about short storage
- WM is about working with that STM (manipulation of the info held in STM) - assumes a more active use of STM
|
|
