Psychology: Models of Memory

Key Terms:

• Short term memory: a temporary store where samll amount of info can be kept for brief periods- Fragile and info can be easily lost
• long term memory:permenant store-limitless amounts of info can be stored for long periods of time
• Capacity:the amount of info that can be held in memory at one time
• Duration:the length of time that memories can be held
• Encoding:the way in which the info is represented in memory store eg. by sound/meaning or image
• Sensory memory:a set of limited capacity,modality-spevific stores that hold info for only a brief period.

What is a model?

• Model:theories based on information-processing systems are models
• Reseachers use flowcharts as a model becasue designing a computer programme that mimics the brain is too complcated and difficult.

• Structural model-focuses on the storage components of memory system
• A & S thought of memory as a flow of information through information processing system -which is divided into stages that the info passes from one store to another in a fixed sequence.
• Stages are constrained by:
• CAPACITY
• DURATION
• ENCODING

• A & S proposed that info enters from the environment and is first registered on the sensory memory store- staying for very short time before decaying or being passed to STM store.
• STM store very small capacity and duration and memory traces held are quite fragile which can be lost if not rehearsed.
• Items are held usually as sounds in the STM store
• Once rehearsed it passes to LTM store-unlimited capacity and duraction but info can be lost due to damage/interferance or decay.

Control Processes needed to manipulate and transform the info as it passes from one store to the other.

• Encoding
• Retrieval
• Rehersal 

Stimuli from the external environment is first registered with the sensory memory -Tiny duraction: few seconds

A & S prooposed separate sensory store:

• Iconic store: visual
• Echoic store: auditory input
• Haptic store:  tactile

Baddeley ( 1988)

• believed that the visual sensory store is used to allow us to experience visual information in a smooth, continous flow eg. watching a cartoon - it is many frames fired quickly but we experince it as a continous visual.
• The senory memory needs to hold the one image long enough before the next is presented.
• Also used to large amounts of incming info from the environment in order to prevent overloading - the store holds the info for a few milliseconds while it is scanned and to see if it should be passed on as useful for more processing -COCKTAIL PARTY PHENOMENON

• Aim:To investigate the capacity of sensory (iconic) memory

• Procedure: Sperling used a chart containing 3 rows of letter ,which he displayed for 50 milliseconds to his ppt.Ppt were asked to immediatley recall as many letters as possible but found they could only recall 4/5 but they reported having been aware of more letters even though they were unable to recall them.Sperling then changed his procedures (partial report technique) by training ppt to distinguish between 3 tones. He then showed the chart for the same time but this time layed one tone as soon as the chart disappeared.Asked to recall the top letters for high note, bottom letters for low and so on. 
• Findings: Under the changed circumstance the ppt were abel to recall on average 3 items from which ever row had been cued by tone.It must be remembered that the ppt did  oy know which row would be asked to recall until after the diplay had disapeared. This suggests that at that stage they would have been able to recall 3 items from which ever row had been cued by tone as a the whole lot of letters was availabe in ther ionic memory.Sperling estimated that ppt had seen 9-10 out of possible 12
• Conclusion: The no. of items recalled in any one row is x by 3 (no. of rows) = no. of items seen. The reson being why they can only recall 3 is that all the items fade in the time it takes to report it back these 3 items

Methodoligcal issues:

• high level of control
• easily replicated
• stimuli artificial and may not reflect  the use of memory in everyday life
• lacks ecological validity.

Ethical Issues:

• consent is needed from ppts
• Debrief afterwards needed as ppts may have felt frustrated and debriefing is needed to reassure them that they are within the normal range.
• some ppt may have been placed in a stressful environment.

• The encoding (type of information) is either iconic or echoic 
• The capacity is said to be vast but the duration however is only 0.25 to 2 seconds.
• Info remains in the sensory memory for less than 2 seconds-probs even shoter in the iconic store.
• Information from sensory memory is passed to short term memory by attention i.e. taking notice of something.info is relatively unprocessed.
• info is passively registered and only a small amount is passed to the next store.
• separte stores for the differet senses.

Key words:

• Displacment: from of forgetting where the items hel in the STM are pused out before they pass to the LTM to make room for incoming items
• Interference: form of forgetting where info in the LTM is confused with similar info.
• Free Recall: way of testing memory wher ppt can recall a list of items in any order.

A & S model of memory displays a clear distinction between LTM and STM.

• Duration
• Capacity
• Coding- how info is stored.
• Forgetting

LTM                                                          STM

• Capacity:Unlimited                              Very limited -approx 7 items
• Duration:Ulimeited                               Very limited
• Encoding: mainly semantic                   Mainly acoustic
• Forgetting: mainly interference              Maily displacment 

Short term memory

• 30 second and having a capacity of 7 +/- 2 (seven plus or minus two, i.e. 5-9) items. And the encoding is mainly visual and acoustic.
• However, we can often hold much larger amounts of information in a process known as chunking whereby the information is consolidated into 'chunks' of information.

Free recall is often used to investigate the diifernece in LTM and STM- a graph is plotted of the relationship between where the word appeared and the liklihood of it being recalled-produces a characteristic serial postion curve.

• words recalled at the end are recalled best-recency effect ( still circulating the STM)
• resonable recall of words at the begining- primacy effect ( words have been rehearsed and so have passed to LTM)
• words in the middle poorly recalled-asymptote (little time for rehersal and there is displacment)

• Aims: To determine the capacity of STM and LTM
• Procedure: Ppt presented with a list of words one at a time and then their free recall was tested. Condition 1: ppt asked to recall words immediatley after. Condition 2: ppts given a distractor task- asked to count back in 3s for 30 seconds before recalling the words.
• Findings: Condition 1: expected serial position curve found. Condition 2: distractor task disrupted the recency effect and words form the last part were not recalled well.
• Conclusion: The task of counting backwards in 3s had displaced the last few words in the list from the fragil STM store but had bot affected earleir words as they had already been passed into the LTM.

Methodological issues:

• highly controlled lab exp.
• easily replicated with same results.
• lacks ecological validity as it is artificial and may not mimic  how memory works in real life.
• Several trials taken by ppts and averages calculated to avoid unrepresentitive results.

Ethical issues: conesent and briefing needed 

• found other fcators that affcet the of  one part serial position curve and not the other.

 A no. of factors affected the primacy effect but not the recency effect

• rate of presentation ( slower the presentation the better the performance)
• age of ppt ( younger ppt remeber more then elderly)
• familiartity( more familair words are better remembered)
• All of the above are examples of functional dissociation and provide good support for LTM and STm.

Neuropsychological Evidence:

• people who have suffered brain damage - loss of memory is usually selective ie.  may affect STM and not LTM.
• Case study:  (Milner 1966) man HM -suffered form severe epilepsy and had brain surgery to remove part of his temp. lobe and hippocampus- alleviated his epilepsy but left him with severe memory deficits.
• Able to recall his childhood (LTM) but unable to remeber event about 10 years before the surgery and he was unable to learn new info.
• Able to remeber 6 no. in the oredr the had been presented- STM intact
• Repeatedly read the same magazine over again without realizing + unable to recognise his psychologist-LTM no longer effective as he is unable to lay down new memories/or unable to retrieve them.

Case study: Shallice and Warrington (1970) reported the case of KF - young man sustained brain injuries after a motorcycle accident. Intact LTM as he was able to learn new info and recall stored info but the STM was affected so that he had a recency effect of only one.

• Another source of evidence-study of people with Alzheimers-severe memory impairment- researchers interested in chemical neurotransmitters.
• Ptients with Alzhiemers have a low level of the neurtansmitter acetycholine -suggests that it may have an impotant function inmemory

Drachman and Sahakian (1979) investigated this by administering a drug that blocks the action of acetycholine, then gave ppts various memory tests on LTM or STM. found that experimental group performed normal on the STM tests but more poorly on the LTM tasks-suggesting that LTM and STM work as separte stores.

• PETs and fMRI scans provided more suppot for the existence fo 2 separate stores
• Squire et al (1992) found that the hippocampus is active in LTM tasks and the pre-frontal cortex is activated in STM tasks..

• LTM has an unlimited capacity-possibel yo lose info through decay and interference but not due to capacity limitations.

STM:

• Limited space capacity- you can only hold and manipulate small no. of itmes at one time.
• Jacobs (1887) devised the digit soan technique:a way of measuring the capacity of STM-ppt have to repeat back strings of digits in order of the presentation, the no. of digits in the string gradually increases untill recall is not possible.
• Jacobs found that on average ppt could recall on 7 digits in the serial call: a way of testing STM where ppt are required to recall items in the order they were presented.
• Miller  (1956) " The magical no.7 + or -" also believed that memory span is determined by 'chunks' of info we can hold than the no of letters/no. Chunking allows for a rhythm that makes info easier to recall.
• Memory span can be increased by chunking
• Miller's term chunk has been criticised for beign to vague .

• Simon (1974) found that the span as measured in chunks depends on the amount of info in a chunk.
• Procedure: Experimented using s immediate serial recall of 1,2 and 3 syllable words for 2 and 8 word phrases.
• Findings:span in chunks was less with larger chunks i.e. 8 word phrases than with smaller chunks.

Glanzer and Razel (1974)

• used the recency effect rather then the digit span as a measure of capacity of STM.
•  Findings: found that recency effect was 2.2 items when the material was single unrelated words but it increased to 1.5 sentances when unfamiliar sentences and 2.2 proverbs when they were familiar.

Factors:

•  Influence of LTM -Cowan ( 2000) - thought that Miller overestimated the no.of chunks that can be held in the STM. Thought that performance on span tasks is affected by rehersal and LTM and does not reflect the actual capacity STM -estimated that capacity was 4 chunks when the factors were controlled.
• Bower and Winzenz (1969)-found that repitition of digit strings became easier for ppt to recall - suggests the strings were rehearsed and stored in the LTM which temporaliy increase the capacity of STM.
• Example:Short term memory -draws on LTM in order to increase STM .If asked to look at the table behind you quickly hear the names and then recall their names there is no way that you are soley relying on short term memory you are also drawing form the LTM.
• Reading aloud: digit span increases if ppt read the digits aloud instead of reading them subvocally (Baddley 1999-says this is because the digits are shotened briefly in the echoic store-strenghthens memory trace.)
• Pronounciation time:many studies have shown thta the capacity of STM is due to time constraints rather thne structural limitations-Naveh-Benjimen and Ayers(1986) digit span and pronounciation with English and Arabic speakers- found that digit span was less for arabic speakers as it took longer for them to pronounce.

• Hitch,Halliday and Littler( 1984) found that the immediate memory span of young kids was related to the time it took for them to articulate the words.
• Schweickert and Boruff (1986) found that ppt can recall the no. of items that can be articulated in 1.5 seconds.
• Individual differencesn :people who are highyl anxious seem to have a shorter memory span ( Macloed and Donnellan 1993)

Examples of STM: 

• examples of short-term memory in action are the holding on to a piece of information temporarily in order to complete a task (e.g. “carrying over” a number in a subtraction sum, or remembering a persuasive argument until another person finishes talking), and simultaneous translation (where the interpreter must store information in one language while orally translating it into another). What is actually held in short-term memory, though, is not complete concepts,but rather links or pointers (such as words, for example) which the brain can flesh out from it's other accumulated knowledge.