Memory KASARA

Memory is the process of reaining learned information and accessing it when needed. Process is ENCODING --> STORAGE -->  RETRIEVAL     

Encoding is putting the information into a meaningful form.    

Storage is keeping the information within the memory system till its needed    

Retrieval is recovering information stored in the memory system when required

Sensory Memory is where the information from the five senses goes to. Short Term Memory holds small amounts of information. Long Term Memory is where information is permanently stored until its needed. The STM and LTM are unitary stores operating like a single unit with no separate compartments.

They all have different capacity and duration.

Duration is the amount of time information's held in a memory store before its lost to decay.

Capacity is the amount of information held before its lost due to displacement.

Capacity: Sensory has a limited amount, e.g. visual is 9-10 items. E.g. Jacobs did a digit span test to determin STM and on average 7 +- 2 was the capacity for STM. Lacks MR and EV. If you chunk the digits it helps remember more information. The LTM has an infinite capacity, its unmeasureable.

Encoding: E.g. Baddeley gave 4 lists of words. List A sound same, List B diff sounds, List C mean same, List D mean diff. He found the STM encodes acoustically and LTM encodes semantically. For the LTM he had a 20min delay and found C and D had a difference.

Duration: Sensory duration is a few seconds unless attentions paid to it. E.g. Peterson used nonsense trigrams to test STM. Asked to count back from 100 in threes to prevent maintance rehersal. After 3 secs 80% were right, after 18secs 10% was right. So info in STM lasts less than 30 seconds without rehearsal. Cant test duration of LTM as you cant always recall info, may suffer from tip of the tounge syndrome. Info in LTM is never lost, its just retrieval failure. E.g. Bahrick found 90% recall faces and names after 34 years from school. 80% accuracy for names and 40% faces after 48 years. So LTM durations infinite.

By Atkinson and Shiffrin.

Sensory Memory: Receives info from senses and enters SM. Retains unprocessed impressions of info with seperate store for each sense. Limited capactiy. Short duration unless attentions paid to it and info is lost due to decay or its never processed. Small fraction of info  received by SM is attended to, and futher processed in STM.

Short Term Memory: Info is encoded acoustically. Its an active memory store with currently thought about info. Temp store for info from SM and can be recalled. Capacity of 7 +-2 and infos displaced to new info. Duration of less than 30 secs, without maintanance rehearsal its lost. Elaborate rehersal means goes to LTM permanently.

Long Term Memory: Info from STM is semantically encoded. Capacity and duration is infinite. Info is never lost when in the LTM. People never forget skills learnt throughout their childhood. A permanent store holding limitless amounts for long periods. Info retrieved to STM. If we cant access we need retrieval cues to remove the retrieval failure.

Neurobiological Evidence: Shallice & Warrington, study of KF who haad accident, reducing STM to 1-2 digits but LTM was unaffected, supports idea of distinct stores but KF verbal was poor, visual was fine so suggest more than one STM. Scoville attempted treat epilespy by remove area of brain. This meant people were unable to encode new LT memories but previous were unaffected. Supports seperate store idea but LTM retrieval is by STM and STM damaged so shouldnt be able to retrieve.  CASE STUDIES are in depth but only suitable for one person. Cant replicate for ethics and cant generalise. High EV and MR.

Laboratory: Murdock did free recall test and found the serial position effect, words at beggining and end were remembered best. Beggining words go to LTM, primary effect & end words stay in STM, recency effect. Support seperate stores. Peterson, Baddeley & Bahrick support idea of seperate stores. LAB well controlled, lack MR and EV.

Natural: Berkenan & Baddeley found people were unaware of changes to the BBC Radio wavelengths despite being told 1000 times day, so rehersal doesnt always transfer to LTM.

General: MSM was influential and inspired research. Developed WMM which lead to more depth. MSM is also oversimplified, assuming theres one STM & LTM but research shows more than one STM and LTM. e.g.non/verbal, then episodic, semantic and procedural.

Central Executive: filter that determins what info is attended to. Processes info in all sensory forms and directs info to other slave systems. Limited capacity, trys to attain balance between task when attention is needed. Theres a two way flow of info between CE & slave systems.

Phonological Loop: Slave system, temp store dealing with auditory sensory info, similar to rehersal system of MSM. Capacity is info spoken in last 2 secs. Confusion can occur with acoustic words. Has sub parts. P-Store: Inner ear, holds speech in memory for last 1-2 secs and holds info directly from sensory register. Articularly Loop: Inner voice, verbal rehersal system and holds by articulating sub vocaly info we want to retain. Primary acoustic store: Baddely and Lewis added to this model afteer was discovered. Holds recent speech/sound.

Visuo-Spatial Sketchpad: Inner eye, holds and manipulates visual and spatial info from LTM/senses. Temp store with a visual and spatial relationship. Divided to visual cache for info in colour/meaning and inner scribe for spatial relationship. Transfer info to VC and CE.

Episodic Buffer: third slave, general store to operate propely. Deal with process specifc info. No storage capacity. Limited store intergrating from all systems.

Neurobiological: KF has a short term memory of 1-2 after his accident. Poor verbal and good visual, supports PL and the VSS as part of WMM for STM. Smith found PET scans showed that the left side brain was visual tasks and right was spatial tasks, supporting STM has diff components. Visual cache and inner scribe. CASE STUDY rich in detail, unethical to replicate, cant generalise.

Lab: Baddeley and Hitch gave dual task, use CE by reasoning and use PL by reading. They could do this simultaneously which supports the idea of seperate STM components that can do more than one task at a time. Baddely also gave participants a list of words, long or short. They were asked to recall and more short were recalled. This supports PL as it can hold as many items said in 1-2 secs. LAB EXP are well controlled but lack MR and EV.

General: WMM has practical applications, helps with development of reading and understanding of dyslexia. Influential to cognitive psychologists, call STM, working memory. WMM has stimulated more research and findings. CE is too simplistic, doesnt explain full role when its most important component.

Retrieval Cues: Act as prompts to trigger recall. External (context-dependent): Physical environment where material was learnt, context - state dependent cues facilitate recall. E.g. Abernethy asked to recall instrcutors and rooms, they rememebered instructor in that room as memory facilitated. Internal (state-dependent): Psychological state affects memory. Its facilitated if similar internal environment. E.g. Overton asked drunk and sober people to recall info. Recalled info better in the state they were told.

Mnemonics: any structured technique which assists learning info and recalling accurately. Verbal: Acronym: Word made up of first letter from words in a phrase. e.g. NASA        Acrostic: Sentence where first letter of word relates to something e.g. Richard Of York Gave Battle In Vein.  Rhyme: Collection of words with a rhythm when said togehter. e.g. DIFFICULTY.  Chunking: Organising large sets of data into small amounts of info. e.g. 123568 = 12 35 68

Visual: Method of Loci: associating info with locations    Keyword method: associating a word with a viusal image    Peg Word Technique: Associating a list of words to remember with rhyme with number up to ten e.g. One bun two stew three tree

Organisation: Mnemoics work as help us to organise the info we need to recall. Brain remembers by building associations with previously known info, called memory hooks. E.g. Bower gave participants 112 words. If organised to heirarchies could remember them 3x better.

Active Processing: MSM-rehearsal enough but has to be elaborate for it to be effecting. You must think about it and encode semantically. Set up associations so there are several routes. E.g. Woodrow found learning techniques aided memory but practise didnt. E.g. Moms found footie fans recalled more scores than non footie as actively processed info, so active and elaborate is effective.

Dual Processing Techniques: Pavio claimed words and images are separately processed. Patient who damanged temporal lobes only process words so concrete words(words than are made to image) has 2 retrieval routes. Agrees with WMM about VSS and PL. E.g. Bower gave participants 100 cards, 2 unrelated words. If linked to image, rememerebed 58% otherwise 45%. E.g. Beni found mnemonics usefull as keyword is good for verbal presented info.

Its the evidence supplied by people who see an event based on memory of event. Can be crime scene or criminal. Juries are heavily influenced, 75% people in USA convicted.

Reconstructive Memory:memories are reconstructed over time and influenced by schemas(ready made expectations based on prev exp). Helps fill in gaps in our memory. Has an implication for EWT as may be based on schema not actual recall of memory. Turkey & Brewer: participants gave schemas on bank robberies, e.g. male, balaclava, dark clothes etc. They were shown a tape of a stimulated robbery. They had good memory recall of info that fitted with schema but info irrelevant to schemas werent good, e.g. colour of car. So info which isnt consistent with schemas isnt encoded into memory. Yuille said EWT can be accurate and over time. Interviewed EW of a crime where one shot and one injured. Found the police interviews(3month before)& current ones were accurate. High EV.

Leading Questions: Questions are phrased so the lead a witness to give certain answer. Can have misleading info to distort memory. Loftus & Zanni 45 students shown a crash and estimated speed. Diffe verbs used contact, smash. Asked week later if saw glass. 32% smashed saw broken glass compared to 12%. Contact 31mph, smash 40mph. Loftus & Zanni, showed particpants an accident, 20mins later asked if saw the or a broken headlight. 17% said the and 7% a.

Children are more willing to accept innacurate info given by adults than other adults due to authority. Have a reliable memory when event is serious. Accuracy of answers depend on the questions. Younger you are more vunerable. E.g. Flin, children and adults were asked about incident witnessed one day & five months after. Recall was similar after one day but significantly worse after 5 months. Delay in memory is imporant as court happens long time after crime. Children more susceptible to memory reconstruction. E.g. Lindsay did meta analysis on children making identifications. Children less than 5 make false identifications. Those under 13 more like than adults to make a choice even when perpetrator wasnt present. Kids have distortions due to limits on processing and retaining.

Elder people have cognitive abilities which diminish with age, so accuract decreases with age. They have less accurate and detailed EWT and recall than middle aged so prone to misleading information. E.g. Loftus et al. Elderly make false identifications and poorer at recalling specific details. Men are prone to memory distortions from misleading info. E.g. Memon said when there is a delay between the incident and recall of 35 mins thers no diff between adults and elder. If it was a week delay then elderly gave less accurate EWT. So elderly more susceptible to memory reconstruction.

Anxiety is a state of apprehension, uncertainty and fear resulting from a threatening situation. Impairs psychological and physical functioning. It impairs accuracy and detailed recall. E.g. Peters took heart rate of students in an innoculation and two mins after it was done. There were high heart rate who had high anxiety and 31% of them recognised the nurse compared to 59% of low. Anxiety inhibited their ability to remember. Had high EV but didnt control extraneous variables. E.g. Christianson surveyed 100 people who witnessed robbery. Victims had more accurate recall, suggesting there is a relationship but its not clear.

Weapon Focus Effect: people obersving the crime pay attention to the aspect which poses most threat to them. They can describe this in great detail. E.g. Loftus, particpants overheard an arugment between 2 people, furniture thrown in uncontrolled. Uncontrolled lead to one emerging carrying a bloody letter opener and controlled had a greasy pen. 49% remembered face of greasy and 33% of bloody letter opener. So it had threat. The participants were decieved and they didnt give consent, cause psychological harm, high EV and experimental realism so can generalise. E.g. Valentine meta anaylsis of 300 identifications. WFE had no effect on correct identification so sometimes it doesnt occur.

Fisher studied real police interviewsd and found faults. Witnessess were inturrupted and not allowed to exand. This was the standard interview.

Gieselman developed CI to obtain more accurate info. STAGE ONE: CONTEXT REINSTATEMENT, they were asked to recreate the crime scene internally and externally. Ask general questions here. STAGE TWO: REPORT EVERYTHING, they were asked to report everything they could recall even if in different order. STAGE THREE: RECALL FROM CHANGED PERSPECTIVE, put theirselves in someonelses shoes and recall in that view. STAGE FOUR: RECALL IN REVERSE ORDER, recall back to front or start with an aspect and work backwards.

There were additional guidelines like minimise distractions, ask open ended questions, pause after each response, avoid interuption and judgement, adapt language and listen.

Gieselman showed a stimulated crime then tested recall by standard, cognitive and hypnosis. Cognitive gave most information. Fisher trained policemen in Miami to use cognitive and 46% increase in amount of info given and 90% info was accurate. Its time consuming and a lot of resources needed. Koeknken found produced more innacurate info.