Memory

Memory

1. Short and long term memory

   • SLT
     • Duration: up to 20 seconds
       • LLOYD: used 24 p's tested individually over 8 trials to remember a trigram (THX 512) and asked to count from 6, 9, 12, 15 & 18 seconds. Found 3: 90%, 9: 20%, 18: 2%
     • capacity: 5 - 9 items
       • JACOBS: showed p's a string of numbers and asked them to remember as many as they could. Found average digit span was 5-9 items.
       • MILLER suggets span is actually 7 +/- 2
       • SIMON found size of chunks affects memory- 8 word phrases are harder to remember than 8 smaller ones (e.g. one syllable words)
     • coding: acoustically
       • BADDELEY: used acoustically similar words that were semantically different e.g. cat, cab, cad and different acoustically but similar semantically e.g. big, huge, tall. Found p's struggled to remember acoustically similar words for ST
         • low ecological validity - does not reflect true to life situations as nobody would have to remember a random set of one syllable words.
         • BRANDIMOTE suggests coding is more complicated; found STM uses visual coding too.
     • JACOBS suggested individual differences affects chunks remembered; 8 y/o = 6.6 chucks, 19 y/o = 8.6 suggesting an increase in brain capacity.
   • LTM
     • duration: up to a lifetime
       • BAHRICK: tested memory 400 p's (age 17-74) on identifying classmates faces. Found 90% after 15 yrs, 70% after 48 yrs. free recall 60% at 15 yrs and 30% at 48.
     • capacity: unlimited
     • coding: semantically
       • BADDELEY: "" Found p's struggled to remember semantically similar words.
         • WICKENS found LTM uses semantic coding too.

2. The multi store memory model

   • BEARDSLEY & SQUIRE: PET and fRMI scans showed different areas of the brain for STM and LTM when p's did different tasks. [STM - prefrontal lobe, LTM - hippocampus]
   • GLAZNER & CUNITZ: the serial position effect. showed p's could recall words first and last from a list, suggesting 2 distinct areas of memory for STM and LTM supporting the rehearsal process of TWM. first words had been rehearsed, last were still in STM
   • It is an over simplification and suggests that there is only one area for STM and LTM
   • SHALLIC & WARRINGTON: patient KF had brain damage from a motorbike accident, which caused STM problems for verbal information but not visual information suggesting that STM has more than one part (MSM is too simplistic)
   • SCOVILLE & MILNER: patient HM- had his hippocampus removed to help his epilepsy. His STM was fine but he could no longer make new LTM. suggesting you need the hippocampus for LTM process.
   • MSM does not account for interest- why is it that you can reread information over and over and to remember it but you can read something that interests you once and remember it. This suggests that there are other factors (like interest) that affect memory

3. The working memory model

   • central executive: directs attention to particular tasks, limited capacity and cannot store data.
   • visuospatial sketch pad: used when planning a spacial task, where visual/spatial data is stored, visual = what things look like, spacial = physical relationship between things
   • episodic buffer: extra storage that integrates info from CE to VSS & PL, maintains the sequencing.
   • phonological loop: limited capacity, auditory information
   • BADDELEY: found participants made for more mistakes when completing tasks from the same slave system then using different ones e.g. reading silently and repeating 'the' compared to tracking an object and repeating 'the' showing you can do tasks that use different slave systems but not two tasks that have to compete to use the same slave system.
     • study lacks ecological validity!!
   • SHALLICE & WARRINGTON: patient KF had trouble remembering auditory information more than visual information- supporting that memory has different sections.
   • SHEPPARD & FENG: given a net of a cube and found that time taken to work out if the arrows would meet was similar to time taken to actually build the cube. proves we must have a visuospatial part.
   • ESLINGER: patient EVR performed well in reasoning tasks but could not make decisions (took hours to decide what to have for lunch) suggesting there are different parts of the central executive.
     • limited research on the central executive (supported by patient EVR)
   • doesn't account for being able to listen to music and read, as both use the phonological loop
   • can't generalise case studies

4. Types of long term memory

   • episodic
     • concerned with personal experiences, time and place, surroundings including specific detail, context and emotion.
     • e.g. first day of school, a family holiday
   • semantic
     • knowledge about the world that is shared by everyone and isn't personal, could include languages, behaviour, maths or social customs.
     • e.g. 2+2=4, the capital of England is London and what a rabbit looks like.
   • procedural
     • concerned with skills, remembering how to do them without having to think about it, acquired through repetition and practice.
     • e.g. knowing how to tie a shoe lace, read, drive, swim, write etc.
   • Evidence from brain scans: episodic - hippocampus, temporal lobe near hippocampus, frontal lobe. semantic - temporal lobe, procedural - cerebellum
   • Patient HM had the ability to recall LTM but not make them is a SIMPLIFICATION as he could, still form procedural memories but not episodic or semantic.
   • Evidence of separate parts to LTM:
   • Difficult to distinguish whether it is possible to form semantic memories independently from episodic
   • case studies are hard to determine exact brain location until the patient is dead- most work is done on living case studies. Therefore it is difficult to determine which part of the brain damage causes the defects, it may just be acting as a relay station as malfunction on the relay station would impair performance.

5. Explanations for forgetting: interference

   • proactive interference
     • when an older memory interferes with a newer memory e.g. a teacher struggling to learn new names because she's learnt so many names in the past.
     • UNDERWOOD: found that pps struggled to learn more lists of words the more lists they had already learnt.
   • retroactive interference
     • when a newer memory interferes with an older one e.g. a teacher has learnt so many new names that she struggles to remember the old ones.
     • MÜLLER: found pps performance was less good if they had been given a task in-between initial learning and recall.
   • retro = "old" interferes
   • SCHMITT: asked 211 pps aged 11-79 to recall 48 street names near the school they attended. He found a positive association between the number of times the pp had moved house outside the neighbourhood to the number or street names forgotten.
   • BADDELY & HITCH: asked pps who had played in rugby games to remember as many of the teams they had played against as possible. He found that forgetting was due more to the number of games played than the time passed supporting interference rather than decay theory.
   • It only explains forgetting when two pieces of information are similar; cannot explain most real-life situations
   • most research is lab and lacks mundane realism. and ecological validity. Remembering word pairs does not reflect true to life scenarios
   • although research proves interference is a true effect; it does not explain how the cognitive process works.

6. Explanations for forgetting: retrieval failure

   • the encoding specificity principle
     • memory is most effective if information that was present at encoding is also available at the time of retrieval
       • BADDELEY suggests its impossible to test for as its circular- if a stimulus leads to the retrieval of a memory then it must have been encoded in th memory. If it does not, then it can't have been encoded but its impossible to test for an item that hasn't been encoded so it cannot be proved
     • TULVING: pps had to learn 48 words (12 categories) e.g. fruit- apple, fruit-pear. some pps had free-recall, some category titles. Found 40% of words when free-recall and 60% with cues of titles.
       • lacks ecological validity
     • Nairn criticised that cues do not cause retrieval; they are just associated with retrieval. He suggests the relationship between the two is a mere correlation.
   • context-dependant forgetting
     • people remember better if the context is the same as when they learnt the information
     • ALBERNETHY: arranged a group of students to be tested each week after being taught with different testing conditions. Some changed teacher/room, some remained the same. She found that those in same room with same teacher performed the best, and that the better students were least affected by this change.
     • BADDELEY: test scuba divers who learnt words wither on land or underwater, and recalled them either underwater or on land. He found the highest recall occurred when the initial context stayed the same as the recall context.
     • too simplistic- there are more factors that determine if you can recall something or not other than context.
     • SMITH showed that thinking of the room is just as effective as being in the room during recall
     • outshining hypothesis suggests that a cue's effectiveness is reduced by the presence of a better cue (meaningful material)
   • state-dependant forgetting
     • the mental state you're in at the time of learning can also act as a cue
     • GOODWIN: asked male volunteers to remember a list of words when they were either drunk or sober, and found that pps remembered better if they recalled in the state that they learnt in.
   • retrieval failure: occurs due to the absence of cues, that the memory is there but not accessible without a cue.

7. Accuracy of eye witness testimony: misleading information

   • leading questions
     • LOFTUS & PALMER: 45 students shown films of traffic incidents, asked questions: "About how fast were the cars going when they ______?" [smashed, collided, bumped, hit, contacted]
       • Findings: smashed = 40.8, collided = 39.3, bumped = 38.1, hit = 34, contradicted = 31.8 showing that leading questions affect the response given by the pps
       • may be bias- response may have just been altered rather than information being stored before its stored.
     • LOFTUS & PALMER: eps returned one week later and were asked if they recall seeing glass. (there wasn't any)
       • findings: smashed = 16/50, hit = 7/50, control = 6/50 proving leading questions affect memory.
     • can't be generalised because all of the pps are students and research has proved that elderly people are less accurate and driving experiences will be less limited.
     • although the experiment had mundane realism- it doesn't entirely reflect real life situations as people who witnessed a crash would have an emotional involvement.
   • post-event discussion

8. accuracy of eye witness testimony: anxiety

9. Improving the accuracy of eyewitness testimony: the cognitive interview

Bilder und Dateien der Mindmap