Memory

Memory

1. EWT

   • Misleading information
     • Loftus and Palmer - showed film traffic accident to 45 student volunteers. Asked to write a brief description of what happened as a distracter task, then had to
   • Anxiety
   • Age

2. Memory improvement techniques

   • Method of loci
     • Pair objects from a list to places on a familiar walk/in a familiar surrounding. Form a mental image of this, then when come to recall, the places on the walk act as retrieval cues for the object that needs to be recalled
   • Narrative chaining
     • Create a story out of the list of words to link them together. When recall story, will say the words and so remember them.
   • Acronyms
     • Create a word from the starting letters of the words that need to be remembered. When come to recall, the letters act as retrieval cues for the words.
   • Face-name system
     • Link the person's name to a prominant feature on their face. This mental image acts as a retrieval cue for the person's name.
   • Acrostics
     • Create a sentence which is easy to remember with each word in the sentence beginning with the same letter as the words that need remembering. When come to recall, the words act as retrieval cues for the other set of words

3. Multi-store model

   • Memory is linear - information passes from store to store in order, and can't skip a store
     • Glanzer and Cunitz - primacy effect (words from beginning of list remembered as had moved to LTM) and recency effect (words from end of list remembered due to still being in STM). Been to different stores in order.
     • Glanzer and Cunitz - unusual words from middle of word list remembered (Von Rostorf effect) - seemed to have skipped straight to LTM from sensory, missing out STM.
   • There are separate stores for sensory, STM and LTM
     • Clive Wearing - damaged hippocampus from an illness, and his LTM was damaged (couldn't remember his wife entering room), but STM was fine
     • KF - STM was damaged, but LTM wasn't
     • Brain scans - for STM tasks, prefrontal cortex is active, and for LTM tasks, hippocampus is active
   • Sensory, STM and LTM are unitary stores - there is only one part to each
     • KF - had damaged verbal STM but not visual so must be separate stores
     • Clive wearing - damaged hippocampus and could still remember procedural LTMs (who his wife is, how to play piano etc) but not form episodic LTM (couldn't remember wife entering room) - suggests LTM is not unitary
   • Information is lost from STM by displacement
     • Glanzer and Cunitz - middle words from list couldn't be remembered but more recent words could be - they had displaced the middle words.
   • Rehearsal is key to information reaching LTM
     • Some words reach LTM without rehearsal - e.g. Glanzer and Cunitz found that unusual middle words from list were remembered (Von Rostorf effect) without them being rehearsed
     • Glanzer and Cunitz - words from middle of list not remembered due to lack of rehearsal
   • Information is lost from LTM by interference
   • Information is lost from sensory by decay

4. Working memory model

   • Central executive
     • Directs attention of its slave systems to the information entering STM
     • Acts as the boss/as a control system of STM
     • Limited capacity
     • Know little about the most important part of STM - for example, don't know the capacity of it, just that it is limited
     • Physiological evidence - brain scans seem to show that central executive is the pre-frontal cortex
     • Some critics believe that there is more than one part to central executive - patients have appeared to have only part of central executive functioning
   • Phonological loop
     • Phonological store
       • Processes verbal information when it enters STM - acts as the inner ear
     • Articulatory control process
       • Silently repeats the verbal information - acts as the inner voice
     • Stores verbal information - speaking, reading, learning word lists etc
     • Physiological evidence - brain scans seem to show that phonological loop is Wernicke's (for phonological store) and Broca's (for articulatory control) areas in brain
     • Doesn't explain why verbal tasks are not impaired by listening to instrumental music - maybe should include a store for non-speech auditory information
   • Visuo-spatial sketchpad
     • Visual and spatial information temporarily stored here
     • Forms images - the minds eye
     • Tasks such as recalling images, watching DVDs, mentally counting the number of an object somewhere
     • Physiological evidence - brain scans show that visuo-spatial sketchpad seems to be occipital lobe
     • Lieberman critisizes as blind people have excellent spatial awareness, even though they have never had visual information. However, model implies that all spatial information was once visual, but this must not be the case. Should separate them into two different components.
   • Episodic buffer
     • Added later to the model
     • How information from LTM is integrated with LTM,
   • STM is not unitary - there are different parts to it
     • Better at explaining some cases of amnesia than multi-store model. For example, KF had functioning STM for visual tasks but not for verbal, suggesting that there are separate stores for visual and verbal information
     • Supported by dual task experiments - better performance when complete two tasks with one being visual and one being verbal, than two verbal or two visual. Suggests that if goes to one store, gets overloaded and impairs performane but if goes to two separate stores, they are not overloaded and performance is better.
   • Only involves STM so not a comprehensive model of memory

5. Parts of memory

   • Sensory
     • Capacity - large
       • Sperling - participants told to focus on whole 12 character grid (shown for 50 milliseconds), then told to recall one of the three rows. Whichever row was chosen, recall was 3/4 (75%), so capacity must be large as all the information had gone in
     • Duration - iconic is 1/2 a second, echoic is 3 seconds, but people with photographic memories can hold up to 2-3 minutes of iconic
       • Sperling - participants told to either focus on whole 12 character grid or one of the three rows (shown for 50 milliseconds). People who had to recall one row got 3/4 (75%) but people who had to recall grid got 5/12 (42%), which suggests that duration is only short and the information decays while recalling.
     • Encoding - iconic (stored as images), echoic (stored as sounds) and haptic (stored as feelings)
   • STM
     • Capacity - 7+/-2 pieces of information
       • Miller - showed participants either dot patterns, trigrams or musical notes. Then had to recall - counted the number they could recall before making a mistake. Was mainly between 5 and 9.
     • Duration - 15-30 seconds
       • Peterson and Pererson - showed pariticpants trigrams, then had them count back in 3s (a distracter task to prevent rehearsal) for varying lengths of time - 3, 6, 9, 12, 15 and 18 seconds. Recall after 3 seconds was 78%, and was 6% accurate after 18 seconds, so some information remains in STM for 18 seconds.
     • Encoding - acoustic
       • Conrad - showed pariticipants a series of letters - either acoustically similar or acoustically disimilar - found that it was more difficult for participants to recall the acoustically similar letter. which suggests that encoding must be acoustic and the similar sounding letters got mixed up with eachother in the brain.
   • LTM
     • Duration - very long (years)
       • Bahrick - got participants who had either left school 15 years or 48 years ago. Had to recall the people from their year group. Used verbal recognition (shown names), visual recognition (showed pictures of them), or free recall (just remembered classmates with no prompts). Found that free recall was poor (60% for 15 years, 30% for 48), visual and verbal recognition was good (90% for 15 years, 70-80% for 48). Suggests that the information lasts for years, but prompts help to retrieve the information.
       • Inability to recall is due to difficulty retrieving information, not due to it not being there.
     • Encoding - semantic (according to meaning and understanding)
       • Baddeley - showed participants one of four word lists 4 times (rehearsal to send information to LTM) - either acoustically similar, acoustically disimillar, semantically similar or semantically similar. Recall was delayed for 20 minutes (completed an unrelated task) to ensure that the information had gone to LTM. Then had to sort these words into order, found that recall of semantically similar words was 55%, but for other lists it was 70-85%.
     • Capacity - impossible to test, but seems unlimited

6. Cognitive interview

   • Report everything
     • Recall everything that happened, even if it seems irrelevant or trivial. Recalling the small details may act as retrieval cues for other events that happened
   • Context reinstatement
     • Recall the context that events happened in - for example the weather, feelings and smells. Create a mental image of what happened, and the surroundings act as retrieval cues for other information
   • Change the perspective
     • Try to recall the event from the point of view of a person that was present at the time. Prevents own schemas from distorting memories
   • Recall in reverse order
     • Start from a key event, and work recall back from that point
   • Evaluation
     • It is effective - Kohnken et all carried out a meta analysis on 53 studies and found that C.I. obtains 34% more detail than standard interview
     • Godden and Baddeley got participants to learn a word list either on land or 15 feet underwater, then got them to recall either in the same environment as learning or in a different environment to learning. Found that recall was 30% better when in the same environment as learning, so supports context reinstatement
     • Gieselman reports that it is not suitable for children under the age of 8 - find the instructions confusing so would obtain less info than standard interview
     • Requires skilled and trained individuals - so more expensive to train people
     • More time consuming than standard interview
   • Enhancced cognitive interview
     • Felt there was lack of communication between interviewer and interviewee on normal cognitive interview which may affect recall accuracy. Encouraged to relax, asked fewer questions, given more instructions and adapted questions to individuals.
     • Fisher et al found that enhanced obtains 57.5% more info than standard compared to 34% for normal cognitive interview

Bilder und Dateien der Mindmap