miller - research using digit span technique - asked to read series of digits that get progressively longer - asked to immediately repeat in right order - recalled on average 5-9 items
new info displaces old info - capacity increased by chunking
duration of STM
18-30 seconds
peterson & peterson - 24 students listened to consonant trigram - heard random 3 digit number & counted backwards in 3’s to prevent rehearsal of trigram - asked to recall trigram after 3, 6, 9, 12, 15, or 18 seconds - highest recall after 3 seconds (90%) - only 2% after 18 seconds
coding in STM
acoustic (by sound)
baddeley - ppts presented with list of words that were acoustically similar and asked for immediate recall they made more errors than presented with acoustically different words
capacity of LTM
unlimited capacity
anokhin - number of possible neural connections in the brain is 1 followed by 10.5 million kilometres of noughts - no human exists who can use all their potential of their brain
duration of LTM
unlimited duration
bahrick - 392 ppts aged 17-74 to name old classmates from high school (free recall) - then given 50 photos and asked if they could recognise classmates (recognition test)
15 years after graduation - free recall was 60% accurate - dropped to 30% after 48 years.
recognition was 90% accurate after 15 years - dropped to 70% after 48 years
coding in LTM
mainly semantic
baddeley - if ppts presented with semantically similar words and asked for recall 20 minutes later they made more errors than semantically different words
multi store model of memory
involves sensory register:
• coding : senses
• capacity: very large
• duration: milliseconds
short term memory:
• coding: acoustic
• capacity: 5-9 items
• duration: 18-30 seconds
long term memory:
• coding: semantic
• capacity: unlimited
• duration: lifetime
how does info move from sensory register to stm
through attention
how does info move from stm to ltm
elaborative rehearsal
how does info stay in stm
maintenance rehearsal
evaluation of multi store model pt 1
• support from KF - suffered brain damage in a motorcycle accident - STM was damaged for verbal STM but unaffected for visual STM, LTM remained intact
• evidence suggests STM is not unitary - multi store model argues there is only one type of STM however verb STM was damaged but visual STM was unaffected
• support from murdock - found no matter how many words a person is shown & asked to recall, items at beginning have greater recall than the middle and words at end also have greater recall than middle (primacy-recency effect)
evaluation of multi store model pt 2
• evidence suggests LTM is not unitary - argues only one type of LTM but research criticised this - clive wearing lost episodic memory but kept procedural memory
• support for sensory register from Sperling - shown letter chart for 50 milliseconds and asked to recall as many letters as possible - could only remember 3 letters on average. however when a high, medium & low tone was played immediately after, could recall 3 letters from any row
what makes up the working memory model
central executive, visuospatial sketchpad, phonological loop, episodic buffer
central executive
coding: any modality e.g visual, acoustic
capacity: no storage capacity
• the supervisory component - directs attention to incoming info and starts the rehearsal process
• delegate’s info to slave systems
visuospatial sketchpad
coding: visual and/or spatial
capacity: limited
• known as the ‘inner eye’
phonological loop
coding: acoustic/auditory
capacity: limited
• acts as temporary storage for acoustic/auditory info - deals with written & spoken
• preserves the order that info arrives
what are the 2 parts of the phonological loop
phonological store - ‘inner ear’ rehearses sounds
& articulatory process - ‘inner voice’ holds and silently repeats words we are preparing to speak
episodic buffer
coding: any modality
capacity: limited
• added by baddeley
• storage component of the central executive - integrates info from all stored
• links working memory to long term memory
evaluation of working memory model pt 1
• supporting evidence - KF - damaged STM for verbal info - visual STM remained intact
• support from dual task research - can explain why ppts struggle to process 2 similar tasks at once - difference tasks use different slave systems - baddeley found ppts had more difficulty tracking a light and describing letter F (both visual) than doing 1 verbal and 1 visual task
evaluation of working memory model pt 2
• support from brain scans - PET scans shown that different areas of the brain are used when doing visual and verbal tasks
• real life applications - allows psychologists to target certain kinds of memory - e.g educational psychologists developed memory training specific to verbal memory to improve students learning
evaluation of working memory model pt 3
• fails to account for musical memory - can’t explain why people are able to process music differently to other sounds - Berz ppts could listen to instrumental music without imparing performance on other acoustic tasks - suggests may be more than one acoustic store
what are the 3 types of long term memory
episodic, semantic, procedural
episodic memory
memory of personal experiences, events or episodes from a specific point in time
• stored with reference to context e.g where it happened, who was involved etc
• requires conscious effort to recall
examples: birthdays, first day of school, family holiday
semantic memory
factual knowledge of meaningful information about the world. may be concrete (facts) or abstract (concepts in maths or language)
e.g what a word means, what an orange tastes like
• requires conscious effort to recall
procedural memory
memory of skills, actions & how to do things i.e muscle memories
• automatic as the result of repeated practice
e.g how to tie your shoe, ride a bike or read
• doesn’t require conscious effort to recall
evaluation of types of LTM pt 1
• support from PM - professional cellist who suffered brain damage - episodic and semantic affected - couldn’t remember musical facts but could read and play music
• support from clive wearing - lost episodic memory (couldn’t remember his wedding) but still had procedural memory (could play piano)
• issues with evidence from brain damaged patients - relatively rare and may not apply to everyone e.g clive wearing LTM may function differently to others - low external validity
evaluation of types of LTM pt 2
• may ignore a 4th type - research has identified another type of LTM known as ‘priming’ (where implicit memories influence responses to a stimulus) - e.g given a list of words including yellow & asked to name a fruit, more likely to answer banana than if not ‘primed‘
• real life applications - allows psychologists to target specific types of memory - belleville able to develop memory training to improve episodic memory
what are the two explanations for forgetting?
interference & retrieval failure
what is interference
when one memory disrupts our ability to recall another
most likely when 2 memories are similar
what are the two types of interference
retroactive & proactive
retroactive interference
when newly learned information disrupts our recall of older memories
e.g learning spanish word for thank you then the italian word for thank you and struggling to recall the spanish
proactive interference
when old memories disrupt our ability to learn & recall new information
e.g learning spanish word for thank you then italian word for thank you and struggling to recall italian word
evaluation of interference pt 1
• support from postman - ppts had to remember list of paired words - experimental group given a second list where paired word was different - recall of control group was more accurate than experimental group - learning new info interfered with ability to recall old info
• support - found when ppts learned a word list with 100% accuracy then given new lists the more similar the new list was the poorer the recall of old list - even strong memories can be disrupted by new learning if similar
evaluation of interference pt 2
• support from schmidt - 211 ppts from dutch school 11-79 years old given a map of the neighbourhood with street names replaced with numbers - asked to remember as many as possible - positive correlation between number of times ppts moved house & street names forgotten
• support for proactive - meta-analysis - found when ppts have to learn word lists they don’t learn the lists given later as well as lists given earlier on - earlier lists disrupted learning of later lists
evaluation of interference pt 3
• supporting research uses meaningless materials - materials used do not reflect the kinds of info we use our memory for in real life - e.g remembering word lists doesnt reflect real life memory - may not be representative of real life forgetting
what is retrieval failure
when a memory is available but we cannot access it because we do not have the necessary triggers/cues
• memories are coded with context (e.g where we were) & state (e.g how we felt) cues
context dependant forgetting
inability to retrieve a memory due to an absence of environmental triggers
e.g learning a fact in one classroom & struggling to recall in another
state dependent forgetting
inability to retrieve a memory due to being in a different mental state
e.g learning info whilst upset and struggling to remember when happier
evaluation of retrieval failure pt 1
• supporting evidence - if ppts learn info in one room and then were tested they forgot more when tested in a different room by different instructor - scuba divers learnt word list either on land or underwater struggled recall when opposite
• real life applications - used to reduce forgetting in real life - thinking of the room where you did individual learning as effective as being there
evaluation of retrieval failure pt 2
• supporting evidence - male volunteers learned list of words while drunk or sober - forgot most when opposite - applies to intoxication as well as emotion
• ppts hid money while high on marijuana - less able to find when not high than when high again
• criticism from meta analysis - retrieval failure only occurred for word lists and the influence of context for real life information was largely eliminated