multi store model of memory

sperling 1960

• investigate iconic information

• Ps were flashed a grid for milliseconds

• full report = recall rate was 40%

• partial report = recall rate was 80%

• SR has large capacity but low duration

• strengths = highly controlled lab experiment with precise timing

• limitations = low ecological validity

jacobs 1887

• investigated the STM capacity

• digit span technique

• Ps read aloud a sequence of numbers at 0.5 sec intervals

• if recalled correctly, a digit is added until they fail
  mean span for digits = 9.3
  mean span for letter = 7.3

• STM has a capacity of 5-9 units

• strength = reliable methodology replicated many times

• limitations = lacks mundane realism

miller 1956

• investigated STM capacity

• meta reviewed existing research

• ppl consistently reached limits around 7+/-2 units of information

• STM capacity increased when info is put into meaningful chunks

• strengths= explained how organising on overcomes capacity limits

• limitations when reviewed it was found that the pacity is closer to 3-5 so miller overestimated

peterson and peterson 1959

• investigated STM duration

• uni students were given a consonant trigram and a 3 digit number

• counted backwards immedietly after to avoid verbal rehearsal

• retention intervals increased = recall dropped

3 secs = 80%

6 secs = 50%

18 secs = less than 10%

• duration of STM is around 18-30 secs

• strength = extremely rigid protocol

• limitation = very little control over variable

baddeley 1966 (STM)

• to identify the STM coding

• Ps were presented with = semantically similar/dissimilar, acoustically similar/dissimilar

• recall was worse for acustically similar words

• recall for semanticlly similar words had minimal impact

• STM codes acoustically = similar words caused confusion

• strength = established clear distinction betwen ST & LTM

• limitations = lacks mundane realism

baddeley 1966 (LTM)

• identify LTM coding

• same as STM - but with a 20 min distraction before recall

• recall for semantically similar words was worse

• acoustic similarity had minimal impact

• LTM codes semantically

• strength = evidence that STM & LTM are seperate stores

• limitations = 20 min is relatively short time

bahrick et al. 1975

	duration of LTM
	392 ex grads aged 17-74 tested using their HS yr books
	50 photos = some from yr book
	recall as many names as pssible
	within 15yrs of grad = 60% accuracy
	48 yrs = 30%
	strength = high ecological validity
	extraneous variables cant be controlled

MSM

• by atkinson and shiffrin 1968

• memory flows through a series of 3 unitary stores in a linear sequence

• info is processed passively

how does memory flow

• SR gathers unprocessed sensory info

• if attention is paid it passes into the STM

• unattended info is forgotten

• info can be kept in the STM thru rehearsal

• without rehearsal it decays

• if rehearsed enough it transfers to the LTM

• to use memories stored in the LTM you must retrieve it to STM

SR

• coding = senses (iconic, echoic)

• capacity = very large

• duration = 0.5 secs

STM

• coding = primarily acoustic

• capacity = 5-9 units of info

• duration = 18-30 secs

LTM

• coding = primarily semantic

• capacity = unlimited

• duration = unlimited