Memory studies

STM duration

Peterson+Peterson

Gave ppts a consonant trigram, asked them to count backwards in 3s to stop repetition (from 3secs to 18secs), then recall the trigram

After 3 secs, ppts could rememeber about 90% of trigrams

After 9 secs, 20%

After 18 secs, 2%

Conclusion: STM duration is 18-30 seconds without rehearsal

Duration of LTM

Bahrick

Tested 392 American graduates on recalling former classmates, either free recall or using photo recognition

Photo - 15 yrs: 90%, 48 yrs: 70%

Free recall - 15 yrs: 60%, 48 yrs: 30%

Conclusion: LTM may last up to a lifetime

Capacity of STM

Jacobs

Used digit span method - researcher reads out 4 digits, ppt has to recall in correct order, number of digits increases until they get it wrong

Mean span for digits - 9.3 items

Mean span for letters - 7.3 items

Conclusion: capacity of STM is 7+-2

Coding of memory

Baddeley

Gave different list of 4 words to ppts (acoustically similar, acoustically dissimilar, semantically similar, semantically dissimilar), had to recall in correct order either immediately or after 20 mins

Immediately - worse with acoustically similar

After 20m- worse with semantically similar

Conclusion: STM - acoustically coded, LTM - semantically coded

Multi-store model recency/primacy effects

Glanzer and Cunitz

Tested ppts ability to remember a list of words and found they were more likely to remember words at the beginning and end of the list

KF case study

Shallice and Warrington

KF had amnesia, his STM for digits was poor when read aloud but better when he read them to himself.

Suggests MSM is wrong in claiming there is only one STM store for processing diff types of info

He also had poor ability for auditory info but processed visual info normally. Phonological loop damaged but VSS intact

Type of rehearsal

Craik and Watkins

Elaborative rehearsal needed for long-term storage (linking info to existing knowledge), means that info can be transferred to LTM without prolonged rehearsal

Dual-task performance

Baddeley

Ppts had to follow a light spot with a pointer while carrying out either a verbal or visual task

Much harder to do visual task

Conclusion: cannot do 2 visual or 2 verbal tasks at the same time as they compete for the same sub-system

Effect of similarity on interference

McGeoch + Mcdonald

Studied retroactive interference, ppts had to learn list of 10 words until remembered with 100% accuracy, then had to learn new list (synonyms, antonyms, unrelated words, consonant syllables, 3 digit numbers, no new list)

Worst recall - synonyms

Interference effects in everyday life

Baddeley + Hitch

Asked rugby players to recall the names of the teams they played against in one season, some players missed matches due to injury

Players who played the most games had the poorest recall

Cues help interference

Tulving + Psotka

Gave ppts lists of words in categories, one list at a time, recall - 70% for first list and became worse as they learned each list, given cued recall test (told name of categories) and recall rose again to 70%

Context-dependent forgetting

Godden and Baddeley

Studies deep-sea divers, had to learn list of words either underwater on land and then recall on either

Accurate recall was 40% lower in non-matching conditions (external cues at learning different from ones at recall)

State-dependent forgetting

Carter + Cassaday

Gave antihistamine drugs to ppts as they have a drowsy effect, had to learn lists of words then recall info

Memory significantly worse in non-matching conditions

Leading questions

Loftus + Palmer

Had 45 ppts watch a clip of car accident and guess the speed of the car, asked the critical question using different verbs (smashed, hit, bumped, contacted, collided)

Contacted - mean speed of 31.8mph

Smashed- mean speed of 40.5mph

Post-event discussion

Gabbert

Studies ppts in pairs, watched video of same crime from different perspectives so they could see things the other couldn’t, then discussed what they saw before doing recall test

71% recalled aspects they hadn’t seen

Control group with no discussion - 0%

Post-event discussion altering EWT

Skagerberg + Wright

Showed ppts clips of mugging, in one the hair was dark brown, light brown in other, ppts discussed before reporting what they saw

Reported a blend of the two - ‘medium brown’

Anxiety negative effect on recall

Johnson + Scott

Ppts believed they were doing lab study, low-anxiety condition heard casual convo and saw man walk past carrying a pen, high-anxiety condition heard argument and saw man carrying bloody knife

Low-anxiety condition- 49% correctly identified man

High-anxiety condition - 33%

Anxiety positive effect on recall

Yuille + Cutshall

Interviewed 13 witnesses of shooting in gun shop 4-5 months after the incident. Compared interviews with original police ones, also asked to rate their stress levels at time of incident

Those who reported highest stress levels were most accurate (88% compared to 75%)

Unusualness causes weapon focus

Pickel

Conducted experiment using scissors, handgun, wallet or raw chicken as items in hairdressing salon video. EWT accuracy significantly poorer in high unusualness conditions

Effectiveness of cognitive interview

Kohnken

Meta analysis of 55 studies comparing CI and standard police interview

CI gave average 41% increase in accurate info (also found increased amount of inaccurate info recalled)

Only 4 studies showed no difference between the interview types

Some CI elements more useful

Milne + Bull

Each of the techniques produced more info than standard police interview, combination of report everything and reinstate context produced better recall than any other elements/combination