Case studies

Sperling study

grid of letter (3 rows of 4) flashed up for 50 milliseconds, either whole report or partial report conditions

Sperling whole report condition and results

participants had to try recall all 12 letters, on average 42% recalled (4-5 letters), other letters initially remembered had faded before they could be properly stored

Sperling partial report condition and results

a tone sounded after letters taken away corresponding to a certain row they had to recall. on average 75% recalled (3 out of 4 letters), shows most/all information was available and stored immediately before quickly fading

Jacobs procedure

researcher read out 4 digits, participant recalled them. then read out 5 digits… continued until digits could no longer all be recalled in the right order

Jacobs findings

mean span for digits was 9.3, mean span for letters was 7.3

who did a study on digit span capacity of STM?

Jacobs

positive evaluations of Jacobs study

it has been replicated, his study is very old but has been validated by more modern studies

negative evaluations of Jacobs study

his study is very old and may have lacked adequate controls e.g. some participants’ digit spans might have been underestimated because they were distracted (confounding variable)

Who did the study on chunking capacity of STM?

Miller

Miller procedure

made observations of everyday practice,

Miller findings

noticed things come in 7s (days of the week, deadly sins), decided capacity of STM was 7±2 chunks, also found it’s easier to recall five words than five letter- called chunking

Miller positive evaluation

none

Miller negative evaluation

he may have overestimated STM capacity, Cowan reviewed other research and concluded STM capacity is 4±1 chunks, suggests the lower end of Millers estimate is more accurate

Who did a study on STM duration?

Peterson + Peterson

Peterson & Peterson procedure

tested 24 students in 8 trials. at each trial the student was given a consonant syllable and a 3 digit number. They counted backwards from a number until told to stop and recall the consonant and 3 numbers. Stopped at 3,6,9,12,15, or 18 seconds, varying at each trial

Peterson & Peterson findings

after 3 seconds average recall was 80%, after 18 seconds average recall was 3%. suggested STM duration was about 18 seconds without rehearsal.

positive evaluation of Peterson and Peterson’s study

study is not completely irrelevant because we do sometimes try and remember fairly meaningless materials e.g. phone numbers

negative evaluation of Peterson and Peterson’s study

stimulus material was artificial: recalling consonant syllables does not reflect most everyday memory activities- lacked external validity (ecological validity)

who did a study of duration of LTM?

Bahrick

Bahrick LTM duration procedure

• 392 american participants between 17 and 74

• high school yearbooks

• photo recognition test: 50 photos, some from the yearbook, asked to pick out the people they knew in highschool

• free recall: asked to recall all the names of their graduating class

  
  

Bahrick LTM duration findings

photo recognition:

• participants tested within 15 years of graduation= 90% accurate

• participants tested after 48 years of graduation= 70% accurate

free recall:

• after 15 years= 60% accurate

• after 48 years= 30% accurate

Bahrick positive evaluation

high external validity due to investigating meaningful memories. other studies on LTM with meaningless pictures to remember had lower recall rates. Bahrick has a more ‘real’ estimate of LTM duration

Who did a study on STM coding?

Conrad

Conrad procedure

• letters flashed quickly on a screen

• participants asked to record them in order

• first set were acoustically similar, second dissimilar

Conrad findings

more remembered acoustically dissimilar set, suggests STM acoustically coded despite items presented visually, similar meanings don’t interfere with STM

Who did a study on STM+LTM coding?

Baddeley

Baddeley’s study on coding procedure procedure

different lists of words given to four groups of participants

Group 1: acoustically similar

Group 2: acoustically dissimilar

Group 3: semantically similar

Group 4: semantically dissimilar

Baddeley’s study on coding findings

when recalled immediately (from STM) acoustically similar words were worst remembered

after 20 minutes (from LTM) semantically similar words were worst

STM coded acoustically, LTM coded semantically

positive evaluation of Baddeley’s study on coding

it identified a clear difference between two memory stores, his findings are mostly still correct today, an important step in understanding the memory system- led to MSM

negative evaluation of Baddeley’s study on coding

it used artificial stimuli instead of meaningful material- could not tell us about coding in every day life e.g. when processing meaningful information people could use semantic coding even for STM, study findings could have limited application

Who had a damaged LTM but fine STM

HM- Henry Molaison

who studied HM

Milner

Why did HM require brain surgery

he cracked his skull and starting having seizures, which was thought to be epilepsy

What part of HM’s brain was removed

hippocampus

what happened to HM after his surgery

seizures disappeared, couldn’t form episodic memories, performed as normal on STM tests

who could only remember non-verbal information

KF

information about KF

his STM for digits was very poor when they were read out loud to him, but recall was much better when he read them to himself- shows there must be different stores in the STM

who studied KF

Shallice and Warrington

who did studies on dual tasks

Baddeley et al.

Baddeley et al. dual tasks findings

participants who did visual and verbal tasks at the same time had similar performances to when the tasks were carried out separately, when both tasks were visual or verbal performance declined substantially

who couldn’t make new episodic memories

Clive Wearing

what happened to Clive Wearing

he got a viral infection that attacked is brain and damaged his hippocampus, he could recognise his wife and remember how to play piano and sing, but couldn’t remember anything from a few minutes before

who discovered interference is worst when the memories are similar

McGeoch and McDonald

McGeoch and McDonald procedure

studied retroactive interference by changing the similarity between two sets of materials. Participants had to learn a list of 10 words until they could fully remember them. Then they learned a new list. 6 different types of new lists: synonyms, antonyms, unrelated words, consonant syllables

McGeoch and McDonald findings and conclusions

the most similar material (synonyms) produced the worst recall of the first list, numbers had the bets recall (apart from control group). this shows interference is strongest when the memories are similar

Baddeley and Hitch rugby players procedure

interviewed rugby players, asked them to recall names of teams they had played in the season, players all recalled teams from the same time interval (over one season) but the number of games varied

who studied retroactive interference

Baddeley and Hitch- rugby players

Baddeley and Hitch rugby players findings and conclusions

players who played the most games had the poorest recall- there were more chances for interference

internal validity of Baddeley and Hitch

invalid: naturalistic setting meant couldn’t control some confounding variables e.g. injured players will remember the match better due to their experience during it

valid: controlled extraneous variables e.g. length of time

external validity of Baddeley and Hitch

valid: used a realistic scenario

internal validity of McGeoch and McDonald

valid: in a lab- can control extraneous variables e.g. number of words in each group, relationship between lists

external validity of McGeoch and McDonald

invalid: not likely for people in everyday life to have to remember words and numbers in such a short space of time

who did research on interference theory with drug studies

Coenen and Luijtelarr

Coenen and Luijtelarr procedure

gave participants a list of words and later asked them to recall them, one group learnt the words on diazepam, one group learnt them and then took diazepam, and one took a placebo (control)

Coenen and Luijtelarr findings and conclusions

recall of the list of words when under the influence of diazepam was worse than the control group, recall when diazepam was taken after learning was better than the control group, John Wixted suggested the drug prevents new information reaching parts of the brain that involve processing memories so it can’t retroactively interfere

who used cues to disprove interference theory

Tulving and Psotka

Tulving and Psotka procedure

participants were given lists of words one after another, the lists were organised into categories but the participants weren’t told that, had to recall the words in the lists, at the end they were told the categories and had to recall the words again

Tulving and Psotka findings and conclusions

recall was about 70% for the first list, became progressively worse, recall rose back to 70% when given the categories as cues, shows interference causes a temporary loss of accessibility to material still in LTM

who did research on context dependent forgetting

Godden and Baddeley

Godden and Baddeley procedure

studied deep-sea divers, divers learnt a list of words either on land or underwater and recalled the words on land or underwater (4 conditions)

Godden and Baddeley findings and conclusions

recall was 40% lower when learning and recalling the information in different conditions, the external cues available at recall were different from those available at learning, led to retrieval failure- context-dependent forgetting

who did research on state-dependent forgetting

Carter and Cassidy

Carter and Cassidy procedure

gave antihistamine drugs to some of the participants, changed their internal state, had to learn words and passages and recall them in one of four conditions (learning on drug/not on drug, recalling on drug/not on drug)

Carter and Cassidy findings

performance in memory test was much worse when the states of learning and recall were different- state-dependent forgetting

who did a context-related forgetting study on a museum

Aggleton and Waskett

Aggleton and Waskett procedure and findings

recreated smells of a Viking museum and found recall of details were more accurate with the smells- context-dependent forgetting

what did Godden and Baddeley do in their replication of their experiment

used the same 4 conditions, participants this time had to say whether they recognised a word instead of recalling them

findings of Godden and Baddeley’s replication experiment

performance was the same in all four conditions, no context-dependent effect

who did the study on leading questions and misleading information

Loftus and Palmer

Loftus and Palmer procedure

45 participants (students) watched a film of clips of a car accident and then asked them questions about the accident- one leading question, five different groups each got different verb describing speed of car in the leading question: hit, contacted, bumped, collided, smashed

Loftus and Palmer findings

mean estimated speed calculated for each group, contacted was 31.8 mph, smashed 40.5 mph

Loftus and Palmer second experiment

supported substitution explanation, participants who heard smashed were more likely to report seeing broken glass (there was none) than those who heard hit, critical verb altered memory

who investigated post-event discussion on EWT

Gabbert et al.

Gabbert et al. procedure

studied participants in pairs, each pair watched a video of the same crime at different angles so they could only see specific elements, the pair then discussed what they had seen before individually completing a test of recall

Gabbert et al. findings

researchers found 71% of participants mistakenly recalled aspects of the event that they did not see but had picked up in the discussion, in the control group where there was no discussion it was 0%

negative evaluation of Loftus and Palmer

• low population validity- students more likely to believe what they are told and care about performing well

positive evaluation of Gabbert et al.

• high internal validity- well controlled

negative evaluation of Gabbert et al.

• low ecological validity- low emotional involvement, no incentive to pay attention

• few consequences unlike in court, no consequences to themself or anyone else if they don’t remember anything, little incentive to pay attention

who investigated negative effects of anxiety on recall

Johnson and Scott

Johnson and Scott procedure

lab study, participants seated in a waiting room, low anxiety condition- heard casual conversation next door, saw man walk past with a pen and grease on his hands, high anxiety condition- heard a heated argument, sound of breaking glass, man walked out holding a knife covered in blood

Johnson and Scott findings and conclusion

participants picked man from 50 photos, 49% correct in low anxiety condition, 33% correct in high anxiety condition, tunnel theory- people have enhanced memory for central events, weapon focus as a result of anxiety has this effect

who investigated positive effects of anxiety on recall

Yuille and Cutshall

Yuille and Cutshall procedure

study of an actual shooting, shop owner shot thief dead, witnesses interviewed 4-5 months after, compared with police interviews that happened at the time of the shooting, accuracy determined by number of details in each account, rated stress on a 7 point scale and if they had emotional problems since e.g. sleeplessness

Yuille and Cutshall findings and conclusion

witnesses were very accurate in their accounts, little changed in recall accuracy, participants who reported the most stress were most accurate, suggests anxiety does not have a strong effect on accuracy of EWM in real world, may even improve it

who investigated anxiety effect on recall in a labyrinth

Valentine and Mesout

Valentine and Mesout procedure

visitors to a horror labyrinth volunteered to wear heart monitors to measure anxiety, after divided into high anxiety or low anxiety group, all asked to describe a person they encountered in the labyrinth

Valentine and Mesout findings

75% of low anxiety group correctly identified actor, 17% of high anxiety group correctly identified actor

positive evaluation of Johnson and Scott

• lab study- good control over variables e.g. level of drama- high internal validity

negative evaluation of Johnson and Scott

• may have measure surprise not anxiety

• ethical issue about inducing anxiety

positive evaluation of Yuille and Cutshall

• used witnesses to real event compared over several months- high external validity

• more ethical- doesn’t cause any unnecessary anxiety

negative evaluation of Yuille and Cutshall

• proximity to the shooting may have improved memory and increased anxiety

positive evaluation of Valentine and Mesout

• two measures of anxiety (heart monitor and questionnaire)- increases validity

negative evaluation of Valentine and Mesout

• quasi experiment so no random allocation to conditions- risk of participants variables