Memory in everyday life

There are 'other factors affecting accuracy of EWT

• Anxiety

• Age of Witness

• Consequentiality

• Individual Differences

There is conflicting evidence about the effect of stress & anxiety on witness recall. Lab based studies generally shown impaired recall in people who have witnessed unpleasent situations.

Loftus & Burns 1982, had some p's that were shown a violent crime where a boy was shot in the face. These p's had a significantly impaired recall for events running up to the incident. Loftus also looked at 'weapon focus', this demonstrated the powerful role that anxiety can lay in undermining the accuracy.

However it could be said that lab does not reflect real-life. Chrsitianson & Hubinette 1993, reported that incidents that had high levels of stress the memory was accurate, detailed & long-lasting. They carried out a survey of 110 people, 22 had seen genuine bank robberies, the victims (subjected to greatest anxiety) showed more details.

A lab experiment by Loftus 1979

Aim was to find out if anxiety during a witnessed incident affected accuracy of later identification.

Procedure was p's were exposed to 1 of 2 situations:

1. they overheard a low-key discussion in a lab about an equipment failure, a person then emerges from the lab holding a pen with grease on his hands

2. they overhead a heated & hostile exchange between people in lab, after the sound of breaking glass & crashing chairs, a man emerged from lab with paper knife covered in blood

P's were then given 50 photos & asked to identify the person who had come out of the lab.

Findings 1. those in group 1 had accurately identified the person 49%

               2. those in group 2 were successful only 33%.

Conclusion This finding has come to be known as the 'weapon focus' phenomenon, where the witness concentrates on the weapon & this distracts attention from the appearance of the perpetrator. Loftus concluded that the fear/ anxiety induced by the sight of a weapon narrows the focus of attention & gives rise to very accurate recall of the central details of the scene.

Evaluation Later Research by Loftus and Burns 1982 has provided support for this findings. P's watched either a violent, non-violent short film of crime. Those who saw the violent version, in which a boy was shot in the face, were less accurate in recalling information about the crime.

However it is mainly lab that produced the results, they could be accused of lacking validity as rather diff picture emerges if look at later real life studies.

The experiment raises ethical issues of welfare of p's deceived & upset by knife.

As children sometimes are witnesses, psychologists were intreguied to see if the same factors of acuracy in adults operate in kids.

Dekle et al 1996 found children are more willing than adults to make a positive identification even though they often choose the wrong person in an indentity parade.

Children seem to be more susceptiblee than adults to absorbing post event information into their original memory representation. As found by Poole and Lindsay

Flin et al 1992 found that EWT becomes less accurate over time especially with children. Children and adults were questioned 1 day and 5 months after the incident. After 1 day there was no difference in amount/ quality of recall but for the children 5 months there was significant decline. This is important as court normally is much later than the crime.

Gordon et al 2001 concluded young children can produce detailed & accurate recall but susceptible to suggestion so their account should be viewed with caution.

Davies 1994 believed that the differences have been over stated that kids can provide valuable testimony.

There is some evidence that the elderly are prone to errors of recall more than young adults. Yarmey 1984 found that when asked questions about a staged event, 80% of elderly p's compared to 20% of younger adults failed to mention the attacker having a knife in his hand.

Cohen & Faulkner 1989 showed a film of kidnapping to a group of middle-aged & elderly p's. They than read a narrative account of the scene. For 1/2 the p's, the narrative account was consistent with what they had seen, the other carried misleading information. In a recall test, elderly p's were found to have been susceptible to the effects of misleading q's.

Poole and Lindsay 2001

Aim was to see whether ages of witness has a effect on EWT

Procedure Engaged children aged 3-8 in a science demonstration. The parents of the children than read them a story, which contained some of the elements of the science demonstration but also included novel information. The kids were than questioned about the demonstration/ in another phase, the kids were asked to think v.carefully about where they got the information from (source).

Result It was found that they had incorporated much of the new information into their original memory. Some of the older kids revised their accounts and extracted the post-event information but younger kids did not.

Conclusion Children are more susceptible than adults to absorbing post event information into their original memory representation

Evaluation  Methodological issues- although experiment was hard to eliminate the extraneous variable than using artificial stimuli in a highly controlled lab.

The instructions had to be clear as kids young, short attention span.

Ethical issues- parents must given informed consent but this case helpful parents involved so children with familiar people and less susceptible to investigator affects.

Foster et al (1994) tried to see whether witnesses were more likely to b accurate if they believed that their evidence would influence the conviction. P's watched a video of a bank robbery & were than asked to pick out the robbers from an identity parade. 1//2 the p's were told it was genuine & their influences would affect the trial, while the others were told it was a stimulation. P's were more accurate in the condition they thought to be real. This study could be criticized for artificiality since hard to test to know if p's believed the crime to be genuine. But it is an indication that real-life situations have different factors to those in experiment.

Some people appear to be more susceptible to misinformation than others. Tomes & Katz 1997 said that people who are more likely to do so are:

• generally poor recall for the event
• score high on measures of imagery vividness
• high score on measures of empathy

it also seems that people resist obvious mislead information, Loftus 1979 red purse study. She gave p's a set of slides that showed a red purse being stolen from a handbag. They were later given an account of the theft that included several errors including the 'fact' that the purse was brown. In a subsequent recall test, all but 2 of the p's resisted the misinformation about the colour of the bag although there were influenced by misinformation about less central elements of the theft. Loftus concluded that memory for information that is particularly striking is less susceptible to the effects of misinformation than memory for peripheral details.

Schemas are knowledgeable packages which are built up through experience of the world & which enable us to make sense of familiar situations & aid the interpretation of new information. Cohen 1993 has suggested 5 ways in which schemas might lead to re constructive memory:

• we tends to ignore aspects of a scene that do not fit the currently activated schema.
• we can store the central features of an event without having to store the extract details
• we can make sense of what we have seen by 'filling in' missing information
• we distort memories for events to fit in with prior expectations
• we may use schemas to provide the basis for a correct guess

This means that schemas, which are usually useful to us because they help us direct our attention & make our experiences more predictable, may also lead to distortions in memory

Lindsay et al 2004

Aim to see the effects of previous knowledge of a topic on the memory for events

Procedure Read accounts to p's of either a palace burglary/ school field trip to a palace. On the next day, all the p's were shown a video of a museum burglary and than asked to recall events from the video.

Result  P's who had previously heard the account of the palace burglary made more errors in recall than people who had heard the account of the school trip/

Conclusion This suggests that memory for events can be distorted by previous knowledge of a similar topic.

Tuckey & Brewer 2003

Aim to find out whether bank robbery schema would have an affect on recall of a staged bank robbery hold-up.

Procedure T+B found that most people think the following: bank robbery schema

• bank robbers are male
• they wear some kind of disguise
• dark clothes
• demand money
• getaway car

They showed a video of a staged bank hold up, & asked people to recall the scene..

Result  P's had better recall for elements of the film that conformed to their schema than to elements that did not

Conclusion This study as with List study indicates schema has an effect on EWT

List 1986 lab experiment

Aim to identify whether peoples individual schemas affect EWT

Procedure Drawing a list of elements that might occur during shoplifting, asked to rate the events in terms of how likely they were to occur, she than complied a video showing 8 different shoplifting incidents & included some elements that people had rated as high probability and low probability, she showed the video to new p's and asked them a week later to recall what they had seen.

Result  P's were more likely to recall high probability events than low. They are often reported see high probability elements that had not actually been included in the video at all.

Evaluation Used a pilot study, tried to make her video realistic but this isn't the same as real life. needed to obtain consent & debrief afterwards. PEOPLE USE SCHEMAS TO RECALL FOR EWT

Loftus 1975 lab experiment

Showed 150 p's a a film of a car accident. After they had seen the film, p's were divided into 2 groups & each group was asked 10 q's about what they had seen.

• group 1 was asked 10 q's which were all entirely consistent with the original film
• group 2 was asked the same q's, except for one: 'how fast was the white sports car going when it passed the barn when travelling along the country road?

After 1  week, the p's were all asked a further 10 q's & both groups were asked a final a 'did u see a barn?'

• only 2.7% of p's in group 1 gave incorrect answer whereas 17.3% in group 2 did!

Those in group 2, the non-existent barn had been added to original memory

Loftus et al 1978 lab experiment

Aim to see whether p's would recall an event inaccurately if they were asked misleading q's

Procedure

• p's were divided into 2 groups & were shown a set of slides showing the events that led up to a car accident
• the slides for each were group were identical, except for 1 slide: group 1 saw a red car stopping at a junction with a 'yield' sign (give way) group 2 saw the same car stopping at a junction with a 'stop' sign
• after the slide presentation, both groups were asked a set of 20 q's
• 1/2 the p's in each group were asked the q 'did another car pass the red one while it was stopped at the 'stop' sign?' for the other 1/2, the critical q was 'did another car pass the red one while it was stopped at the 'yield' sign? ( this meant 1/2 the p's were asked a misleading q+ 1/2 were asked consistent q)

• After 20 minutes,  all the p's were given 15 pairs of slides, presented in random order, to look at. They had to pick from each pair of slides presented, the slide that had been included in the original set (recognition test). The critical pair of slides consisted of 1 slide showing the car stopping at YIELD sign and other STOP sign.

Result 

• 75% of p's who had received consistent q's picked the correct slide compared to 41% who had misleading q's
• When the recognition test was delayed for a week, accuracy in the group that had been misled fell even further to 20%!

Conclusion

• The misleading q had served to delete the correct info from memory & replace it with false info. That is, the original memory was no longer stored.
• The effect of misleading q's becomes more pronounced over time.

Evaluation

• the conclusions of this study has been supported by other research. e.g Loftus & Loftus 1980 found that accuracy in the misled group did not increase, even when p's were offered money for picking right slide.
• Although the results of this study were statistically significant, it is important to note that by no means everyone in the group that was misled was inaccurate in the recognition task.
• P's were simply shown static slides, we cannot conclude that memory would be affect by misleading q's in same way with real life.
  

Cognitive interview technique (CIT)

In the legal system inaccurate EWT can lead to serious repercussions. One suggestion to improve EW account, by Fisher 1987 who studied real life interviews by detectives in Florida over a 4 month period and found that witnesses were frequently bombarded with a series of brief, direct & close-ended q's in order to elicit facts.

The sequencing of these q's was seen to be out of sync with the witness's own mental representation. Witnesses were often interrupted and not allowed to talk freely about their experiences & disrupted their concentration.

On the basis of research like this Geiselman 1985 developed the CIT

The cognitive Interview                                         Instructions to witness

• Context reinstatement (CR)     Mentally reinstate the context of the       target event, Recall the scene, the weather, what were you thinking & feelings at the time, preceding events etc 
• Report everything (RE)   Report every detail you can recall even if it seems trivial.
• Recall from changed perspective (CP) Try to describe the episode as it would have been seen from different viewpoint, not just your own
• Recall in reverse order (RO) Report the episode in several different temporal orders, moving backwards & forwards in time
  

This is designed to enhance retrieval of the original memory, designed to provide extra cues that may jog witnesses' memory for more central details. Subsequent research led to 'enhanced cognitive interview' after looking at current police practise through detailed analysis of taped interviews. Fisher et al 1987 suggested to add extra features

e.g. they recommended that the interviewer should minimise distractions, actively listen to the witness ask open-ended q's, pause after each response, avoid interruption, encourage use of imagery, adapt their language to suit the witness & avoid any judgemental comments

Geiselman et al 1985 tested p's by showing them videos of a stimulated crime and then testing different groups with CI/ a standard police interview/ interview under the influence of hypnosis. The found CI elected more info from the p's. A no. of subsequent studies have confirmed these findings although Koenken et al 1999 found that witnesses questioned using the CI also recalled more incorrect info than those with standard technique, this is probably cause the CI procedure elicits more info overall than the other.

Fisher et al 1990 have also demonstrated the effectiveness of the CIT in real police setting in Miami. They rained detectives to use enhanced CIT with genuine crime witnesses and found that its use significantly increased the amount of info recalled. More recently Kebbel et al 1999 carried out a survey of police officers in the UK and found that there was quite a widespread use of CI.

However, while officers generally found it useful, they expressed some concern about the amount of incorrect recall generated and amount of time it took to complete the enhanced version. In practise it seemed that the officers were using RE+CR instructions but rarely CP + RO.

Psychologists have tried to test these police perceptions that certain elements of CI are more useful than others Milne & Bull 2002 e.g tested all the Ci procedures used singly and than in combo.They found that all 4 procedures used singly produced more recall from witnesses than the standard technique.  But the most effective combo was RE + CR, which is in line with what practising police officers had expected.

One area where the CI has not proved particularly successful is in the questioning of young children. Geiselman 1989 reviewed a no. of studies and concluded that kids under the age of 6 actually reported events slightly less accurately in response to CIT, this is probably because they find the instructions difficult to understand. The CI can be used to quite effectively for children aged +8.

On balance the CIT has proved v.useful in increasing the amount & accuracy of eyewitness statements.

All the research in memory has enabled education lists to devise/ develop strategies for teaching and learning so that students of all ages can learn and retrieve relevant materials, knowledge and understanding and apply them to work. All of us want our memories to be efficient and accurate in exams, remembering lists of tasks, names of people and other everyday necessities. Research has bought several important strategies.

Few techniques that have been made to improve memory, called mnemonics. All use same principle of organisation, many depend on visual imagery. 

Peg word system for example , one way through rhyme. One is bun. Two is shoe.

Another way is to try form a mental image of each word & then 'hang' it on one of the pegs. Example  milk & butter. Imagine a bottle of milk balancing on a pack of butter which is in a shoe, The stranger the mental image the more likely you are to recall it!

Another method is method of loci where the memory pegs are places rather than rhyme. You think of a regular route you take e.g school and visualise say 10 key locations on way e.g jaugar shop, odeon, sainsburys, children clothes store. Than the list of things you need to remember are associated each with 1 memorable location. When the time comes remember the route & words will be memorable

Visual imagery has been shown in many studies to enhance recall. Pavio 1965 found that p's could recall concrete nouns better than abstract, he explained his findings in terms of dual coding hypothesis. Meaning the concrete words are encoded twice (once verbal, once visually). De Ben i and Moe 2003 while confirming importance of visual imagery for improving memory found this was more helpful got items that had been present verbally. In terms of WM system it could be that visually presented items are an accompanying visual image would be competing for same storage resources (visuo-spatial sketchpad) whereas verbally presented items & visual images held in separate loops.

we are more likely to remember material we have actively processed i.e. we have given deep meaning Craik & Lockhart 1975 suggests that we have to give meaning to what we are learning in order to learn it well. We have to meaningfully process it.

Craik 1977 investigated this by testing recall under different conditions. He presented a list of printed words to 4 different groups of p's and asked each to carry out a diff task as follows: group 1 had to answer a structural question e.g. is words in CAPS group 2  had to carry out an acoustic task e.g does the word rhyme with dog? group 3 had to carry out a semantic task e.g. is it a name of a living thing? group 4  had no particular task to carry out, simply asked to commit the words to memory (the control).Results in group 4 the p's were the control. group 3 performed better than 1&2 as they had to give meaning to their learning. groups 1 & 2 performed similarly to group 4 which overall suggests that meaning engagement  in learning leads to better retention i.e. learner effectively associate new with old learning & make connections

Bransford & Johnson 1972 showed the importance of the role of understanding what has been learnt is to the actual retrieval of the memory of that learning. They gave some p's a passage to read with a title (which gave the passage the meaning) and some group without a title. The former group were able to recall far more of the passage and this shows how understanding underpins the role of organisation

The role of chunking in increasing STM capacity, involves organising material & understanding it, this allows us to integrate and unify material which inturn reduces load on memory. Katona asked people to remember a string of digits in serial order e.g 12345 (was larger digit span!) so nearly impossible but 1.4.9.16.25.36 ; is 1 squared, 2 squared, 3 squared! once you see this pattern all you need to learn is that the digit string consists of square numbers 1-11, an impossible task has been reduced to learning a simple rule!

Specificity principle whereby you recall things better if retrieval context is like encoding text. Geiselman & Glenny 1977 investigated this effect by asking p's to imagine a list of words being said aloud by a familiar person. some p's were told to imagine a familiar female voice and other male. After a delay p's were read a list of words that contained the previous words they had been given before, p's were more successful at this recognition task if the gender of the voice presenting it matched gender they had imagined previously. Its been show that recall for information can be more successful if it is tested in the same room where it was learned. If that is not possible it seems that simply imagining the original learning environment can help later recall (Jerabek & standing 1992).

Also, been suggested that your emotional/ physical state at time of encoding can affect likelihood of recall. Ucros 1989 reviewed many  studies of mood-state-dependant memory & found some evidence that recall is better if mood at time of learning matched mood at time of retrieval, but the effect is only slight & seems that if material  been well learned, context is less important.

The amount of attention placed by the learner into the task set/ the amount work to be learned has a v.significant effect on retention and how something can be remembered. Ericsson & Chase 1981 studied an individual called SF who was able to memorise lists of up to 80 digits- he did this by practising for an hour a day over a 2 year period and it was found that his general memory span for other things was about average!

Practise is important for exam rev especially & spacing rev/learning rather than massing it all together is vital for long turn retention. A lot of memory strategies outlined are rather simplistic

Herrann 1991suggests to take a mutli-modal approach to improve memory including maintaining the mind & body in the best condition. Matlin 1998 stressed importance of meta-memory, this refers to knowledge & awareness of your  own memory & how it works. You need to recognise what strategies work for you & you need to be honest about your own strengths and weaknesses.