Eye-witness Testimonies (EWT)

• what is an eyewitness testimony?

  legal term, referring to the use of eyewitnesses to give evidence in court.
• what does the innocence project suggest about the importance of (EWT)?

  • an organisation which works to clear the names of wrongly convicted people.
  • claim that eyewitness misidentification is the greatest single cause of wrongful convictions in the USA, convictions that were later overturned by DNA evidence.
• what are the 3 stages of EWT?

  1. the witness encodes info into LTM (the event, people involved), may be partial as the event occurs quickly, e.g. at night and accompanied by rapid, complex, violent action.
  2. witness retains info for a time. memories may be lost or modified during retention, other activities may interfere w/ the memory itself.
  3. witness retrieves memory from storage. what happens next is there may be a presence or absence of info that may affect the accuracy of the memory.
• what are the 2 factors effecting reliability of EWT?

  • misleading info= post-event discussion and leading questions.
  • anxiety= does high levels of anxiety improve recall, or make it worse? what is the impact of the weapon focus effect?
• misleading info: post-event discussion

  • occurs when there is more than one witness to an event.
  • witnesses discuss what they have seen (w/ witnesses or other people).
  • info is added that may be misleading.
  • the accuracy of the witness's recall may be reduced.
  • false memories can be stimulated by misleading post-event discussion.
• what is the conformity effect?

  witnesses go along w/ each other to win social approval or because they think others are right.
• gabbert (2003):

  • p's put into pairs. both p's watched a diff video of the same event. in each video there were unique differences.
  • control cond- recall straight after watching video.
  • experimental cond- pairs encouraged to disccus the event then recall.
  • results: 71% of those who had discussed the event went on to mistakenly recall items acquired during discussion.
  • HOWEVER, study lacks ecological validity.
• counter-argument to argument that gabbert (2003) lacks ecological validity:

  oklahoma bombing-

  • one witness stated the murderer had an accomplice. no other witnesses had stated this. after time, they claimed they too had seen someone else.
  • the 1st witness later realised this was a mistake in their recall.
• leading questions- response bias explanation:

  when a person gets a leading question it biases their response w/out changing the memory.
• leading questions- memory substitution explanation:

  • when a person gets a leading question, it actually changes the stored memory.
• loftus and palmer (1974)- leading questions:

  • students watched film clips of car crashes, then were asked how fast cars were going when they: 'contacted', 'bumped', 'collided', 'hit' and 'smashed' into each other.
  • found that the mean estimated speed increased depending on which description was used in the question.
  • e.g. 'contacted' at 31.8 mph and 'smashed' at 40.8 mph.
  • this showed the effect of leading questions.
• loftus & pickrell (1995)- post-event discussion:

  • found that they could plant false memories of being lost in a store aged 5 in their sample of 24 p's.
  • nearly 30% of the sample recalled being lost & were able to provide details purely from suggestions.
  • this suggests that memory is reconstructive and not like a video re-enactment.
• loftus & pickrell (2003):

  further research by loftus and pickrell showed that a false memory could be created of a childhood visit to disneyland.

  • this time, they were able to plant the false memory by using a picture of a fake ad. this is a more subtle way of suggesting the memory but equally effective.
  • around 30% of p's recalled seeing bugs bunny at disneyland, which isn't possible since he isn't a disney character.
• evaluation of research into misleading information:

  • strength- research is very useful when applied to criminal justice system and has led to improvements in procedures.
  • weakness- early research very artificial and lab-based, e.g. loftus car crash work used police crash videos lasting a few seconds and had limited samples (psych students) so was lacking in validity and generalisability.
  • research could be abused to plant suggestions, e.g. by advertisers to make people buy their products.
• contradictory evidence to leading questions (A03):

  • yullie & cutshall interviewed 13 witnesses to a fatal shooting in canada 5 months after being interviewed by police.
  • when they interviewed them, they also asked 2 leading questions.
  • these questions had no effect on the accuracy of recall.
  • suggests that leading questions may not be the only influence on memory. incident was more relevant to them.
• why is much of the research in eye witness testimony now focused on the effects of arousal?

  • anxiety = emotional state where we fear that something bad is about to happen. people often become anxious when they are in stressful situations. this anxiety tends to be accompanied w/ physiological arousal (increased heart rate, shallow breathing).
  • impacts memory and recall. some studies suggest that moderate levels of stress can improve memory (weapon focus effect) others highlight that high stress can negatively affect accuracy.
• research that supports anxiety negatively affecting EWT- loftus & burns (1982):

  • p's shown a violent version of a crime where a boy is shot in the face. p's had significantly impaired recall for events running up to the violent incident.
  • study shows that p's were more likely to focus on a weapon in the high anxiety situation, thus impairing their recall of events, suggesting anxiety can negatively impact the reliability of EWT.
• research that supports anxiety negatively affecting EWT- johnson & scott (1976): 


  • p's sat in a waiting room overhearing an argument in another room. they then saw a man run through the room carrying either a pen covered in grease (low anxiety condition) OR a knife covered in blood (high anxiety/weapon focus condition).
  • results:

  1. mean accuracy in low anxiety condition (pen)- 49%
  2. mean accuracy in high anxiety condition (knife)- 33%

  • the results show us that anxiety leads to poorer recall in EWT, the higher the anxiety, the less accurate the recall.
• what is the weapon focus effect?

  when an eyewitness focuses on the weapon, increases anxiety, and reduces attention to other details of a crime.
• evidence suggesting anxiety positively impacts EWT- christianson & hubinette (1993):

  • questionned 58 real victims of a bank robbery.
  • found that those who had actually been threatened were more accurate in their recall, compared to those who were onlookers.
  • this continued to be true 15 months later.
  • suggests that anxiety has a positive effect on EWT as recall was more accurate from those who were actually threatened than those onlooking.
• evaluation (A03)- the yerkes-dodson law:

  • deffenbacher suggests that this apparent contradiction in research findings could be best explained w/ reference to the yerkes-dodson law...
  • which states that performance improves w/ increases of arousal up to some optimum point and then declines w/ further increases.
• the yerkes-dodson law ~

  what the yerkes-dodson inverted-U hypothesis says about anxiety:

  • moderate amounts of anxiety can improve accuracy & detail of recall.
  • this increases up to an optimal level.
  • after this level increasing, anxiety leads to a decline in recall.
• how can the yerkes-dodson law be used to explain why p's in loftus & burns and johnson & scott's studies showed poor recall in the weapon focus conditions?

  • when faced w/ high anxiety conditions, the p's were past the optimal level of arousal and therefore their performance decreased, leading to a decrease in their ability to recall effectively.
• however, this goes against what happened in christianson and hubinette's study. the witnesses saw a real crime and had very high levels of anxiety. how would the yerkes-dodson effect suggest their performance was so strong and their recall was good?

  • it could be argued that in this case, for these p's, they were at an optimal level of arousal and therefore their recall was at their highest.
• evidence against the yerkes-dodson effect ~

  yuille & cutshall:

  • interviewed 13 witnesses to a fatal shooting in canada 5 months after being interviewed by police.
  • found those w/ high anxiety had LESS accurate reall than those w/ low anxiety. but those w/ VERY high anxiety had extremely accurate recall.
  • as those w/ higher anxiety showed better recall it doesn't support the YD effect as the higher the anxiety, the less we should be able to recall as it has surpassed the optimal point.
  • however, one weakness of the study was that the witnesses who experienced the highest levels of stress were actually closer to the event, and this may have helped w/ the accuracy of their memory recall.
• alternative explanation of poor reliability of EWT:

  • freud states that in times of extreme emotional distress, such as w/ traumatic events, that we repress (remove the memories into our unconscious) to protect us from the distress.
  • he would therefore suggest this is why our recall of a traumatic event may be poor.
• evaluation of anxiety as an explanation into why EWT is unreliable:

  ethics ~

  P- one issue w/ lab based research into anxiety and EWT is that it can be unethical.

  E- it's potentially unethical because of the psychological harm they're subjected to purely for research purposes.

  E- it raises questions to whether there is a benefit of causing harm to others in the pursuit of science.

  L- therefore this why real-life studies may be more effective because psychologists interview people who have witnessed a real-life event, so there is no need to create it.
• another evaluation of anxiety as an explanation into why EWT is unreliable:

  individual differences (can we generalise these findings to all as were are all so different) ~

  P- it's also important to address that anxiety may not be the reason that recall is affected.

  E- this explanation ignores that individual differences may lead to differences in recall.

  E- e.g. bothwell et al. found that neurotic personality types (less emotionally sensitive) showed increased recall in anxiety.

  L- therefore it could be a person's personality AND NOT the stress of the event leading to failure to recall. this weakens support for the theory.