Eye Witness Testimony

• Loftus and Palmer (1974)

  Eyewitness testimony is information given by a witness to a crime. Loftus and Palmer (1974) studied the effects of leading questions in two experiments.
• EWT

  • Eyewitness testimony is the information and details given to police by witnesses. 
  • Facts such as location/time/events of the crime scene and the appearance of the criminal are given. 
  • But they can be inaccurate because of both faults in memory, and leading questions.
• Leading questions

  • Leading questions are questions that potentially guide a witness to a certain answer. 
  • For example, a leading question could be: 
  • “Did you forget to do your homework?” 
  • A better-phrased question could be: 
  • “What happened with your homework?”
• Experiment one

  • Participants were shown a film of 2 cars crashing and were then asked questions about the events of the car crash. 
  • Questions included: 
  • “How fast do you think the cars were going when they hit?”
• Method

  • The word "hit" was then substituted for different participants. - Examples of other words used instead of 'hit' were: 
  • Smashed.
  • Bumped.
  • Contacted.
  • Collided.
• Results 

  • Participants estimated different speeds when different words were used. 
  • Participants tended to give higher speeds when the word "smashed" was used. On average they estimated 41 mph. 
  • The word "contacted" gave the lowest average speed estimate of 32 mph.
• Loftus and Palmer (1974)

  Eyewitness testimony is information given by a witness to a crime. Loftus and Palmer (1974) studied the effects of leading questions in two experiments.
• Experiment two

  • Three groups of participants were used in this experiment:
  • Group 1 were participants that were given the word ‘smashed’.
  • Group 2 were participants that were given the word ‘hit’.
  • Group 3 were participants that were given no suggestion of the speed of the car.
• Method

  • After one week, the participants were brought back in and asked the question: 
  • “Did you see any broken glass?” 
  • The correct response would have been ‘no broken glass’ as there was no broken glass seen in the film.
• Results

  • Group 1 (‘smashed’ group) were more likely to say that they had seen broken glass compared to the other two groups.
• Conclusion

  • Both experiment 1 and experiment 2 support the conclusion that leading questions (even just changing one word like ‘smashed’) can affect the accuracy of the eyewitness testimony. 
• Laboratory study

  • This was a laboratory study and so had good control of extraneous variables. 
  • But because a film was shown and it was an artificial setting, it may not give an accurate depiction of eyewitness testimony.
  • Real-life scenarios, such as a car crash, can be traumatic and emotionally arousing. So this trauma can also have effects on memory.
• Demand characteristics

  • Demand characteristics are when participants in a study may distort the results of the study by being aware of what the aim of the study is. 
  • The participants in this study may have been aware of its true purpose (by being asked leading questions, they may realise the study was about vulnerability to leading questions.) 
  • If this were the case, the reliability and validity of the results would be compromised.
• Usefulness 

  • A strength of this study is the real-life implications. 
  • If police officers have this information, they can conduct more reliable and trustworthy police interviews. 
  • Police want to gain the most accurate facts, so they should be made aware of the effects of leading questions. 
  • Police should set guidelines and give information on how to formulate proper/non-leading questions.
• Loftus and Zanni (1975)

  Loftus and Zanni found that by even changing a single word in a question, the accuracy of recollection can be affected.
• Method

  • Similar to the Loftus and Palmer (1974) study, participants were shown a video of a car crash. 
  • The participants were then either asked: 
  • “Did you see a broken headlight?” or 
  • “Did you see the broken headlight?” 
  • The video did not show a broken headlight.
• Results

  • 7% of the “a” broken headlight group stated they saw one.
  • 17% of the “the” broken headlight group stated that they saw one.
• Conclusion

  • The accuracy of the recollection of events can be altered even by changing a single word. 
  • By using the word “the”, it implied that there was a broken headlight. 
  • By using the word “a”, the implication is there may or may not be a headlight.
• Causality 

  • The study was a laboratory study, so extraneous variables (age/gender/ethnicity/etc.) could be controlled. 
  • So causality could be demonstrated (changing the word causes the effect of altered recollection).
• Ecological validity 

  • But as it was a laboratory study, the setting was artificial. 
  • By being shown film of a car crash rather than observing an actual car crash, this could alter the effect of eyewitness testimony. 
  • This means the study lacks ecological validity.
• Loftus (1979): The Effect of Anxiety on Eyewitness Testimony 

  Anxiety can affect memory accuracy. Loftus (1979) studied the effect of anxiety on eyewitness testimony of violent crime. 
• Anxiety and recall

  • Anxiety can affect memory recall: 
  • Very low levels of anxiety (i.e. not being alert or being sleepy) can result in poor recall accuracy.
  • High levels of anxiety, such as the high levels of stress and arousal after witnessing violent crime, can also result in poor recall accuracy.
  • So psychologists believe that there is a happy medium of optimal memory recall accuracy.
• Method

  • Two conditions were created in this independent design study.
  • In both conditions, participants heard a discussion in a room nearby. 
  • Group 1: participants witnessed a man come out of the room with his hands covered in grease and holding a pen. 
  • Group 2: participants witnessed a man come out of the room carrying a knife covered in blood. 
  • Participants were then asked to identify the man from 50 photographs. 
  • The participants were unaware that this was staged.
• Results

  • The participants in Group 1 (pen/grease) had 49% accuracy in identifying the man.
  • In Group 2 (knife covered in blood), the participants only had a 33% accuracy rate in identification.
• Conclusion

  • Levels of anxiety affected the accuracy of recognition. 
  • The higher levels of anxiety as a result of seeing a knife covered in blood caused Group 2 to be less accurate in identifying the man.
• Conclusion cont. 

  • When witnesses were more anxious, they tended to focus on the weapon rather than the criminal. 
  • Overall, Group 2 participants were more likely to focus on the weapon itself, rather than other details of the crime scene (such as the man). 
  • So when witnessing violent crimes, witnesses tend to focus on central details and ignore other details.
• Evaluation

  • Because the participants were unaware that the scenario was staged, the study has high ecological validity.
  • But there are ethical issues because of the potential trauma of witnessing a man carrying a knife covered in blood. 
  • Psychological participants have a right to be informed and debriefed of the nature of the deception of the study.
• Gabbert et al (2004) - Staged Crime

  Research aim: To investigate the effect of post-event information on accuracy of recall.
• Participants

  • Laboratory experiment.
  • 60 students from the University of Aberdeen and 60 local older adults were the participants.
• Procedure 

  • Participants were allocated to one of two groups. 
  • The control group individually watched a video of a girl stealing a wallet.
  • The experimental group (co-witness group) were put into pairs and watched the video individually. They were told it was the same video, but one half of the pair saw the scene from a different perspective and did not actually see the girl steal the wallet.
  • The co-witness group were then allowed to discuss what they had seen.
  • All participants then completed a questionnaire to test their memory of the event.
• Results & conclusions

  • Results: 
  • 71% of the co-witnesses reported remembering information that they had not actually seen.
  • 60% said she was guilty despite not seeing her commit the crime.
  • Conclusion: Participants in the co-witness group were influenced by the post event information given to them by their partner.
• Evaluation

  • Strengths
  • Reliable – easy to replicate. 
  • Practical applications – demonstrates unreliability of eyewitness testimony. 
  • Limitations
  • Lacks ecological validity – the participants were aware it was not a real crime.
  • Demand characteristics. 
  • Social desirability.
• Valentine and Coxon (1997) - Effect of Age on EWT

  Research aim: To investigate the effect of age on eyewitness testimony and the effects of leading questions.
• Procedure

  • Laboratory experiment. 
  • Three groups of participants:
  • Children (7-9 years);
  • Young adults (16-18);
  • Elderly (60-85).
  • Participants watched a video of a kidnapping and were then asked a series of leading and non-leading questions.
• Results & conclusions

  • Results: 
  • Both the elderly and children gave more incorrect answers to non-leading questions. However, children were more likely to be misled by leading questions.
  • Conclusion: 
  • This shows that children are more susceptible to the influence of leading questions and therefore do not make reliable eyewitnesses. The elderly and children are more likely to have poorer memories for events than young adults. This shows that age does influence the accuracy of eyewitness testimony.
• Evaluation 

  • Strengths
  • Reliable – easy to replicate. 
  • Practical implications for the use of children as witnesses. 
  • Limitations
  • Lacks ecological validity – not real life. 
  • Ethics – showing a crime video to young children.