Reliability of eye-witness testimony

• What are the 4 sections?

  Crimes are emotive experiences, post-event information, memory is reconstructive & child witnesses.
• Davies et al (1989):


  Reviewed literature discussing kids used as witnesses. 6-7 & 10-11 fairly accurate in memories & don't usually make things up or deliberately lie when giving testimony. Memory for important details not significantly altered by adult suggestion after event.
• Anastasi & Rhodes (2006):

  Found all age groups most accurate when recognising offender from own age group. (Much research of child eye-witnesses uses adults as 'target' individual).
• Pozzulo & Lindsay (1998):

  Meta-analysis drew data from no. of studies-over 2000 participants. Kids under 5 less likely than older kids or adults to make correct identifications when target present in line-up. Kids aged 5-13 didn't differ significantly from adults in target-present condition, but were more likely to make choice (inevitably wrong choice) in target-absent condition. (Thought this was due to kids being more sensitive about doing what they were asked- feel they can't say no & have to give an answer).
• Why might kids not be reliable eye-witnesses?

  Prone to fantasy & memories may be affected by suggestions made by others.
• Why might eye-witnesses be reliable (crimes are emotive experiences)?

  Emotionally shocking events which hold personal significance create accurate & long-lasting memory (flashbulb memory). Hormones associated with emotion (e.g, adrenaline) may enhance storage of memories (Cahill & McGaugh 1995). Emotion surrounding crime may lead to more reliable memories.
• Why might eye-witnesses not be reliable?

  Crimes witnessed are emotive, emotionally traumatising & unexpected. Freud argued extremely painful/threatening memories forced into unconscious mind. (Repression ego-defence mechanism).
• Loftus & Palmer (1974):

  Info 'suggested' after event becomes incorporated into OG memory.
• Loftus & Zanni (1975):

  Demonstrated effects of post-event info. 7% of those asked 'Did you see a broken headlight?' reported seeing one. 17% asked 'did you see THE broken headlight?' reported seeing one. Subtle changes in question wording can influence recollection of participant.
• Loftus (1979b):


  Showed participants slides of a man stealing large, bright red purse from woman's bag. Later showed info purporting purse was brown. 98% correctly remembered purse was red. Suggests eye-witness recollection for central or key details may be more resistant to distortion from post-event info.
• What is wrong with eye-witness research?

  Focuses on details hard to estimate (e.g, speed) or details not central to incident-misleading research & thus may be more susceptible to corruption.
• Yuille & Cutshall (1986):

  When research conducted with eye-witnesses to real-life crimes rather than lab-based 'crimes', accuracy is higher. Recollections didn't fade over time & weren't susceptible to leading questions.
• What did RapeCrisis report?

  90% of rapists known to victims. Ability to identify assailant likely very reliable, even when crime was traumatic. (In many crimes, eye-witnesses know perp so don't refer to schemas).
• Yarmey (1993):


  Asked 240 students to look at videos of 30 unknown males & classify them as 'good' or 'bad'. High agreement suggesting similarity in info stored in schemas. Any preconceived ideas about facial features of criminals may influence us when making decisions on suspects in line-up-may not select actual criminal.
• How may schemas affect eye-witness reliability?

  Schemas help process info quickly- info held in schemas may distort memory of event e.g, criminal expectation may be derived from TV & incorporated into memory when recalled.