Lecture 3 Eyewitness Identification PDF

COMMONWEALTH OF AUSTRALIA Copyright Regulations 1969 WARNING This material has been reproduced and communicated to you by or on behalf of the University of Sydney pursuant to Part VB of the Copyright Act 1968 (the Act)....

COMMONWEALTH OF AUSTRALIA Copyright Regulations 1969 WARNING This material has been reproduced and communicated to you by or on behalf of the University of Sydney pursuant to Part VB of the Copyright Act 1968 (the Act). The material in this communication may be subject to copyright under the Act. Any further reproduction or communication of this material by you may be the subject of copyright protection under the Act. Do not remove this notice. Crime Investigation: Witnesses II Associate Professor Helen Paterson Phone: 9036 9403 Email: [email protected] The Plan… Background Evidence of inaccurate identifications (ID) Factors affecting the accuracy of IDs  ID procedures  Composition of lineup  Format of ID procedure  Instructions given to the witnesses  Lineup presentation Juries and ID evidence What can be done? Learning Outcomes At the end of this lecture you will be able to:  Discuss 3 forms of evidence on the inaccuracies of IDs  Discuss how ID composition, procedures, format, instructions, and presentation affect the accuracy of ID evidence  Discuss juror perceptions of ID evidence  Discuss ways to improve ID procedures Background Witnesses are frequently asked to identify a culprit from a lineup Lineups contain the suspect who is placed among a set of individuals who are known to be innocent for the crime in question, called foils or distracters Background To accurately assess the rate at which real witnesses will correctly identify culprits two types of lineups are needed in research:  Target-present lineups – Lineup contains the culprit  Target-absent lineups – Lineup contains an innocent suspect Background Type of Guilty Accurate Identification Decision Lineup Culprit Present Target- Yes Correctly identify culprit Present Target- No Correctly reject all lineup members Absent Background ID Evidence can be dangerous!  It is very convincing  Research shows us that identification evidence is often inaccurate  False IDs have 2 negative consequences: – Guilty person goes free – Innocent person may end up in jail  Thus, eyewitnesses can be sincere, confident, convincing…. And wrong. Evidence of Inaccurate IDs How do we know ID evidence is often inaccurate? 1. Surveyed justice officials 2. DNA Exoneration cases 3. Empirical studies of ID performance 1. Surveyed Justice Officials Surveyed US justice officials reported that over 70% thought that erroneous convictions occurred in less than 1% of the cases (Huff, 1987).  Approx. 3 million arrests for serious crimes/year in US  Approx. 50% result in conviction = 1.5 million  If 0.5% are erroneous, then =7,500 wrongful convictions per year in US 2. DNA Exoneration Cases Since the 1980s, advances in DNA analysis have allowed investigations to link a person to a crime with a very high degree of certainty. DNA analysis has also been used to prove the innocence of convicted inmates (see www.innocenceproject.org) In 2015, 72% of the cases of DNA exoneration were due (at least partly) to mistaken identification The most common factors leading to wrongful Mistaken ID Serology inc Based on first 70 DNA exoneration cases Police misconduct Prosecutorial misconduct Bad science Bad lawyer Hair analysis False witness Informant False confess 70 60 50 40 30 20 10 0 conviction 3. Empirical Studies 2 types of empirical studies on ID evidence:  Field studies  Laboratory studies Field studies are high in ecological validity, but lack experimental control Lab studies allow more control, but typically lack ecological validity. Evidence from Field Studies Several studies have had confederates go into convenience stores or banks and perform an unusual transaction (e.g., Piggott et al., 1990) Overall, about 41.8% of the IDs were correct ID and 35.8% were false. Results were no different for the bank clerks who claimed they had been trained in ID Evidence from Lab Studies Typically participants either watch a video or a staged mock-crime Participants are then required to identify the perpetrator from either photographs or a live lineup A review of many of these studies concluded that the rate of false ID varied from about 0% to about 100%! It is more useful to note that false IDs were often associated with very high witness confidence (Wells, 1993) Problems with Lab Studies Lab studies are not normally designed to estimate overall level of accuracy – rather they are designed to investigate the effect of some particular variable (Cutler & Penrod). Researchers tend to manipulate task difficulty so as to avoid floor and ceiling effects. Thus, we shouldn’t try to estimate overall performance on the basis of these studies. Factors Affecting the Accuracy of Eyewitness ID (Wells, 1978) Estimator Variables System Variables  Viewing conditions  ID procedures*  Fatigue, alcohol etc  Composition of lineup *  Attention  Format of ID  Own race bias  Arousal procedure *  Weapon focus  Instructions given to witnesses *  Lineup presentation * Identification Procedures There are two types of identification procedure 1. Showups 2. Lineups Showups Showup: A one-person lineup containing only the suspect. Can be performed in street or at police station. Acceptable if not enough evidence to arrest and conduct lineup. Have been criticised as biased because the witness knows the person the police suspect. Lineups Lineup: Tests the ability of the witness to identify the suspect when seen with several foils. Lineups Lineup procedures vary greatly between jurisdictions. 25% of Mid-West USA cases involve creation of a lineup made up only of suspects (Wells & Turtle, 1986). Lineups In Australia and UK this would not be permitted as procedures are more tightly controlled. In UK there must be suspect plus 7 or more innocent foils of similar appearance. In UK suspect chooses position in line. In NSW there are often 20 people in a lineup. More important than real size may be the “functional size” (Wells et al., 1994) Composition of ID Procedure Composition: Lineup should consist of foils who look like the eyewitness’ description. Features that the witness mentioned in the culprit’s description should be matched across lineup members (unless doing so would cause the suspect to stand out). Composition of ID Procedure For a lineup to remain fair, characteristics of the suspect should not stand out from those of foils.  E.g., gender and race should always be matched across lineup members. Composition of ID Procedure Gary Wells has argued that a lineup serves to uncover information not available from verbal description. A good foil matches verbal description but differs from suspect in ways not mentioned in verbal description. Format of ID Procedure Lineups can be live, videos, or photos Live lineups are expensive and difficult to arrange. Meta analysis compared formats (live, photo, video) and found no evidence of any differences. Suggested that live lineups might not be worth the trouble (Cutler et al., 1994). Identification from Photographs Can take the form of a:  Photo-lineup (or photospread)  Mugshot search Photo-lineups A photo-lineup is used to test witness ability to recognise a suspect. In the UK the preference is for live or video lineups. However, photo lineups are allowed when this is not possible or if the suspect refuses to take part in a live line-up. In US and AU the majority are photo. Mugshot Searches A mugshot search should only be used to help identify a suspect in the early stages of an investigation. A witness who sees a face in a mugshot search is likely to pick that person out in a lineup, even if not the culprit due to Unconscious transference: When people remember a face, but mistake the circumstances in which they saw the face Instructions Given Lineup instructions: Witnesses should be told that perpetrator may or may not be in the lineup  Witnesses are more likely to identify someone if they’re told the suspect is in the lineup (Malpass & Devine, 1981)  Even if we give the witness the option of saying “don’t know”, biased instructions that imply the perpetrator is present, increase the risk of False ID (Cutler & Penrod, 1995) Instructions Given A police officer who knows who the suspect is might also give verbal and nonverbal cues to witness Lineup Presentation Simultaneous lineup: Present all lineup members at the same time to the witness. Sequential lineup: Members are presented one at a time, must decide if it is or is not the criminal before seeing another photo/person. Did you get it right? Sequential lineups result in fewer false identifications However, may result in fewer correct IDs Lineup Presentation Sequential lineups are thought to be better than simultaneous lineups because witnesses use absolute judgments instead of relative judgments Relative judgment: Comparing lineup members to one another and choosing the one who looks most like culprit Absolute judgment: Each member of the lineup is compared to the witness’ memory Simultaneous vs Sequential? Meta-analysis of 72 experiments (Steblay, Dysart & Wells, 2011) Sequential lineups were less likely to identify any suspect type (whether guilty or not guilty) than simultaneous lineups, but when a suspect was identified he/she was more likely to be guilty when using sequential than a simultaneous lineup. There is an 8% difference in suspect identification between sequential and simultaneous lineups, favouring simultaneous lineups; meaning that simultaneous lineups are more likely overall to identify the guilty suspect. Do jurors recognise problems with eyewitness testimony? Perceived accuracy vs. Actual accuracy  Jurors are most likely to believe confident witnesses, but confidence is not always related to accuracy. Do jurors recognise problems with eyewitness testimony?  When lineups are properly conducted, confidence and accuracy are strongly related for adult witnesses. High- confidence suspect identifications are remarkably accurate.  However, under less favourable ID conditions (e.g., unfair lineups), the accuracy of even a high-confidence suspect ID is seriously compromised (Wixted & Wells, 2017). Do jurors recognise problems with eyewitness testimony? Students read information about a Robbery-Murder Case (Loftus, 1974; 1979) then allocated to one of three conditions: 1. Circumstantial evidence 2. Circumstantial evidence PLUS eyewitness testimony 3. Circumstantial evidence PLUS discredited eyewitness testimony Do jurors recognise problems with eyewitness testimony? % Voted for Conviction 80 60 40 20 0 No Eyewitness Eyewitness Discredited Eyewitness What can be done?  The American Psychology–Law Society solicited a review paper to examine ID evidence and issue recommendations  Wells et al. (2020) produced nine research-based recommendations to help prevent mistaken eyewitness identifications: 1. Conduct a thorough pre-lineup interview of the eyewitness 2. Decide when or even whether to conduct a lineup 3. Use a neutral administrator to conduct the lineup 4. Select lineup fillers so that the suspect does not stand out What can be done?  Nine recommendations cont: 5. Give nonsuggestive pre-lineup instructions to eyewitnesses 6. Take a confidence statement from an eyewitness immediately upon an identification decision rather than later 7. Video-record the entire procedure 8. Avoid repeated identification procedures with the same witness and suspect 9. Avoid “showups” (an identification procedure with a single suspect) in favour of lineups when possible What can be done? Educate jurors about:  Biased lineups  Lineup instructions given  How confidence is not always a good indicator of accuracy  The way in which (mis)leading questions can influence memory Take Home Message ID evidence is potentially dangerous as there is evidence of inaccurate IDs Several system variables influence the accuracy of ID evidence Jurors often don’t recognize problems with eyewitness identification evidence Psychologists have attempted to improve the accuracy of ID evidence