Lecture 4: Eye witness testimony Flashcards
what are EWT used to show
- What can cognitive psychology tell us about eyewitness perception and memory?
- How have psychologists used this knowledge to improve criminal investigations?
how many ewt errors are involved in DNA exoneration cases
- Eyewitness testimony errors involved in 75% of DNA exoneration cases
why are EWT important
- May be the only evidence available in absence of other forensic evidence
- Identifying suspects, describing suspects, building composites, describe an event, set a timeline…etc
how man people in a jury gave a guilty sentence in Magnussen et al., 2010 when EWT was used
78%
what happened and was found in Magnussen et al., 2010)
- Jurors, judges, and general public have limited knowledge of factors affecting eyewitness testimony (Magnussen et al., 2010)
o 18 statements on eyewitness testimony
o Correct responses: Public – 56%; Jurors – 57%; Judges – 65% - Juries cannot judge eyewitnesses
- Eyewitness picked out perp fro line –up. Discredited witness – poor eyesight, not wearing glasses on day, admitted to picking defendant because he looked nervous. - Jurors, judges, and general public have limited knowledge of factors affecting eyewitness testimony (Magnussen et al., 2010)
o 18 statements on eyewitness testimony
o Correct responses: Public – 56%; Jurors – 57%; Judges – 65% - Juries cannot judge eyewitnesses
- Eyewitness picked out perp fro line –up. Discredited witness – poor eyesight, not wearing glasses on day, admitted to picking defendant because he looked nervous.
how did Buckhout, 1980 test an EWT
They showed a recorded crime on TV
– 5 min segment, 2 mins later they showed a line up.
what are the results of Buckhout, 1980
o 2000 people rang in, 1,800 made incorrect ID
o 14.1% gave the correct answer
what this showed:
- Transference – remember seeing the face and confuses and innocent bystander with the actual perp – may incorrectly attribute a person’s familiarity to the crime context
- On tv only a 2 min delay so not a long time to forget the answer
- Good example of unconcious tanscripts may mistake sensor familiarity for being the person asociate event with person, not necasarily concious
what are the stages which make EWT unreliable
perception
encoding
storage
retrieval
how can the retrieval stage affect our memory of EWT
whats happenign at this stage that can affect accuracy, eg how long in between
how can an initial perception stage affect an ewt
She thinks theres an initial perception stage, eg how far bac you were in lecture theatre, very far back in lecture etc not listening
what is an estimator variable
- Estimator Variables: variables that are not under the control of the justice system
what are some examples of estimator variables
Eyewitness factors: emotional state, intoxication etc
Perpetrator factors: disguise, facial distinctiveness
Situation factors: exposure duration, distance, retention interval
what are system variables
- System Variables: variables that can be controlled as part of the legal system
what are some system variable examples
Line up construction: size, structure, and selection of fillers
Cognitive Interview: procedures and training
oCan be controlled by the ‘system’ the justice system
-Might need to know for exam whether estimator or system variable for exam
what type of varuables can be controlled by the criminal justice system
system variables
explain what encoding is
- We remember more of an event, the more we know about it in advance (Ornstein et al., 2006)
- We remember more information that is consistent with our scripts (Holst and Pezdek, 1992)
o Events
o Faces
what factors affect encoding
- We remember more of an event, the more we know about it in advance (Ornstein et al., 2006)
- We remember more information that is consistent with our scripts (Holst and Pezdek, 1992)
o Events
o Faces
what happened in holst(?)s experiment about encoding
- E1: examine people’s scripts for three types of robberies – high agreement about what typically comprises each type of robbery;
- E2: Mock trial – script relevant items stated, some unstated. 1 wk later – unstated items were recalled as having been stated;
- People incorporate these scripts into their memory
define change blindness
difficulty in detecting major changes in perceptual environment (e.g, continuity errors in films); Identity change also – even when differences in physical appearance as substantial – not just a result of inattention (about 19% notice change)
name soe explanations of change blindness by simon 2000
o Overwriting o First Impressions o Nothing is stored o Storage but no comparison o Feature combination
what is Levin et al., 2002 study about and what are the results
- Change blindness correlates with memory (Levin et al., 2002) o Primed (65%) versus not primed (12.5%)
how is change blindness linked to EWT
- Those who notice change are more likely to identify the correct perp in a line-up
- Feature combination linked to eye-witness memory in which combination of perceived event with post-event suggestions
what are the situation and perpertrator factors of change blindness
- Duration of exposure (Memon et al., 2003)
- Distance from incident (Lindsay et al., 2008)
- Awareness of the incident
- Facial distinctiveness (Busey & Tunnicliff, 1999)
- Disguises (Patterson & Baddeley, 1977)
o All of these factors how you percieve the occasion
how can changing your appearance affect recognition and encoding
- Recognition of faces is poor if the face changes between the encoding and the presentation of possibilities.
- This could be a change in glasses, hairstyle, beard or even expression.
- Encoding specificity principle
o People better at remembering events when encoding and retrieving conditions are similar - Reinstatement of environmental context (contextual reinstatement) – episodic memory can be cued by environmental stimuli
- Application in cognitive interview and guided memory technique
- Usefulness in eye-witness testimony inconsistent – positive vs null results in some studies - some suggest arousal could be a moderator (Brown, 2003)
- If face at recognition and encoding are different it will be harder to remember an event
define the - Encoding specificity principle
o People better at remembering events when encoding and retrieving conditions are similar
what do EWT inconsistencies suggest
- Usefulness in eye-witness testimony inconsistent – positive vs null results in some studies - some suggest arousal could be a moderator (Brown, 2003)
- If face at recognition and encoding are different it will be harder to remember an event
name how you can test perception change of appearance
- Study phase:
o Participants viewed a staged robbery
o For 1/2 of the participants the robber
o wore knit pullover cap - Test phase:
o Identified perpetrator from video lineup
o 45% no hat group vs 27% hat group
name some witness factors which affect encoding and EWT
- High stress negatively impacts memory (Deffenbacher et al., 2004)
o Soldier study (Morgan et al., 2004)
o Soldier study: military personnel; undertook survive training (interrogation by guard); Live line up, photo spread, sequential photo; High stress less accurate by ~ 25- 30% (less true positive IDs, and more false positive responses) - Aging effect (Memon et al., 2003)
o Age effect: interaction with delay (no different when short – 35mins); older witnesses (60+) less accurate after 1 wk delay - Ethnicity-bias
- Stressfull situ at time of retrievala nd encoding
- More likely to rmemeber/ recognise a face of same ethnicity of your own
- Age= older less likely to remember
Describe the Soldier study (Morgan et al., 2004)
- High stress negatively impacts memory (Deffenbacher et al., 2004)
o Soldier study: military personnel; undertook survive training (interrogation by guard); Live line up, photo spread, sequential photo; High stress less accurate by ~ 25- 30% (less true positive IDs, and more false positive responses)
Stressfull situ at time of retrievala and encoding
describe the - Aging effect (Memon et al., 2003)
o Age effect: interaction with delay (no different when short – 35mins); older witnesses (60+) less accurate after 1 wk delay
- More likely to remember/ recognize a face of same ethnicity of your own
describe ethnicity bias
- More likely to rememeber/ recognise a face of same ethnicity of your own
how does alcohol/ witness intoxication affect EWT
o Reduced attentional capacity
o Intoxicated Ss had worse memory for peripheral details; less likely to express uncertainty about details (Evans et al., 2009)
o “Alcohol myopia theory” suggests that alcohol increases focus on central detail (Josephs, 1990)
- Impact of alcohol both at encoding and retrieval
- More atttention to centeral events
- Less peripheral memory
- Reduces ability to process and extract meaning from scenes
- This is very common to have intoxicated victims
- Sexual and non-sexual crimes = more likely to be intoxictaed
- Intoxicated victims: ~ 40-50% of victims of sexual and non-sexual crimes
what is alcohol mypoia
alcohol affects cog function either by: restricting range of cues that can be perceived or reducing ability to process and extract meaning from scenes
- Intoxicated witnesses: ~ 72% investigators indicate this is very common
what are the 2 types of storage in EWT
- Retention interval
- Post event suggestion
define Retention interval (storage)
o delay decreases the amount of information that can be recalled
define post event suggestion (Storage)
o Exposure to media reports
o Co-witness discussions
o Choice blindness
o Media reports- eg 9/11 , can lead to miss information
what are the 2 main types of Retention interval (storage)
- Face recognition (Wells et al., 2006)
- Event details (Ebbesen & Rienick, 1998)
what is Retention interval (storage) face recognition according to wells et al 2006
- Face recognition (Wells et al., 2006)
o Immediate, longer delays
o Fewer correct IDs (~ 51% vs 61%)
o Increase false IDs (~32% vs 24%)
o Increased delay -> longer response time
what is Retention interval (storage) event details according to (Ebbesen & Rienick, 1998)
- Event details (Ebbesen & Rienick, 1998)
o Immediate vs 4 week delay
o Reduction in number of recalled facts
o % error consistent
define post event suggestion nd misinformation
- Many witnesses will see or read news reports of events
o Recall post-event information vs what really happened
o Proneness to false memory associated with personality traits (Ost, Granhag, Udell, & Hjelmsater, 2007)
o Social influences and ‘shared reality’ - Delay between event and information plays a role in acceptance of misleading suggestions
o Susceptibility of memories to retroactive interference increases as memories are forgotten (Reyna, 1995)
how does Delay between event and information play a role in acceptance of misleading suggestions
o Susceptibility of memories to retroactive interference increases as memories are forgotten (Reyna, 1995)
what did reyna find about post event suggestion
people who scored higher on fantasy = more likely to be influenced by events
define what happens when 86% of witness discuss their memory with co-witnesses
o Causes conformity (Gabbert et al., 2004)
o Can cause misinformation (Wright et al., 2000)
o Couples vs strangers (French et al., 2008)
what did OST find about post event suggestion
false memories higher in those who score higher on personality traits related to dissociation and fantasy proneness
define choice blindness
- Difficulty detecting manipulation of a choice they made (Sagana, Sauerland, Merckelbach, 2014)
o Blindness for recognition decisions: ‘This is the person you selected’
o 39% - 68% of manipulations remained undetected
o Many people think they would be able to detect this manipulation
what are the different types of retrieval
- Verbal overshadowing
Mug shots:
o Repeated exposure to suspect
o Filler choice
o Facial composites
what did schooler 1990 find about verbal overshadowing
- When you describe a face you become less accurate at recognising it subsequently (Schooler et al., 1990)
o By a reduction of 26% accuracy (64% control versus 38% face verbalization)
when did Meissner & Brigham, 2001 say that verbal overshadowing is most likely to occur
- More likely to occur when (Meissner & Brigham, 2001):
o Identification immediately followed the description
o Participants were given more ‘structured’ recall (as opposed to ‘free recall’)
what happened in the schooler study 1990 of verbal overshadowing
: subjects viewed a 30s video tape bank robbery, then 20 min unrelated task reading several passages, and then answered questions.
- Then either verbal description of face condition (5 mins to write description of facial features) or control condition (unrelated activity).
- Identification task: Shown target and 7 distractors
- Structured recall encourages recall of incorrect or inaccurate information
why does schooler 1990 say that verbal overshadowing occurs
o No relationship between description quality and recognition performance.
o Specific to verbal description, mentally revisualising the face does not interfere with recognition.
o Specific to face verbal description, other verbalizations do not interfere with face recognition
o (Schooler et al., 1990)
o Subjects are therefore not relying in the verbatim description exclusively for recognition
o By providing a verbal description of face interferes with recognition (this doesn’t occur with thinking about face etc)
what are the 2 types of face processing
Expert processing
inexpert processing
describe the features of expert processing of faces
Unconscious Non-verbal Configural (?) Disrupt inversion Holistic Gestalt Global Good for face perception Wide perceptual field
describe the features of inexpert processing of faces
Conscious Verbal Featural Not-disrupted by inversion Each part in isolation Separate Local Bad for face perception Small perceptual field
what is hierarchical stimuli in face processing
o Global feature made up of lots of local features.
o The whole is more than the sum of its parts!
- Identifying parts causes processing change
- Local or global Navon letters followed by face recognition (Macrae & Lewis, 2002)(see graph)
o Huge difference VO priming using verbal overshadowing which is a huge disadvantage
what did Memon et al., 2002 find out about mugshots
- Repeated exposure to a suspect increases the probability of identification and confidence (Memon et al., 2002)
o Even if wrong
what did Wells, Charman, & Olson, 2005 find out about mugshots
- Building face composites can harm line up identification performance (Wells, Charman, & Olson, 2005)
o The composite building process can harm builder’s memory for the face
o Reduction in chance of later identifying the original face
what is o Suspect filler similarity when using mugshots
o Suspect filler similarity – too highly similar problematic – eyewitnesses experience difficulty in identifying the suspect. A higher identification rate is observed when line ups are moderately similar and not highly similar
what happens when similar fillers are used in line ups (Fitzgerald, Oriet, & Price, 2015)
- Too highly similar = problematic – eyewitnesses experience difficulty in identifying the suspect.
- Moderately similar = higher identification and fewer false positives
- (Fitzgerald, Oriet, & Price, 2015)
- Culprit absent – choice of innocent suspect
what happens when people have distinguishable marks in mugshots
- May attempt to replicate distinctive feature across all other distractor faces (time consuming)
- May occlude this distinctive feature by pixelating or placing a solid black box cover it (however, this can sometimes occlude elements important in facial recognition if for example, too close to the eyes)
describe the processes of creating face composites and how they can be affected
- Building face composites can impair subsequent identification performance (Wells & Charman, 2016)
o Composites generally poor likeness to face – but does depend on technique used
o Identification of target under control vs composite - Combining face composites yields improvements in face likeness (Bruce, Ness, Hancock, Newman, & Rarity 2003)
o A morph of all four composites created a better likeness that individual composites
o In a line-up, 4-morphs performed better (produced more correct choices and fewer false positives) than individual composites
what methods are used in enhanced cognitive interviews
- Mental context reinstatement
- Report everything
- Recall in a variety of temporal orders
- Change perspective
how many police interviews in the UK use enhanced cognitive interviews
o Very few police offices engage in it
o 83% interviews in UK are not done with the cognitive interview
Because it is time consuming and the pressures of resources are too great
why do many police interviews not use enhanced cognitive interviews?
Because it is time consuming and the pressures of resources are too great
what are the Memory factors in UK law
- Amount of time under observation
- Distance from suspect
- Visibility (night, day, lighting?)
- Obstructions to the view of the witness
- Known or seen before (when and where)
- Any special reason for remembering the suspect?
- Time lapse since witness saw suspect
- Error or discrepancy between the witness’s description in their first and subsequent accounts
define Postdiction Variables:
measurable variables that correlate with eyewitness accuracy
how can police move towards a proactive approach
use postdiction variables
name some postdiction variables
o Individual differences: facial recognition, time estimation, memory ability
o Self-report decision processes
o Response latency (how long to make an identification)
how do we find out the exposure duration in real life and in the lab
o In lab, predetermined
o In real world, must rely on eye-witness – but how accurate
what does suggest about time estimation tasks of exposure to a crime
- People vary hugely in time estimates, some being really poor judgers of time (16% to 496% of actual duration), but consistent across time estimation tasks
o Can we determine the reliable from the non-reliable?
o Or deduce the possible error and adjust accordingly? - (Attard & Bindemann, 2013)
what did Memon et al., 2003 find out about time estimation in ewt
- Exposure duration to the event and the face is correlated with face recognition performance (Memon et al., 2003)
o More exposure to target (45s vs 12s) = increase identification
what are the limitations of self reported measures in facial recognition
- Self- report measures of face recognition skill linked with identification accuracy for culprit (Olsson & Juslin, 1999).
o Self-reports are subjective
o How much awareness do we really have? (Bindemann et al., 2014)
