Lec 4/ Ch 5 Flashcards
1
Q
The Origins of Criminal Profiling
- 1888 Case: Jack the Ripper
- 1940 Case: NYC’s Mad bomber
- 1970 The FBI and Dr. Douglas
- The RCMP’s Violent Crime Linkage Analysis System (ViCLAS)
- Linkage blindness
A
The Origins of Criminal Profiling
History of criminal profiling
- Investigative technique for identifying the personality and behavioural characteristics of an individual based on analyses of the crimes he or she commits
- Mostly used in cases involving violent serial crimes
- Originally used for narrowing the list of suspects; but now serves additional purposes
- Help set traps to flush an offender out
- Determine if threatening communications should be taken seriously
- Advise on how to interrogate subjects
- Advise how to break down defendants in cross-examination
- RCMP uses the term “criminal investigative analysis”
- Not just from FBI ppl
Early attempts at criminal profiling
- 1888 Case: Jack the Ripper
- Series of murders committed at London
- Victims were all women and mutilated
- Offender sent a letter to the newspaper and signed it
- Police surgeon to looked at wounds to see if it provides clues about the offenders’ behavior and personality
- Jack the Ripper is still not caught
- 1940 Case: NYC’s Mad bomber
- Offender began detonating bombs in public places around NY
- NYPD asked a local forensic psychiatrist – Dr. James Brussel for help
- Dr. Brussel developed a profile for the offender
- Middle aged male
- Suffer from paranoia
- Pathologically self-centered
- Reasonably educated
- Unmarried, maybe a virgin
- Roman catholic
- Would wear buttoned-up double breasted suits
- The offender was arrested and actually wore double breasted suit
The FBI and Dr. Douglas
- FBI’s behavioral sciences unit developed a criminal profiling program in 1970s
- Behavioral Sciences Unit in 1972
- Now called the “Behavioral Research and Instruction Unit”
- Profiles were produced in a systematic way by law enforcement agency
- Training was provided on how to construct criminal profiles
- Similar units have started around the world
- 1976 – John Douglas began interviewing serial predators across the U.S.
- The goal was to develop a database that could be used to help law enforcement
- The killers talked a lot with Dr. Douglas
Investigative psychology
- David Canter made some of the key advances in criminal profiling since early 1990s
- He is the founder of investigative psychology
- Canter was involved in the John Duffy/Railway Rapist rape/murder case
- Canter was called to provide a profile of the unknown offender
- He used his knowledge on human behavior from env psych
- He was successful
- He made advances in inductive criminal profiling
Box 3.6 The RCMP’s Violent Crime Linkage Analysis System (ViCLAS)
- RCMP used modern computer tech to develop an automated system for linking serial crimes
- Violent Crime Linkage Analysis System (ViCLAS) – developed by RCMP to collect and analyse info on serious crimes across Canada
- Issue police encounter when faced w/ possible crime series is linkage blindness
- Linkage blindness: police can’t link geographically dispersed serial crimes committed by the same offender due to lack of communication among police agencies
- ViCLAS was developed to prevent linkage blindness
- ViCLAS is san electronic form officers fill out, it captures crucial info on serious crime (ex. homicides, sexual assaults, missing persons, non-parental abductions)
- Qs ask about Offender’s b, victim and available forensic info
- Info is then entered into a computer and downloaded to a centralized databased to compare w/ other crimes
- ViCLAS analysts determine if there are possible crime linkages
- If there is crimes are highlighted as a series and relevant police agencies are notified and asked to share info
- Some rs qs the reliability of the d at in the system
- Other rs says ViCLAS is one of the best crime linkage analysis system in the world used by UK, Australia and Germany
2
Q
2 Methods of criminal profiling
- Deductive profiling method
- Issue w/ DPM
- Inductive profiling method
- Findings from Aitken et al 1996
- Issue
- Organized/disorganized model
- Organized vs Disorganized b
- Organized vs disorganized characteristics
- Canter et al. 2004
- methods
- results
- Investigative psychology approaches to profiling
- Goodwill et al 2016
- 3 clusters for victim selection
- 3 clusters for background characteristics
- Goodwill et al 2016
A
2 methods: Deductive & Inductive profiling method
- Descriptions of the profiling process are generally vague
Deductive profiling method (SHolmes)
- Use logic to predict offender’s background based on evidence left at the crime scene by the offender
- Ex: Victim noticed that offender had short fingernails on R hand; long nails on left; Concluded the offender is a guitar player
- Issue: the logic arguments can be wrong
- Offender was not a guitar player, but works to repair old tires
Inductive criminal profiling (John Douglas)
- Profile offender’s background characteristics by comparing the offender’s crimes w/ similar crimes committed in other solved cases (past cases)
- Aitken et al 1996
- Developed stat profile of a child murderer
- .96 probability that the offender will know the victim
- .92 the offender will have a previous criminal conviction
- .65 the offender is under age 20
- Downside: Crimes that don’t fit pre-existing profiles are difficult
- Best to use both methods in criminal profiling
Inductive method: Organized/disorganized model
- Dev by John Douglas and his colleagues in the 1980s by interviewing offenders
- Organized-disorganized model: crime scenes and backgrounds of serial offenders can be categorized as organized or disorganized
- These categories are idealized, usually it can be a mix of both
- Organized crime scene b: 1st degree murder (ex. Zodiac Killer)
- Planned offence
- Use restraints on victim
- Ante-mortem (b4 death) sexual acts committed
- Use vehicle
- No post-mortem mutilation
- Corpse not taken
- Little evidence left at scene
- Disorganized behavior
- Spontaneous offence
- No restraints on victim
- Post-mortem sexual acts
- No vehicle used in crime
- Post-mortem mutilation
- Corpse/body parts taken
- Evidence left at scene
- Organized background characteristics reflect a methodical person
- High intelligence
- Skilled occupation
- Sexually adequate
- Lives w/ a partner
- Geographically mobile
- Lives and works far away from crimes
- Follows crimes in media
- Maintains residence and vehicle
- Disorganized characteristics disturbed person who is suffering psychopathology
- Low IQ
- Unskilled occupation
- Sexually inadequate
- Lives alone
- Geographically stable
- Lives and works close to crimes
- Little interest in media
- Does not maintain residence and vehicle
- When encountering disorganised crime scene, the investigator should profile the background characteristics of a disorganized offender
- For organized crime scenes, organized background characteristics
- FBI has revised this model to account for many offenders who have mixtures of organized and disorganized features
Canter et al. 2004 - Organized/Disorganized Serial Crimes: Model or Myth? – not in TB
- Categorize aspects as organized vs disorganized
- Review multiple homocides
- Use stats to determine which aspects cooccur, which do not
* Expected the organized aspects clump together; disorganized – another cluster
- Use stats to determine which aspects cooccur, which do not
- Result:
- graph showed that organized are clumped together (happen more frequent)
- disorganized aspects spread over the graph (happen less frequent)
- IOW: you cannot distinguish serial killers based on their level of organization
- We should use disorganized aspects to distinguish serial killers
- Ex. SK 1 left a weapon in the victim
- SK 2 is unlikely to do this
- We should use disorganized aspects to distinguish serial killers
- Used a multidimensional scaling technique (state tech) to test the hypothesis that crime scene behaviors can be classified using Organized/Disorganized categories
- Behaviors closer together in statistical space, tend to co-occur.
- Behaviors in the center of the plot are observed more frequently.
- H: If the Organized/Disorganized model is valid, then “organized” and “disorganized” behaviors should clump together in different regions of the graph.
- Results:
- Graph shows 39 different characteristics of crime scene
- Elements occur frequently are seen/cluster in the middle; further out = less frequent
- Organized elements are in the middle = tend to occur more; disorganized are spread across the graph
- Conclusion: serial crimes are often very organized (criminal was not detected until there were a series of killings; serial killers are usually organized).
- All killers tend to have a level of organization
-
what distinguishes them are the elements of disorganization (less frequent)
- aka type/quality of disorganization
Investigative psychology approaches to profiling
- These approaches are similar to inductive approaches, like the organized-disorganized model where they rely on solved cases to develop categories of crime scene b and background characteristics
- But investigative psychologists emphasis more on rigorously testing the validity of the categories proposed and the linkage b/w crime scene b and background characteristics
- Goodwill et al 2016
- Used stat technique – cluster analysis, to find out different ways sex offenders search, approach, and assault victims
- Ex. for victim selection, the analysis identified 3 clusters
- Offenders who predominantly target adult females w/ specific physical features (telio specific)
- Offenders who target child and teens w/ physical features (pedo/hebe specific)
- Offenders who don’t have preferred victim type (non-specific)
- Cluster analysis was also used to identify different clusters of background characteristics
- Socially competent offender
- Anti-social generalist offender
- Sexually deviant offender
- Also used another stat technique = multiple correspondence analysis (MCS) to examine the inter-relationships b/w all of the clusters
- There is relationships b/w clusters of crime scene b and clusters of background characteristics
- Ex. offenders who adopt hunter search strategy (actively seeking victims w/in short distance from their home), they are
- telio specific (target adult F w/ specific physical features)
- use home intruder approach style (assault victim in their home)
- employ and assault strategy based on violence and control
- These offenders tend to be sexually deviant, w/ the following background characteristics
- Socially isolated
- Persistent sexually deviant fantasies
- Voyeurism in adulthood
- Poor self-image in adulthood
3
Q
The Validity of Criminal Profiling
- What officers say about it?
- 4 criticisms from TB
- Trait model lack support, ignores situational factors
* Counterargument:- person-situation debate
- Replication crisis in psychology
- Trait model lack support, ignores situational factors
- Profiling lack support
* 2 assumptions
* Goodwill & Alison 2007
* Counterargument- Gosling et al
- lens model
- Profiling lack support
- Profiles are vague and can fit many suspects
* Alison et al. 2003
- Profiles are vague and can fit many suspects
- Professional profilers vs untrained individuals
* Snook et al 2007
- Professional profilers vs untrained individuals
A
What officers say about it?
- valuable, can mislead
Profiling has limitations
- based theory model that lack strong empirical support
- Core assumption of profiling no empirical support
- profiles ambiguous, can fit many suspects
- Professional profilers may be no better than untrained indiv at constructing accurate profiles
theoretical basis
- TB cites “Classic trait model” → Assumes that the primary determinants of behavior are stable, internal traits; Ignores situational factors
- TB is wrong:Profiling is based on classical trait model but personality psych looks at BOTH trait and situational factors
- This is based on the person-situation debate:
- Mischel: Found that personality only predict real world problems w/ correlation r ≤ .30 → “a personality coefficient”
- Replication crisis in psychology
- It’s not affecting personality psych as. Bad
- 80% of the time replication was successful for Big 5 trait model
Is there empirical support for profiling assumptions?
- Rs tested 2 related assumptions
- Offenders behave in a stable fashion across the crimes they commit
* Partial support
- Offenders behave in a stable fashion across the crimes they commit
- Reliable relationships exist b/w the way in which offenders commit their crimes and their background characteristics
* Less support- Previous profiling rs may not hv found results that were supportive of profiling assumption b/c the studies oversimplified the profiling process
- Reliable relationships exist b/w the way in which offenders commit their crimes and their background characteristics
Profiling relies on a “lens model”
- Point: ppl can use env cues to infer stranger’s traits
- Criminal profilers use the “crime scene” as the lens in this approach
What is the impact of ambiguous profiles?
- Profiles can be ambiguous that they can fit many suspects
- IOW: study supports the that ambiguous profiles can be interpreted to fit more than one suspect
- Need to be careful when interpreting results:
- Recent study: profiles used in investigations did not have the same degree of ambiguity
How accurate are professional profilers?
- profilers are slightly better, however the difference is not hugely significant
Point: criminal profiling is based on a strong theoretical basis
4
Q
Geographic Profiling
- Definition
- purpose
- assumption
- how it was built
- How is the graph read
A
Geographic Profiling
- Geographic profiling: uses crime scene locations to predict the most likely area where and offender resides
- Geographical profiling allows help prioritize potential suspects
- Done by ranking the suspects based on how close they live to the predicted home location
- Basic assumption: most serial offenders do not travel far from home to commit crimes
- So it should be possible to make a reasonable accurate prediction about where the offender lives
- Rs supports this assumption
- Serial offenders tend to be consistent in their crime site selection choices
- They commit crimes close to where to live
- For travelling offenders (aka travel in a particular direction to commit crimes), geographical profiling is not helpful
- Case: Yorkshire Ripper in England
- After 5 yr of unsolved murder, the advisory team reviewed the case
- Investigators felt the offender lived in a different part of the country where the crimes were happening
- The advisory team think the offender is a local man
- The team made a geographic profile, plotted the 17 murders, calculated the center of gravity
- It was near Bradford, which is a city close to where most murders happen
- Peter Sutcliffe was arrested
- Ppl built computerized geographic profiling systems to help profiling task
- Input Locations of linked crime sites in system, which appear as points
- System do calculations based on offender spatial behavior, and show the probability the offender lives at particular points where the offences take place
- Each location on the map has an overall probability, and a color
* Ex. top 10% = red
* Colored band = probability the offender lives in the area
- Each location on the map has an overall probability, and a color
- Police can use this map to prioritize their investigations
- These predictions are accurate and help the police
- Rossmo 2000
- Looked at crimes from 15 serial killers
- Calculated the hit % values (% of space that needs to be searched to locate offender’s home base)
- Using this system, the offender’s residence is found in 12% of the time
- Canter et al 2000
- Looked at 79 serial killers
- 51% of offender residences were located in top 5% of prioritized activity space
- 87% of offender residences were located in top 25% of prioritized activity space
5
Q
Eyewitness testimony: the role of memory
- What does eyewitness testimony depend on
- Atkinson Shiffrin Model of Memory
A
- Eyewitness testimony depends on memory
- Atkinson Shiffrin Model of Memory
- Info flow through a series of stages to permanent storage in memory
- Sensory input → sensory memory (it attended to) → STM (if rehearsed) → LTM
- Sensory memory: large capacity, lasts 1-2 sec
- STM: capacity 5-10 items, last 15 sec
- LTM: unlimited capacity, lasts forever
6
Q
4 Factors that influence memory
- What happens after the event (ex. Suggestive/ leading qs in interviews)
- more stress →? memory
- cross-race effect
- Weapon focus
Eyewitness myths
- 1 Memory is not a video recording
- 2 Weapon focus (ex. monkey business illusion)
A
- Wells & Loftus 2003 - Mem can be influenced by subsequent events
* Ex. the way qs are asked in interviews influence what one remembers
* What is remembered can be inaccurate
- Wells & Loftus 2003 - Mem can be influenced by subsequent events
- Loftus and Zanni 1975 - Suggestive/ leading qs can influence responses
* Changing one word in a qs can lead to an increase in incorrect ans
- Loftus and Zanni 1975 - Suggestive/ leading qs can influence responses
- Deffenbacher et all 2004 - Greater stress results in poorer memory for both the perpetrator’s appearance and other crime details
- Meissner & Brigam, 2001 - cross-race effect
* Eyewitnesses are better able to correctly identify member of their own race compared to that of a different race
- Meissner & Brigam, 2001 - cross-race effect
- Saunders 2009 – Weapon focus
* If a weapon is present, witnesses often fous on it, and have a less reliable memory for other aspects of the crime
- Saunders 2009 – Weapon focus
Eyewitness myths
- 1 Memory is not a video recording
- It can change each time we retrieve the event
- Some things are added or guessed b/c we cannot remember all the details
- Mem fallibilities in daily life are insig (ex. coffee from Tims vs Starbucks)
- This is sig if police tries to arrest suspect (ex. is the suspect L or R handed)
- 2 Weapon focus
- Actually, if a weapon is present, people will focus on it and remember other aspects of the crime to a lesser degree.
- Monkey business illusion: only pay attention to things that are relevant to us
- Rs hv confederates (red and blue teams) bounce beachball
- Ppl were told to count the # of times the blue team passes the ball
- When asked about whether they see gorilla
7
Q
2 types of eyewitness memory
- Recall memory
- Recognition mem
A
- Recall memory: reporting details of a prev witnessed event or person
* Ex. describe what the perpetrator did and what he/she looked like
- Recall memory: reporting details of a prev witnessed event or person
- Recognition mem: determine whether a prev seen item or person is the same as what is currently being viewed
* Ex. hearing a set of voices and identifying the perpetrator’s voice; identify clothing worn by perpetrator during the crime
- Recognition mem: determine whether a prev seen item or person is the same as what is currently being viewed
8
Q
Box 5.2 Forensic Psych in the spotlight – Canada’s witness protection program
- What it does
A
- Some provide info that put them in harm (ex. organized crime, terrorist informants)
- Program offer ST protection and permanent relocation (ex. identity changes)
- The Safer Witness Act (2012) from Toews amends the Witness Protection Program Act
- It makes it easier for witnesses in provincial program to get new identities
- It has new restrictions on disclosure of info to make the program more secure
- It increases the amount of time emergency protection maybe provided to witnesses
- Makes the program available to those referred from National Defence and CSIS
- The Act was introduced b/w someone in the protection program was murdered in 2007
- Some ppl in the program sued for improper treatment
- Others were kicked out of the program
9
Q
How do We Study Eyewitness Issues?
- Describe The Laboratory Simulation
- 2 IV
- Estimator v
- System v
- 3 types of DV
- 2 formats of recall
- Open ended recall
- Direct qs recall
- Recall is examined for
- Recognition of perpetrator
- Lineup
- 2 things recognition is examined for
A
- Examine data from actual crimes
- Ex. archived police reports
- Ex examine witnesses in naturalistic env by following police to crime scenes and interviewing witnesses after the police have done their job
- Lab simulations – most common
The Laboratory Simulation
- Have ppl watch a critical event (ex. crime)
- now a witness, the participant is asked to describe what happened
- Witness then examine a line up
- IV: 2 main types
- Estimator v: v that are present at the time of the crime and cannot be changed
* Ex. age of witness, amount of lighting, presence of weapon, whether the witness was intoxicated
* Eyewitness accuracy can only be estimated after the crime
- Estimator v: v that are present at the time of the crime and cannot be changed
- System v: v that can be manipulated to increase (or decrease) eyewitness accuracy
* Ex. type of procedure used by police to interview the witness or the type of line up procedures used to present the suspect to the witness
- System v: v that can be manipulated to increase (or decrease) eyewitness accuracy
- variables in control of the justice system.
- Both estimator and system v can be manipulated in lab studies
- System variable are variables that can be manipulated and can affect recall/eyewitness accuracy, they are usually controlled by the justice system in some way
3 DV
- Recall the event/crime
- Recall the perpetrator
- Recognize the perpetrator
- 2 types of recall event/crime/perpetrator
- A. Open ended recall/ free narrative: witness asked to write or tell what they remember about the event/perpetrator w/o officer/experimenter asking qs
- B. Direct qs recall: witness are asked a series of specific qs about the crime/perpetrator
- Ex. witness maybe asked about the color of the getaway car of length of perpetrator’s hair
- How the witness’s recall of the crime/perpetrator is examined
- The amount of info reported (ex. descriptors of crime/perpetrator)
- The type of info (ex. proportion of peripheral vs central details; proportion of perpetrator vs env details)
- Accuracy of info reported (ex. proportion of correct vs omission errors/fail to report vs commission errors/falsely reported)
Recognition of the Perpetrator
- Line up: set of ppl presented to witness, who must state whether the perpetrator is present, and which person it is
- Other types: voice lineup, clothing lineup
- How the witness’s recognition response is examined
- Accuracy of decision (ex. rate of correctly identifying perpetrator is/ is not in the lineup)
- Types of errors made (ex. rate of identifying an innocent person/ stating the perpetrator is not present when he/she is present in the lineup)
10
Q
Recall Memory
- Fisher et al 1987
- Methods
- General questioning format
- 5 key findings
- Memory conformity
A
- Insufficient info → case not solved, perpetrator can keep committing crimes
- Inaccurate info → wrong person is the suspect
Interviewing Eyewitnesses
Fisher et al 1987
- Looked at the techniques police used to interview eyewitnesses
- Analysed 11 tape recordings from PD
- General interview format:
- Introduce themselves
- ask open-ended qs
- Ask direct qs to determine specific info (ex. age, height of perpetrator)
- ask the eyewitnesses if there was any additional info they can remember
- 5 key findings
- Rs found police often interrupted eyewitnesses when they provided an open-ended recall report
* Issue: Police limits the amount of info eyewitnesses remember by preventing them from speaking or distracting them with questions
- Rs found police often interrupted eyewitnesses when they provided an open-ended recall report
- Police question eyewitnesses w/ short specific qs
* leads to short superficial ans compared to open-ended qs;
* Officers may not ask a relevant qs that provide critical info- Ex. the perpetrator may hv a tattoo, but the officer does not ask about this → so the eyewitness may not report it, and police miss a descriptor that help narrowing the pool of suspects
- Police question eyewitnesses w/ short specific qs
- Officers ask qs in predetermined or random order that was inconsistent w/ the info that witnesses were providing at the time
* Ex. Officer may ask a qs about the perpetrator’s voice while the witness was describing the perpetrator’s clothing
* Fisher et al- Mixing visual and auditory qs decrease recall by 19%
- Officers ask qs in predetermined or random order that was inconsistent w/ the info that witnesses were providing at the time
- Officers tend to ask qs that are “leading”
- contamination of co-witnesses → memory conformity
* Crimes can hv multiple witnesses
* Witnesses can be contaminated if they know what other witnesses reported
- contamination of co-witnesses → memory conformity
- Memory conformity: when one witness reports influences what another witness reports
11
Q
Recall memory - misinformation effect
- Loftus and Palmer 1974
- Part 1: Method & results
- Part 2: Method & results
- Misinformation effect/post-event info effect
- Loftus 1975: Method & results
- 3 main explanations for misinformation effect
- 3 ways misinformation effect is seen in real life
A
The Leading Question: The Misinformation Effect
- Loftus: her studies show that witness’ recall report can be altered by the phrasing of a qs
Loftus and Palmer 1974
- Part 1: Method
- Had uni students watch a videotape of a car accident
- Ppl were then asked identical qs w/ a variation in one critical word “hit”
- How hast were the cars going when they hit each other?
- Hit was replaced w/ either smashed, collided, bumped, contracted
- Results: Participants reported highest rate of speed when “smashed” was used, lowest rate when “bumped” and “contacted” were used.
- Part 2
- Called ppl back a wk later and asked if they had seen any broken glass
- Results: Ppl who were questioned w/ the word smashed were more likely to recall seeing broken glass than others
- But there was no broken glass in the video
- IOW: the wording of a qs can influence memory for the incident
- Misinformation effect/post-event info effect: witness who is presented w/ inaccurate info after an event will incorporate that misinformation into a recall task later
Loftus 1975
- 40 students watched 3 min clip, they were interrupted by 8 demonstrators
- After the clip, ppl were asked 20 qs
- ½ were asked: was the leader of the 4 demonstrators who entered the classroom a male?
- ½ were asked: was the leader of the 12 demonstrators who enter the classroom a male?
- Other qs were the same
- 1 wk later
- Ppl were asked 20 new qs about the film
- Critical qs: how many demonstrators did you see entering the classroom
- Those who were asked about 12 demonstrators report seeing avg of 9d
- Those who were asked about 4 demonstrators report seeing avg of 6.4d
- IOW: incorporating # of demonstrators into the qs affect the # of demonstrators the witnesses recall seeing later
Explaining the misinformation effect
- 3 general positions
- Misinformation acceptance H: explains the misinformation effect, states the incorrect info is provided b/c the witnesses guess what the officer/experimenter wants the response tb
- Source misattribution H: explains the misinformation effect, states that the witnesses can recall both the original/accurate and the inaccurate memory; but witness can’t remember where each memory came from
- Memory impairment H: Loftus’ explanation of the misinformation effect; original mem is replaced w/ new incorrect info
- We still don’t know which position is correct
But the misinformation effect is real, and seen in real life
- Officer may make an assumption on what happened and phrase a qs that aligns w/ their assumption
* Ex. Officer asks “Did you see THE gun?” rather than the neutral qs “Did you see A gun?”
- Officer may make an assumption on what happened and phrase a qs that aligns w/ their assumption
- There maybe more than one witness and the witness overhears others’ statements
* If there are discrepancies b/w the witnesses, the witness may change their report to make it consistent
- There maybe more than one witness and the witness overhears others’ statements
- Officer may incorporate erroneous detail from a prev witness’ interview
* Ex. Officer may ask: What was the perpetrator w/ the scar wearing?”
* The witness may later report the perpetrator had a scar when in fact there was none
- Officer may incorporate erroneous detail from a prev witness’ interview
12
Q
Procedures That Help Police Interview Eyewitnesses
- Hypnosis
- Why is it used
- How does it help
- 2 hypnosis techniques often used
- Findings on the accuracy of info retrieved from hypnosis
- Elizabeth Bain
- What happened
- SCC ruling
A
Hypnosis
- When witnesses are traumatized, they may be unable to recall a crime
- Hypnosis may help -→ relaxation, heightened focus and concentration - → assumes Hypnosis improves recall
2 hypnosis techniques often used
- Age regression
* Witness goes back in time and re-experiences the original event
- Age regression
- TV technique
* Witness imagines that he/she is watching an imaginary TV screen w/ the events being player as they were witnessed
- TV technique
- Lit reviews
- Those under hypnosis will provide more details, but those details are just as likely tb inaccurate
- Perfect et al 2008
- One aspect of hypnosis is helpful: close their eyes can increase recall of details
- When eyes are closed, visual and auditory info is recalled more
- Hypnotized ppl are more suggestible to subtle cues by the interviewer
- Police want accurate info
- Canadian courts do not permit info gained from hypnosis to be used as evidence
In the Media Hypnotically Refreshed Memory Goes to Court, or Not
- Elizabeth Bain –Psych student at UTSC disappeared
- Her BF Baltovich was under surveillance
- Officers hypnotized witnesses who saw Bain and Baltovich
- Baltovich was charged w/ 1st degree murder
- Baltovich’s appealed the case and cited hypnotically refreshed memory should be set aside
- Baltovich spend 9 yr in prison until new trial in 2004; 2007 acquitted
- In 2007, SCC ruled that testimony under hypnosis is inadmissible as those under hypnosis are more suggestible and have a difficult time distinguishing b/w accurate and fake memories
- There’s no way to determine the memories are accurate
13
Q
Procedures That Help Police Interview Eyewitnesses - Cognitive interview
- Cog interview
- 4 memory retrieval it is based on
- Geiselman et al 1985
- Memon et al 2010
- Downsides
A
The Cognitive Interview
- Cog interview: interview procedure for eyewitnesses; based on principles of memory storage and retrieval
- This process is not recommended for unwilling participants (ex. suspects)
- Cog interview is based on 4 memory retrieval to increase recall
- Reinstating the context
- Reporting everything
- Reversing order
- Changing perspective
Geiselman et al 1985
- Ppl watched at police training film of a crime
- 46 hrs later, ppl were interviewed w/ one of the procedures by law enforcement professionals
- Results: Geiselman et al. (1995) found that cognitive interview produce more accurate info than standard police interview and hypnosis
Fisher and Geiselman 1992
- Expanded the cog interview into the enhanced cog interview
- Enhanced cog interview: interview procedure that includes principles of social dynamics and memory retrieval principles used in the original cog interview
- Additional components
- Rapport building
- Supportive interviewer b: witness’s free recall is not interrupted; officer should wait out pauses, who express attention to what the witness is saying
- Transfer of control: witness (not officer) should control the flow of the interview; witness is the expert/ person who saw the crime
- Focused retrieval: qs are open ended and not suggestive; after free recall, the officer should use focused memory techniques to help retrieval
- Witness-compatible questioning: officers’ qs should match the witness’s thinking
* If the witness is talking about clothing, the officer should be asking about clothing
- Witness-compatible questioning: officers’ qs should match the witness’s thinking
Memon & Bull 1991
- Compared enhanced cog interview, original cog interview, and standard police interview
- Both cog interview produde more accurate info w/o increasing inaccurate info than standard interview
- No diff b/w the 2 cog interview
Memon et al 2010
- Meta analysis
- Cog interview produces an increase in accurate info, and small increase in errors
- Both original and enhanced version produced the results
- Dunno which components increase the accurate info
Wright & Holliday 2007
- Tested cog interview in UK w/ diff aged ppl
- Young adults 17-31
- Older adults 69-74
- Older older adults 75+
- Cog interview increased the amount of accurate “person, action, object, and surrounding” details for each age group compared to standard interview
- No increase in errors
- Downsides: Time intensive; officers are reluctant to use it (discretion)
14
Q
Recall of The Perpetrator
A
Recall of The Perpetrator
- Witness may be asked to describe the perpetrator’s appearance
- These descriptions are vague and apply to many ppl
Quantity and Accuracy of Descriptions
- Witness descriptions are limited in detail and accuracy
Webber 1994
- Examine description by adults in real and staged crimes
- Witnesses to staged crimes report 7.4 descriptors
- Witnesses to real crime report fewer 4 descriptors
- Common report – hair and clothing
Fahsing et al 2004
- Examined archive
- 250 offender descriptions made by witnesses to armed robberies at banks
- Written descriptions from case files were compared to surveillance vids
- 2348 attributes were described
- All included gender in description
- 90% of descriptions are about the upper body, clothing, and height
- Witnesses usually mentions the offender’s build, weapon, pants, bag, and age
- The descriptions provided were accurate but very gender and non-specific
- Description were accurate and reliable compated to videotapes
- Witnesses report gender, build, height, age, and ethnicity
- Gender was most accurate
- Age, height was inaccraute
Ganhag et al 2013
- Looked at witnesses’ descriptions of murder of Anna Lindh – Swedish Foreign Minister
- 42% of described attributes were incorrect
- Witnesses were unreliable when describing basic features lije age and height
- # of errors was higher due to brief time ppl saw the crime (12-14s)
Wagstaff et al 2003
- Hair color and hairstyle were reported most accurately
Other studies
- features, like weight eye color and footwear are lower in accuraxy
- IOW: descriptions are limited in quantitiy and accuracy, and not that useful
Kask et al 2006
- Examined the effectiveness of having a standard for witnesses to use when answering qs about the target
- ½ ppl were asked a “standard- qs”
- When the experimenter was used as a standard, experimenter asks: my hair is this long. How long was his hair?
- Others were asked no-standard qs – How long was his hair?
- Results: use of a standard did not help witnesses to recall accurate person info
- It did help to reveal descriptors that are not usually remembered well (ex. appearance of eyes, nose, mouth)
Sauerland and Sporer 2011
- Change in modalities of how a descriptions is elicited can influence the amount of info recalled
- Looked at whether asking witnesses for a verbal description differed from asking them to write down what they remember
- H: writing may produce less description info as it put demands on WM
- Also writing is slower and less practiced that speaking
- Results: asking witnesses to write their descriptions produce shorter and less accurate descriptions than oral
15
Q
Recognition Memory
- Define
- 3 ways recognition memory is tested
- # 1 Lineup
- 2 uses for lineup identification
- Foils/distractors
- 2 strategies
- Similarity-to-suspect strategy
- Match-to-description strategy
- 2 strategies
- Fair lineup
- How to: Estimating identification accuracy
- 3 types of identification decisions can happen w/ target-present lineup
- Correct identification
- Foil identification
- False rejection
- 3 types of identification decisions for target-absent lineup
- Correct rejection
- Foil identification
- False identification
- 3 types of identification decisions can happen w/ target-present lineup
- Identification decision implications
- Foil identification
- False rejection
- False identification
- 5 reasons to use photoarrays
- Simultaneous lineup
- Relative judgement
- Sequential lineup
- Absolute judgement
- # 2. Videos
- Lindsay & Wells (1985) RESULT
- Show-up
- 2 acceptable uses of a showup
- Walk-by
- Lineup biases
- 3 types of biases that increase false +ves
- Foil bias
- Clothing bias
- Instruction bias
- 3 types of biases that increase false +ves
A
Recognition Memory
- Recognition memory: determine whether the previously seen item or person is the one currently being viewed
- Witness’ recognition memory is tested by
- Live lineups or photo arrays
- Video surveillance records
- Voice identification
Lineup Identification
- Typical method; witness views a group of possible suspects and determine whether one is a the perpetrator
Why conduct a lineup?
- Suspect: a person the police suspect who committed crime
- Can be innocent or guilty
- Perpetrator: guilty person
- Use 1: Lineup identification reduces the uncertainty of whether a suspect is the perpetrator beyond verbal description provided
- Witness identifying the suspect increases the likelihood that the suspect is the perpetrator; vv
- Use 2: Lineup also provide police w/ info about the physical similarity b/w the lineup member chosen and the perpetrator
- Police have some notion of what the perpetrator looks like based on the person selected from the lineup
Lineup distractors
- Other lineup members are included
- Foils/ distractors: the are known to be innocent of the crime in qs
- 2 types of strategies to decide on the physical appearance of the lineup distractors
- Similarity-to-suspect strategy: match lineup members to the suspect’s appearance
* Ex. if the suspect has brown hair, blue eyes, and moustaches, then each lineup member have these characteristics
* Issue: there are many physical features that can be matched, like eyebrow width, length of nose, thickness of lips
* Can produce a lineup of clones
- Similarity-to-suspect strategy: match lineup members to the suspect’s appearance
- Match-to-description strategy: sets limits on the # of features that need to be matched
* Distractors are matched only on the items that the witness provided in their description- Ex. if a witness state the criminal had brown hair, blue eyes, round face, and no facial hair; then those would be features the lineup member is matched
- Match-to-description strategy: sets limits on the # of features that need to be matched
Lindsay et al. (1994)
- Fair lineup: lineup where the suspect does not stand out from the other lineup members
- Ex. if skin color was not mentioned, then a lineup can be made w/ one white face (the suspect) and 5 black faces biased lineup
- Some characteristics, ex. sex and race, should be matched even if not mentioned in the witness’s description
Others rs
- To avoid biased lineup, if a feature is provided in the witness’s description but does not match the suspect’s appearance, then the distractors should match the suspect’s appearance on that feature
- Ex. if the perpetrator is described as having brown hair but the suspect has blond hair, then the distractors should have blond hair
Estimating identification accuracy
- To determine how often witnesses make an accurate (or inaccurate) identification decision, we need to create the condition when police hv arrested the right suspect - guilty
- Also, need to create the condition when police hv arrested the wrong suspect – innocent
- This creates 2 lineups
- Target-present lineup: contains perpetrator
- Target-absent lineup: no perpetrator, only innocent suspect
- 3 types of identification decisions can happen w/ target-present lineup
- Correct identification: witness identify guilty suspect
- Foil identification: witness identifies a foil
- False rejection: witness state the perpetrator is now present
- 3 types of identification decisions for target-absent lineup
- Correct rejection: witness can state the perpetrator is not present
- Foil identification: witness can identify a foil
- False identification: witness identify an innocent suspect
- Rs many refer false identifications and foil identifications from a target-absent lineup as false positives
Identification decision implications
- correct decision w/ a target-present lineup = correct identification
- correct decision w/ target-absent lineup = correct rejection
- Foil identification (w/ target-present or target-absent lineup): known error to the police, so the person identified will not be prosecuted
- The witness may be perceived as having a faulty memory
- The other details provided by this witness may be viewed w/ some skepticism b/c a known recognition error was made
- False rejection: unknown error and may result in the guilty suspect going free and possibly committing further crimes
- False identification: unknown error in reality, may result in innocent suspect being prosecuted and convicted for a crime he/she did not commit
- The real criminal remains free to commit further crimes
- The most serios type of identification error a witness can make
- Foil identification (w/ target-present or target-absent lineup): known error to the police, so the person identified will not be prosecuted
Live lineups or photoarrays
- Photo arrays are more common than live person lineups
- Reasons
- They are less time-consuming to construct
- They are portable
- Suspect does not have the right to counsel being present when a witness looks at a photo array
- Photos are static
* Police need not worry that the suspect’s b may draw attention to him or herself, thus invalidating the photo array
- Photos are static
- Witness may be less anxious examining a photo array than live lineup
- Can use video-recorded lineups
- Pros:
- Can enlarge faces or focus on particular features
- Lineup members are shown walking, turning, and talking
- Pros:
- Cutler et al 2010
- Examined whether kids and teens performed differently w/ live vs videotaped lineups
- Results: only for teens the video lineup had a slight advantage when the target was not present (for correct rejections)
- No differencts b/w 2 lineups
- UK uses video lineups
- Video lineups are shown in sequence to witness
- Witness can see lineups x2
- Horry et al 2012
- Examined the identification decisions from 1039 actual lineups in UK
- Found that witnesses requesting additional passes through lineups were more likely to guess for foil identifications
Lineup presentation procedures
- Simultaneous lineup: presents all lineup members at one time to the witness
- Wells 1993
- Suggest this encourages witness to make a relative judgement
- Relative judgement: lineup members are compared w/ one another and the person who looks most like the perpetrator is identified
- Sequential lineup: present the lineup members serially to the witness
- Witness must make a decisions as to whether the lineup member is the perpetrator b4 seeing the next lineup member
- Witness cannot ask to see previously seen photos and witness is unaware of the # of photos shown
- May reduce the likelihood witness can mak ea relative judgement
- Witnesses make an absolute judgement
- Absolute judgement: each lineup member is compared w/ the witness’s memory of the perpetrator and the witness decides whether it is the perpetrator
Lindsay & Wells (1985)
- compared the identification accuracy rate w/ simultaneous and sequential lineup procedures
- uni students witness videotaped theft and were asked to identify the perpetrator from 6 photos
- ½ the students saw target-present lineup
- ½ students saw target-absent lineup
- Across target-present and target-absent conditions, the lineups were either presented w/ simultaneous or sequential procedure
- Correct identification (target-present lineups) rates: did not differ across lineup procedures
- Correct rejection rates sig differ across lineup procedures
- 42% of the participants made a correct rejection w/ simultaneous lineup
- 65% of the participants made a correct rejection w/ a sequential lineup
- IOW: if perpetrator was not included in the lineup, witnesses were more likely to correctly indicate that he/she was not present if shown in SEQUENTIAL lineup
- high correct rejection rate using the sequential than simultaneous procedure is replicated
- 2 meta-anlysis
- Correct identification rates are lower for sequential lineups than sequential lineups
- This difference lessens as real world conditions are considered
- Sequential lineup: used in some Canadian jurisdictions (ex. ON) and US states (NH)
- Other rs
- If certain method factors are used, the simultaneous procedure produces a correct rejection rate similar w/ the sequential procedure w/o a drop in correct identification
- Show-up: identification procedure that shows one person to the witness – the suspect
- Witness is aksed whether the person is the perpetrator
- Cons
- Since there is no other lineup members shown, the witness is aware of whom the police suspect
- This increase the chance of making a false identification
- Gonzalez et al 1993
- Did not find false identifications is higher w/ a showup than a lineup
- The witnesses were more likely to reject a showup than all members of a lineup
- Witnesses are more cautious w/ their decision making when presented w/ a showup rather than a lineup
- They will make a rejection rather than an identification
- Yarmey et al 1996
- Lineups produced lower false-identifications rates than showups
- Steblay et al 2003
- False identifications were higher w/ showups than lineups
- Analysed 271 cases: suspect was more likely to be identified in a field showup than in photographic lineup
- IOW: showups are suggestive
- Wetmore et al 2015
- Have 1600 ppl view crime video and perform a showup or simultaneous lineup identification immediately or 2 days later
- Showup never resulted in higher accuracy compared to lineup, even when lineups were biased instead of fair
- Pont: need more rs
- Only 2 acceptable uses of a showup
- Deathbed identifications: fear that the witness will not be alive by the time a lineup is assembled
- Used if a suspect is apprehended immediately at or near the crime scene
- Walk-by: identification that occurs in naturalistic env; police take witness to a public location where the suspect is likely tb
- Once the suspect is in view, the witness is asked whether he/she sees the perpetrator
- In lineups w/ 1+ ppl, person administering the lineup need to decide where to place the suspect’s photo
- Usually done randomly
- Megrey et al 2012
- Examine eye movements of witnesses examining lineup
- Faces on the L were looked at first
- Foil on the LS were more likely tb inaccurately identified as the target
Lineup biases
- Biased lineup: a lineup that “suggests” whom the police suspect and thereby whom the witness should identify
- Biases that increase false +ves
- Foil bias: suspect is the only lineup member who matches the description of the perpetrator
* Ex. suspect has a beard and moustache while other lineup members shaved
- Foil bias: suspect is the only lineup member who matches the description of the perpetrator
- Clothing bias: Suspect is the only one wearing clothing similar to that worn by the perpetrator
* Ex. perpetrator was described wearing a blue baseball cap
* Suspect is wearing the cap while foils are not
- Clothing bias: Suspect is the only one wearing clothing similar to that worn by the perpetrator
- Instruction bias: Police fail to mention to the witness that the perpetrator may not be present
* Police imply that the perpetrator is present and that the witness should pick hem/her out
- Instruction bias: Police fail to mention to the witness that the perpetrator may not be present
Box 5.3 Canadian Researcher Profile: Dr. Rod Lindsay
- Study eyewitness issues and real world applications
- Frustrating cases
- Case 1: First nations suspect as placed in 12 person lineup in which no other member was First nations
- Case 2: Filipino suspect is placed in 12 person lineup, all other lineup members were First nations
- Court decided nothing was wrong w/ the lineup
- Lindsay disagree
- Canada should hv smth similar to the UK Hone office
- it research police procedures and decide on the best practices and distribute this info through law enforcement
- Throws our evidence that did not meet standards
