Making A Case
Geiselman & Fisher
Mann et al
Case Study of John Duffy
Being able to accurately identify a face is key to many convictions. However, cases
of mistaken identity show that this is not always as ea
sy as it seems.
Bruce et al
shown that recognising a familiar face is something which we are all quite good at,
are a completely different matter.
Look at the images of the unfamiliar faces below; are they of the same man? (We
l revisit this later)
Sinha et al ( 2006) after
completing a meta
analysis of face
recognition research, found 8
factors used to decide how to
you believe to be important in
face recognition and leave out t
We can recognise familiar faces even when the resolution is low (poor
Familiarity does not impact recognition
The more familiar they are the easier it is to recognise faces.
Facial features are processed holistically (all to
gether as a single unit).
Eyebrows and hairline are most important features.
Illumination changes influence recognition (the suspect may look very
different in poor light or side light).
to be one system to process identity and expression.
Motion of the face helps recognition (We can better recognise a video than
Illumination changes do not have any real impact on recognition, it is the
features that matter.
There appear to be specialised neuron
s for face recognition.
and expression may be processed by two different systems in
High resolution (good quality CCTV) is needed to recognise faces.
Eyes and nose are key factors in helping process a face.
Piece by piece facial reconstructions are important.
recognition has the same problems as other types of memory recall:
Length between event when you saw the face and recall of it?
Content and cue
dependent memory of the face or snapshots
of intense emotion associated with the face.
How much meaning to the individual has the face to be
How much individual variation is there in the ability to
Internal and External Features in facial recognition
Familiar faces enjo
y high recognition
even under difficult circumstances,
whereas unfamiliar ones are often
misidentified even in good conditions.
Familiar faces are recognised
using their internal features such as
eyes, brows, nose, mouth
whereas unfamiliar ones are recognised by external features such a
s head shape,
hair and ears...
Witnesses tend to see suspects for a very short time and therefore use unfamiliar
face perception wh
en creating a composite. This means that external features will
be described best. However, witnesses are often trying to describe unfamiliar faces
using internal features to draw up a photofit, so this will be unreliable.
A photofit picture of a murder suspect in 1992 from
descriptions from m
f the public (Unfamiliar face
Bruce et al (1988) The impor
tance of external and internal features in facial recognition
To investigate the relative recognisability of internal (eyes, brows, nose and
mouth) and external (head shape, hair and ears) features of a face composite.
Methodology & P’s
. Exp 1: 15 male and 15 female staff & students (av age
29.2) from Stirling Uni
paid £2 to sort composites. Exp 2: 21 males & 27 female
undergraduates from Stirling Uni
This was an
independent measures design
with three conditions. All the
p’s were given pictures of 10 celebrities, the task was to match the correct composite
image to the celebrity from the 40 composites given. Group 1 were given complete
composites, group 2 composites containing only interna
l features (eyes, nose, mouth)
and group 3 external feature (head shape, hair and ears) composites. Each face was clean
shaven and spectacles were avoided.
used a photo array; the task was to identify the celebrity
es from the array. The task was made either easy (very different to target face) or
hard (very similar to target face). Once again the composites were composed of either
external or internal features.
42% of whole composites an
d those of external features were sorted
correctly, compared to only 19.5% of internal features.
External features were identified 42% of the time, compared to just 24%
of internal features.
In exp 1 & 2
P’s performed at jus
t above chance level for internal features.
All P’s performed equally well with both whole composites and external features.
This shows that external features are more important for facial recognition and
that faces are processed holistically. This has
implications for facial reconstructions
which involve witnesses picking internal features from a book. Newer face
reconstruction software takes these finding into account. E
fit or EvoFIT both
change faces holistically.
Evaluate the study
Method: Lab st
Reductionism or holism?
Is the research ethnocentric?
Is the study ecologically valid?
been applied to real life…Evofit
new facial co
for victims and witnesses to construct a likeness of a
criminal's face. Traditional composite systems used by the police require witnesses to describe an
assailant's face and then to select individual facial features (producing a "composite"
process does not work well: we are not good at describing faces nor selecting individual features.
Unlike current systems, faces in EvoFIT are modelled in their entirety and are not separated into
component parts. A facial composite (the new t
erm for 'photofit') is created by first displaying a
number of faces containing random eyes, noses, mouths, etc. A witness selects a few of these faces
that are most similar to a criminal. The selected faces are then mixed or 'bred' together to produce
ther set for selection. Repeated a few times allows a composite to be 'evolved'. The most recent
version has been shown to produce composites that can be named on average about 6 times better
than composites from a traditional 'feature' system.
h is theoretically better, with witnesses selecting whole faces to allow a composite to be
'evolved' over time; identification is better too. It was developed by
Professor Peter Hancock
University of Stirling,
Professor Vicki Bruce
at Newcastle University and
University of Central Lancashire. The most recent version has been shown to produce composites
that can be named on average about 6 times better than composites from a traditional 'feature'
EvoFIT was first used in a criminal inve
stigation with Northants police as part of
The case involves a series of sexual offences carried out by a Caucasian male believed to be in his
late twenties in Southern England (all of which have been linked by DNA evidence). The
ction of an EvoFIT was carried out in the normal way by the selection of shapes and
textures using an updated hairstyle. Three generations (phases) of faces were required to produce
"The Beast of Bozeat". Below are the best intermediate images and the EvoF
IT for each phase. The
final image is a good likeness to the assailant.
A photograph of 32
old James Ben Davies, from Camberley, Surrey who was
sentenced to four and a half years in prison at Northampton Crown Court in
2005, in relation to thr
ee indecent assaults on lone females in Hampshire,
Cambridgeshire and Northamptonshire between 1998 and 2000
The photofit is of the person who murdered Rachel Nickel in 1992 by stabbing her
numerous times in front of her 2 year old son. The images at the
top of the page are of
Colin Stagg (on the left) and Robert Napper (on the right). Police arrested Stagg for the
crime that had actually been committed by Napper. A string of police errors led to the
wrongful arrest. The two men do look alike, but the phot
ofit is less convincing…perhaps
Evofit technology would have helped catch the correct suspect sooner and prevented her
murder and the murder of another mother & child and numerous rapes of other women
Extension work: Various websites will pro
vide info on this case to follow
up. To create your own face composite and have some fun visit
Factors Influencing Identification
refers to the concentration of a wit
ness’s attention on a
weapon which results in them having difficulty recalling other details of
the scene and identifying the perpetrator of the crime.
Loftus et al (1987) Weapon Focus
To provide support for the Weapons focus effect when witnessin
g a crime.
P’s told this was a study of
(when something you have
learned earlier interferes with your memory of the present) All p’s were shown 18 slides of
a series of events in a Taco restaurant
. For both groups the slide was the same except for
one slide. This slide was the independent variable. In the control group the slide shows the
second person in the queue handing the cashier a cheque, in the experimental group the
slide shows the same p
erson pulling a gun on the cashier (IV). The dependent variable was
measured by a twenty item multiple choice questionnaire.
What was the DV?
P’s were also show 12 photos in random sequence and asked to rate how confident they
were of their identificatio
n. Gaze concentration was measured by reviewing where eye gaze
was concentrated and then plotting these measurements
Answers to the questionnaire about the slide show showed no significant
difference between the two groups. In the control cond
ition (cheque) 38.9% made a
correct identification, in the experimental condition (gun) it was only 11.1%.
indicated a significant difference. Eye fixation data showed an average fixation time of
3.72 secs on the gun, compared to 2.44 secs on
the result indicated a
A second experiment with the same procedure using 80 psychology students found
Questions to answer
Why were P’s told this was a study of proactive interference
The IV was one slide in a series of 18, why did she keep all other factors constant?
What are the limitations of using a lab study to investigate this?
What does it mean, when results are found to be ‘significant’?
Discuss the sample in terms of
how representative it is
The experiment was highly controlled, provide examples of this and explain why
this is important
What is positive about doing a second experiment that provides the same results?
Has the results of this study had a real life i
Example exam question
Outline any relevant research which can inform us about how a witness should be
Evaluate the methodology used to investigate the interviewing of witnesses ( 15)
Hints on how to answer such question
Expected content be drawn from the specification is, recognising and recreating
faces by E
fit (e.g. Bruce 1988). Factors influencing accurate identification (e.g
The cognitive interview (e.g Fisher et al 1989)
We have not yet covered
p band marks require
Correct and comprehensive use of psychological
terminology. Description of evidence is accurate, relevant, coherent and detailed.
Elaboration/use of example/quality of description is very good and the ability to
evidence selected in the context of the question is very good.
The answer is mostly grammatically correct.
You should be able to evaluate the scientific methodology used in research in
terms of its design, sample, validity, reliability, ethics and usefu
lness. You may also
choose determinism, reductionism and freewill to answer the question.
Top band marks require
Evaluative points covering a range of issues. The
argument is competently organised, balanced and well developed. The answer is
ated to the context of the question. Effective use of examples. Valid
conclusions that effectively summarise issues and arguments is highly skilled and
shows thorough understanding.
The Cognitive Interview (CI)
Forensic psychologists combined vari
ous ideas and designed a
more effective way of questioning witnesses that avoid any
chance of leading the victim.( which led from the work by
Loftus). These have been shown to produce more reliable recall
Fisher and Geiselman
(1992) designed th
The technique is based around four main cognitive components: Match up the technique
with what it means, then how it might work. Insert the correct letter in the table below
that corresponds to the statement.
a) It encourage
s witnesses to recall events in different orders, for example starting half way through a sequence
of events and then working backwards
b) It tries to recreate the scene of the incident in the mind of the witness, this includes the sights, sounds and
s but also crucially it attempts to model the emotions and feelings of the person at the time.
based on the concept of cue dependent memory.
c) Evidence suggests that we are more likely to recall information if it is in a similar context to when i
t was first
experienced or learned, so putting ourselves in a similar state of mind should aid recall.
d) It encourages witnesses to view the scene as others present may have seen it, for example as other witnesses,
the victim or the perpetrator may have s
een the incident.
e) If one route fails then try another.
So if working through from start to finish hasn’t worked try to accessing
the memory by sneaking up on it from a different angle e.g. backwards.
f) These two techniques are designed to get the witn
ess to mentally revisit the scene and mentally reconstruct
the incident in their mind.
g) It encourages witnesses to report all detail that they can remember regardless of how trivial it may appear,
h) These two techniques are based on the idea that onc
e a memory has been stored there is more than one
way of getting at it or retrieving it.
Stages of the interview
What this means
How does this work
Recall in reverse order
Evidence FOR the cognitive interview
Geiselman & Fisher
(1985) got participants to watch a video of a violent crime.
days later they were interviewed in one of 3 ways: standard police interview, cognitive
or under hypnosis.
The cognitive interview was found to trigger the most
Note: hypnosis is not as effective as films would lead us to believe (the so
Witnesses often do recall more under hypnosis and are more
confident in their recall.
Unfortunately much of what they recall is inaccurate.
Additionally, their confidence in what they recall can be very influential in court room
situations, particularly with jurors so is doubly dangerous.
Kohnken et al
carried out a
of 53 other studies and found that the CI
could elicit an average of 34% more detail than the standard interview and crucially
without the loss of accuracy you get with hypnosis.
Interestingly, when the four components of the
interview are used individually, e.g. recall
in a different order, there is little gain over the standard interview.
It’s only when two or
more components are used that there is significant improvement in recall.
Milne and Bull
(2002) The ‘report everyth
ing’ and ‘context reinstatement’ combinations appear most
Can valid comparisons be made between police in different countries or different
forces in the same country?
Any problems with the technique in practice?
Does this techni
que work with young children?
Fisher et al (1989)
Field test of the cognitive interview
To test the cognitive interview in the field
Field experiment using real police interviews of witnesses
16 detectives from the Rob
bery division of Dade County, Florida, police. 7 of
the detectives were trained in the cognitive interview technique (CI). All had minimum of
5 years experience.
Detectives asked to record their interviews using Standard interview
(SI). Detectives then divided into two groups, with one of these groups (7
officers) being trained in Cognitive interview technique (CI). The two groups continued
to record their interviews.
The interviews were recorded and analysed by a team at the Univer
sity of California, who
were blind to the conditions (i.e. they didn’t know if the interview they were analysing was
by a trained CI detective or not). The amount of information gathered by the two groups
of detectives was collated.
63% more in
formation was obtained by the detectives trained in CI compared
to those untrained
Cognitive Interview Techniques do seem to work, more information is
gathered. This could help the Police solve more crimes.
How does ca
rrying out a field experiment affect the validity of examining
the Cognitive interview technique?
Reductionism & Holism:
Can witness behaviour really be reduced to one variable
such as interview style or is reductionism of limited use when applied to re
Individual V situational explanations:
Can the witness experience and subsequent
interview they are in explain behaviour or their individual personality? Can the two
was it representative?
Have these findin
gs been consistent with other work in this area?
Have the research findings been applied to everyday policing ad been
You do not have to say anything, but it may harm your defence if you do
mention when questioned something which you later rely on in Court. Anything
you do say may be given in evidence”.
After arrest the purpose of the interview is to gather further evidence to establish
guilt or innocence. In the Bri
tish legal system a defendant is presumed innocent,
until proven guilty. Police officers like to think they are good at detecting lies and
are trained to look for body language clues such as looking down, putting their
hands over their mouths, inconsisten
t responses etc.
But are police officers good at detecting lies?
In the USA polygraph testing is widely used, but is this any more reliable?
Some people are just good liars!!
Mann et al (2004) Police officers’ ability to detect suspects’ lies.
Mann et al (2004) Police officers’ ability to detect suspects’ lies.
To test police officers’ ability to distinguish truth and lies during interviews
This was a field experiment involving 99 Kent police officers (24
emale and 75 males mean age 34.3). These participants were shown 54 video clips
involving 14 suspects of real
life police interviews; each clip was backed up by
further evidence which established truth or lie. The clips were of head and torso
in length from 6 to 145 seconds. The participants were asked to fill out
questionnaires about their experiences of detecting liars and decide if each clip was
truthful or a lie. They were also asked to make a note of cues they used to detect
Police Officers performed better than chance (50%) they were 66.2%
accurate on lies and 63.6% accurate on truths. The most frequently mentioned
cues were gaze, movements, vagueness, contradictions in stories and fidgeting.
ice officers were good at detecting lies. However, to
establish whether they were any better than the general public would require a
control group. This study could not have had a control group because of issues of
confidentiality and privacy (the clips
were real police interviews).
Use the terms below to create evaluative questions with answers
these will be
swapped around to question each other.
Methodology, demand characteristics, sample,
ethnocentrism, reductionism v
holism, individual V situational explanation, reliability, usefulness
Create descriptive questions about the details of the study
these will be
swapped around to test each other.
What is the difference between an interrogation and an interview?
The investigator starts by telling the suspect that there is no doubt as their guilt. It
usually takes place after an interview. A
n interrogation is designed to increase
arousal and anxiety. The Police and Criminal Evidence Act (PACE) (1984) requires
all interviews to be recorded in triplicate, one for the police, one for the solicitor
and one for the court, to ensure evidence cannot
be tampered with.
Inbau believes that it is justifiable for the police to lie, deceive or use trick to force
this may be acceptable in the USA but is not allowed in the UK
(however false confessions have still been known to occur).
Chance to shift blame (either to some else or circumstances)
Never allow the suspect to deny guilt
Reinforce sincerity, eye contact, first names
If suspect cries, infer guilt
Pose the ‘alternative question’
give two choi
ces, one more socially
acceptable, but both inferring guilt
Admit guilt in front of witness
Real life interrogation technique
Read the case of how these interrogation techniques work in real life
this is a transcript
an actual interrogation of a non
violent activist in the USA who threw custard pies at
the Mayor of San Francisco in protest to his overspending on re
gilding the dome at City
Hall, while homeless people died on the streets.
Inbau et al (1986) The Reid ‘nine st
eps’ of interrogation
Can these techniques be justified?
What type of offender might be particularly vulnerable?
Could they lead to false conf
2.3 False Confessions
suggest that there are three types o
f false confession:
The ‘voluntary confession’
no obvious external pressure
The ’coerced compliant confession’
after forceful or persistent
questioning the suspect confesses to escape the interview
The ‘coerced internalised confession’ where the suspect becomes
temporarily persuaded that they are guilty.
Persuasion is possible because are memories are fragile and susceptible, especially
when we are in an anxious and vulnerable state. So
me groups are more susceptible
to this pressure than others:
the young, the mentally ill, those with a low IQ etc.
Gudjohnsson et al (1990)
A case of false confession
to document a case of false confession of a youth who was at the time
sed and susceptible to interrogative pressure.
This was a case study of FC a 17 year old youth who was accused of
two murders. In 1987 two elderly women were found battered to death in their
home, their saving’s were missing and they had bee
n sexually assaulted. FC was
arrested because of inconsistencies in accounts of his movements during an earlier
routine enquiry and because he was spending more money than usual. There was
no forensic evidence to link him to the case.
During the inter
view he was repeatedly accused of lying, the questioning
was leading and accusatory and the police suggested he was sexually impotent. He
found this very distressing and after 14 hours of aggressive questioning without a
break FC confessed. He retracted
his confession the next day, only to confess
again under pressure about his failure to have successful relationships with women.
In prison he was assessed by psychiatrists who found no evidence of mental illness,
an IQ of 94 and a score of 10 (very high)
on the Gudjohnsson Suggestibility Scale.
After a year in jail he was released when another person pleaded guilty to the
This is a clear case of ‘coerced compliant false confession’. He gave
in to pressure to escape from the intervie
What step/s in Inbau’s interrogation scale does the questioning fall into?
What situational factors would induce a false confession?
What individual factors may induce a false confession?
What is the problem with looking at case
Below are examples of
life false confessions
. You will research these further in
groups and feedback your findings to the rest of the class.
Netta Bell murder
Corethian Bell in Cook County
Bakewell murder of legal se
Stephen Downing 1974
High School rape and murder
Jeffrey Mark Deskovic 1990
Stephanie Crowe 1998 murder
Gauger parents murder 1993
Central Park jogger attack and rape 1989
5 teens confessed
Creating a Profile
The US approach (top
down approach) is based on the work of the FBI in response
Explain the term
his contrasts with a bottom
up approach which
details and creates the big picture. Both of these approaches
have been used to build up profiles of crimes to aid police in
solving crimes and apprehending criminals. Both have flaws, in
reality it will depend on the situation and type of cr
ime as to
which approach is used.
i楫e汹l䍨a牡r瑥物獴楣s of l晦e湤敲
pho睳 獥s映co湴牯l a琠瑨e c物re 獣e湥
s楣i業 楳ia 瑡牧t瑥搠獴牡湧er
A瑴敭灴p 瑯 co湴牯l
A扯ve ave牡re 䥑
poc楡汬y a湤n獥s畡汬y com灥瑥tt
瑩te o映瑨e o晦e湣e
co汬o睳 me摩d cove牡re o映捲業e
i楴瑬e a瑴敭灴p瑯 h楤i 瑨
e ev楤i湣e a琠
j楮im畭 畳e o映co湳瑲a楮t
i楶e猠a汯湥Ⱐ湥l爠瑯 c物re 獣e湥
pex畡汬y a湤n獯c楡汬y 楮a摥d畡瑥
published their account of the ‘lust murderer’,
they advanced a theory that lust murderers are mainly catergorised by two types:
Organised and disorganised. This is an example of a top
e main source of FBI data is interviews with offenders in prison. The obvious
characteristics of ‘serial killers’ are that they tended to be white, unmarried, male,
in unskilled occupations, have a history of psychiatric problems and come from
l family backgrounds.
FBI investigators in 1979 interviewed an opportunity sample of 36 sexual
murderers in prison. The interviews were unstructured and developed in an ad hoc
fashion. They were able to categorise them as either ‘organised’ or ‘disorganis
from their responses to the interview q
24 organised and 12
What are the issues with building offender profiles? Do they help or
hinder the police in their search for perpetrators of crime?
leads an ordered life and kills after some sort of
critical life event. Their actions are premeditated and planned, they are
likely to bring weapons and restraints to the scene. They are likely to be of
average to high intelligence and employed.
is more likely to have committed the crime in a
moment of passion. There will be no evidence of premeditation and they
are more likely to leave evidence such as blood, semen, murder weapon etc.
behind. This type of offender is though
t to be less socially competent and
more likely to be unemployed.
Canter et al (2004) Investigation of organised/disorganised theory of serial murder.
To test the reliability of organised/disorganised typologies.
A content analysis o
f 100 cases of serial killers from the USA. The
third crime committed by each serial killer was analysed using
(Douglas et al 1992).
Twice as many disorganised as organised crime
scenes were identified. In
of cases the body was concealed and in 75% of cases sexual activity occurred.
Further analysis failed to reveal any significant differences between organised and
Canter concluded that instead of their being a distinc
two types of serial murder, all of the crimes had to have an organised element to
them as they hadn’t been caught after three killings. It would be better to look at
Canter (1994) criticises the technique
used in the American approach. He
suggests that interviews with criminals are unreliable
Why is this?
Other criticisms of the FBI approach are:
The samples of serious offenders interviewed in order to draw up the
classification system were a small
mple of serial rapists and murderers
Why was this, a problem?
The crime classifications for murderers and rapists were crude and not very
helpful in terms of detection
Details of the development and efficiency of the methods were not
ach flawed because it lacks scientific rigour and is more akin to
Bottom up approaches
is the UK’s foremost profiling expert, his bottom
approach looks for consistencies in offenders’ behaviour during the
e. No initial assumptions are made about the offender and the
approach relies heavily on computer databases. It can be the little
details that are often overlooked that can be crucial to the success of a
most famous case is that of the ‘Ra
ilway Rapist’ John Duffy.
Explain the term Bottom
Why is this type of approach thought to be more objective and reliable than the top
ur of the offender at the
time of the crime will be comparable to
what they’re like in everyday life.
Degrees of violence used in serious
crimes, especially rape, may reflect how
the criminal treats other women in his
eir crimes close to where
they live or feel secure.
Usually disorganised criminals
Commit their crimes away from where
they usually live.
Canter & Heritage (1990) Developments in Offender Profiling.
To identify a beh
aviour pattern from similarities between offences.
A content analysis was carried out on 66 sexual offences from
various police forces committed by 27 offenders. The data was subjected to
Canter’s smallest space analysis (a computer progra
mme devised by Canter)
Five variables were found to be central to the 66 cases:
No reaction to the victim
Victim’s clothing disturbed.
This suggests a pattern of behaviour where the atta
ck is impersonal and sudden and
the victim’s response is irrelevant to the offender.
This has become known as the five factor theory. These five aspects
have now been shown to contribute to all sexual offences, but in different patterns
or different individuals. An analysis of these factors can enable Police to decide if
an offence has been committed by the same individual.
Extension Task: The case of Madeleine McCann
Madeleine was reported missing in May 2007
and has yet to be found. David Canter joined other
leading professionals in the criminal and forensic field in October 2007 to revisit the resort that she
disappeared from to try to identify what could have happened. David Canter and the others use the
tish bottom up approach to piecing together the possibilities that could explain her abduction.
TASK: View complete documentary (5 parts) on YouTube under “Searching for Madeleine
McCann.” Identify the aspects of the bottom
cases that you can report back on where the police have used the bottom up
approach to identifying offenders or potential likely offenders in real life cases.
This data is correlational, why does this limit our understan
ding about the causes
of the offender’s behaviours?
What is a content analysis?
This is a cognitive social approach
The effectiveness of this techniques depends on good record keeping by the
police. What factors can affect the accuracy
of police records?
Be瑷ee渠N㤷㔠a湤nN㤸V ㈳ wome渠睥we
扥瑷敥渠N㔠a湤nP㈠a琠牡r汷ly 獴a瑩t湳n楮 a湤na牯畮搠
io湤n渮 䍡C瑥爠扥came 楮瑥牥獴敤t楮 瑨e ca獥sa晴e爠
牥r摩湧 牥灯牴猠楮 the 䕶e湩湧 pta湤n牤⸠ 䥮f瑨e ea牬r
N㤸V猠瑷o po汩ce of晩fer猠睥we a灰p楮瑥搠to he汰 h業
睡猠a汲la摹 歮ow 瑨a琠
a渠accom灬楣攠楮vo汶e搠楮 瑨e c物re献s 䍡C瑥爠灬ace搠a汬
瑨e ca獥猠o渠a ma瀠a湤nth楳ia汬o睥搠h業 to 獰scu污瑥t
a扯畴uwhe牥r瑨e 牡灩獴 m楧h琠汩ve⸠ 䡥 ca瑥to物獥搠
perpetrators as ‘marauders’ or ‘commuters’. Depending on whether they str
from within their home base ‘marauders’ or travel away from home ‘commuters’.
Has lived in the areas circumscribed by the first three cases since 1983
Probably lived in that area at the time of the arrest
Probably lives w
ith a wife or girlfriend, possibly with no children.
Mid to late twenties
About 5 feet 9 inches
skilled or skilled involving weekend work or casual labour
from July 1984 onwards
His job most likel
y does not bring him into contact with the public
Keeps himself to himself, but has one or two close men friends and
probably very little contact with women, especially in the work situation.
Has knowledge of the railway system, along which the a
The variety of his sexual actions suggests considerable sexual experience.
Was probably arrested some time between 24 October 1982 and January
1984 and this may have had nothing to do with rape, but having be
aggressive under the influence of drugs or alcohol.
This profile represents the first attempt to use behavioural characteristics to search
for a criminal instead of purely forensic evidence from the crime scene.
identified 17 profile points…late
r it was found that the perpetrator matched 13/17
of the profile points.
In November 2000 John Duffy who was serving life for the rape
and murder of several women, confessed that he was responsible
for many more and that he committed some with an accomp
David Mulcahy. He was the Railway rapist.
John Duffy had been interviewed by the police for an unrelated
offence, a ‘domestic’ against his ex
wife. He was one of 2000 suspects connected
to the crime by their blood groups. He lived in Kilburn North
worked on the railways as a carpenter and was the right age. The profile did fit on
However, Duffy was a lot shorter than victims remembered and many had
described him as having black or even ginger hair. But police, sugges
t that the
‘weapons effect’ may have played apart as he threatened victims with a knife.
Profiles can be useful, but police must be careful not to be blinded to
other possibilities by them. Occasionally criminals do not fit the profile. Ov
could lead to miscarriages of justice. E.g. Paul Britton’s misleading profile in the
hunt for the killer of Rachel Nickell.
Rachel Nickell case
Offender profiling captures the imagination. But with the Rachel Nickell case, its use has
‘Rachel Nickell was a young woman who was brutally murdered in mid
walking on Wimbledon Common in south London. As part of the investigation into the
killing, a profile was commissioned from a psychologist. A suspect was eventu
was noted that he seemed to fit the profile well. An elaborate opera
drawing partly but not only on the profile, was put together in which a police woman
befriended the socially isolated and inadequate suspect, offering the pro
mise of an
tionship in exchange for descriptions of his sexual fantasies and a confession
that he murdered the woman on the Common. The confession was not forthcoming, but
he was still arrested. The case fell apart.’
Discussion on effect
iveness of profiling:
Pinizzotto and Finkel (1990)
Kocsis et al (2002)
Canter and Heritage (1990)
Canter and Alison (1999)
Check out more on profiling at the following website
1a) Outline any relevant research which can inform us about how a witness should
be interviewed (10)
1b) Evaluate the methodology used to investigate the interviewing of witnesses
2a) Describe one approach to offender profiling. (10)
2b) Assess the
effectiveness of offender profiling. (15)
research which has investigated the interviewing of suspects. (10)
b) Evaluate the effectiveness of police interviews. (15)
4a) Describe one piece of research into how lies can be detected when suspe
b) Discuss the reliability of information gained from suspects in interviews. (15)