Handbook of Psychology of Investigative Interviewing: Current Developments and Future Directions
Page 43
Steblay , N. M. ( 1997 ). Social infl uence in eyewitness recall: A meta - analytic review of
line - up instruction effects . Law & Human Behavior , 21 , 283 – 297 .
220
Handbook of Psychology of Investigative Interviewing
Stretch , V. & Wixted , J. T. ( 1998 ). On the difference between strength - based and
frequency - based mirror effects in recognition memory . Journal of Experimental
Psychology: Learning, Memory and Cognition , 24 , 1379 – 1396 .
Technical Working Group for Eyewitness Evidence
(
1999 ).
Eyewitness evidence: A
guide for law enforcement . Washington, DC : US Department of Justice, Offi ce of
Justice Programs .
Wells , G. L. ( 1988 ). Eyewitness identifi cation: A system handbook . Toronto : Carswell
Legal Publications .
Wells , G. L. & Bradfi eld , A. L. ( 1998 ). ‘ Good, you identifi ed the suspect ’ : Feedback
to eyewitnesses distorts their reports of the witnessing experience
.
Journal of
Applied Psychology , 83 , 360 – 376 .
Wells , G. L. & Bradfi eld , A. L. ( 1999 ). Distortions in eyewitnesses ’ recollections: Can
the postidentifi cation - feedback effect be moderated?
Psychological Science ,
10 ,
138 – 144 .
Wells , G. L. & Lindsay , R. C. L. ( 1980 ). On estimating the diagnosticity of eyewitness
nonidentifi cations . Psychological Bulletin , 88 , 776 – 784 .
Wells , G. L. & Luus , E. ( 1990 ). Police line - ups as experiments: Social methodology
as a framework for properly - conducted line - ups . Personality and Social Psychology
Bulletin , 16 , 106 – 117 .
Wells , G. L. & Olson , E. A. ( 2002 ). Eyewitness identifi cation: Information gain from
incriminating and exonerating behaviors
.
Journal of Experimental Psychology:
Applied , 8 , 155 – 167 .
Wells , G. L. , Olson , E. A. & Charman , S. D. ( 2003 ). Distorted retrospective eyewit-
ness reports as functions of feedback and delay . Journal of Experimental Psychology:
Applied , 9 , 42 – 52 .
Wells , G. L. , Memon , A. & Penrod , S. D. ( 2006 ). Eyewitness evidence: Improving
its probative value . Psychological Science in the Public Interest , 7 , 45 – 75 .
Chapter Thirteen
Recent Developments in Eyewitness
Identifi cation Procedures in the
United Kingdom
Tim Valentine
Goldsmiths, University of London
Carwyn Hughes
Sussex Police
and
Rod Munro
Devon and Cornwall Police
Introduction
In the initial stages of an investigation, the potential of eyewitness identifi ca-
tion may be assessed from an interview with a witness. An initial interview will
often involve taking a description of the perpetrator. In the absence of physical
forensic evidence such as a DNA profi le, investigators may rely on an eyewit-
ness to provide the necessary identifi cation. The interaction between the
eyewitness and the police is critical to the reliability of identifi cation
evidence.
The frailty of eyewitness identifi cation has long been recognized. In
response to public concern about the reliability of eyewitness identifi cation
following several wrongful convictions in the early 1970s, the British govern-
ment established an enquiry into eyewitness identifi cation chaired by Lord
Handbook of Psychology of Investigative Interviewing: Current Developments and Future Directions
Edited by Ray Bull, Tim Valentine and Tom Williamson
© 2009 John Wiley & Sons, Ltd.
222
Handbook of Psychology of Investigative Interviewing
Devlin. More recently, over 230 people wrongfully convicted in the USA have
been exonerated by DNA evidence that was not available at their original trial.
Investigation of the causes of these wrongful convictions has shown that
mistaken identifi cation was a cause in three - quarters of the cases (Innocence
Project, n.d. ).
Devlin r evisited
Devlin (1976) reported an analysis of the outcome of all cases from England
and Wales which involved disputed identifi cation in 1973. From an analysis of
2,116 live line - ups (known as identity parades in the UK), Devlin noted that
the police suspect was identifi ed in 45% of line - ups; no identifi cation was made
for 46% of line - ups; and a volunteer was identifi ed in 9% of line - ups. Thus a
known mistaken identifi cation (of a volunteer) was made in almost one in ten
line - ups.
Eyewitness identifi cation is a very powerful form of evidence. Devlin (1976)
reported that 90% of suspects who were identifi ed in a line - up were prosecuted,
and 82% of these prosecutions resulted in a conviction. In contrast, only 9%
of those not identifi ed were prosecuted on the basis of other evidence, and
86% of these cases resulted in conviction. Devlin reported separately cases
involving disputed identifi cation in which the only evidence consisted of iden-
tifi cation by one or more eyewitnesses. Seventy - four per cent of such cases
resulted in a conviction.
Devlin (1976) made the following recommendation:
We do however wish to ensure that in ordinary cases prosecutions are not
brought on eyewitness evidence only and that, if brought, they will fail. We think
that they ought to fail, since in our opinion it is only in exceptional cases that
identifi cation evidence is by itself suffi ciently reliable to exclude a reasonable
doubt about guilt. (para. 8.4 , p. 150 )
It is instructive to compare this recommendation with a Court of
Appeal judgment made almost 30 years later (
R v Mitchell ,
2005 ). The
judgment concerned the issue of whether corroboration is required to
convict on the basis of the opinion of a single expert, who concluded that his
analysis of CCTV imagery provided very strong support for the contention
that the appellant was the person in the CCTV image. Lord Justice Tuckey
stated:
as a matter of principle it is open to the jury to convict on the evidence of an
expert qualifi ed in facial mapping … There is no need for some independent
support for such evidence … Were it otherwise it would not be possible to
convict a defendant on the identifi cation evidence of a single witness and we all
know that this often happens. (para. 11)
Recent Developments in Eyewitness Identifi cation Procedures
223
Notwithstanding Devlin
’ s recommendation, in 2005 convictions on the
evidence of a single eyewitness occur ‘ often ’ .
Legal s afeguards
A direct outcome of the Devlin inquiry was a landmark ruling in the Appeal
Court in London ( R v Turnbull , 1976 ) which established the principles a trial
judge must use to caution the jury of the special need for caution in cases of
disputed eyewitness identifi cation.
The Turnbull g uidelines
English case law is based on the premise that a distinction can be drawn
between good and poor eyewitness identifi cation evidence. When identity is
disputed, a trial judge
must advise the jury to consider carefully the circum-
stances of an identifi cation. The requirements for a ‘ Turnbull warning ’ are
summarized in the acronym ADVOKATE. The jury must be advised to con-
sider the following: A mount of time for which the perpetrator was in view.
D istance of the witness from the perpetrator. V isibility of the perpetrator. How
good was the lighting?
O bstruction to the witness
’ view?
K nown to the
witness? Has the witness seen the suspect before? How often? A ny reason to
remember? If only seen occasionally before, did the witness have any reason
to remember the suspect? T ime delay between the incident and the formal
identifi cation procedure. E rror. Is there any material discrepancy between the
description given to the police at the time of the incident and the appearance
of the suspect?
The Police and Criminal Evidence Act 1984
The Police and Criminal Evidence Act 1984 (PACE) established a code of
practice (known as Code D) which governs the conduct of identifi cation pro-
cedures in England and Wales. The current code came into effect in 2008 and
covers a wide range of circumstances of which the major provisions are as
follows. If identifi cation is disputed, a video identifi cation procedure contain-
ing moving images must be offered unless it is not practicable or a
‘ live ’
identifi cation procedure is more suitable. Line - ups should consist of a minimum
of eight foils and one suspect. The foils should, ‘ so far as possible resemble
the suspect in age, general appearance and position in life ’ . Witnesses must be
advised that the person they saw may not be present and must view the entire
line - up at least twice. They are instructed that if they cannot make a positive
identifi cation, they should say so. The person who runs the procedure should
224
Handbook of Psychology of Investigative Interviewing
not be involved in the investigation. The suspect has the right for his or her
legal representative to be present when the witness views the line - up. The
suspect ’ s representative may make reasonable objections to the procedure (e.g.
the selection of foils). If the objections cannot be resolved (e.g., by substitut-
ing a different foil) the reason for the objection must be recorded, together
with a reason why it could not be resolved. If the witness has previously been
shown photographs, details of the photographs shown should be recorded.
Anything the witness says should be written down before he or she leaves the
identifi cation room.
Is PACE w orking?
There have been a number of studies of live line - ups conducted by the police
under the PACE guidelines (Slater, 1994 ; Wright & McDaid, 1996 ; Valentine,
Pickering, & Darling, 2003 ). In all three studies approximately 40% of wit-
nesses identifi ed the suspect, approximately 40% of witnesses did not make
any identifi cation and approximately 20% of witness identifi ed a foil. With
fi eld data such as these it is impossible to know how many line - ups contained
the actual perpetrator. However, we do know that at least 20% of witnesses
made a mistaken identifi cation of an innocent person. It is interesting to note
that the rate of mistaken identifi cation in these studies is higher than the rate
noted by Devlin (20% vs. 9% respectively). At fi rst sight this is a cause for
concern. However, it could be argued that the increased rate in identifi cation
of foils refl ects fairer construction of line - ups. Using foils that resemble the
suspect may make it more likely that a witness will make a mistaken identifi ca-
tion. Nevertheless, the data show that the instruction that ‘ the person seen
may or may not be present ’ did not inhibit many unreliable witnesses from
making an identifi cation.
Archival data collected by the police from 1,776 identity parades showed
that the suspect was identifi ed in 48% of cases, but did not distinguish non -
identifi cations from identifi cation of a foil (Pike, Brace
& Kyman,
2002 ).
An archival analysis of 58 live line - ups conducted in US criminal cases found
that the suspect was identifi ed in 50% of cases, a foil was identifi ed in 24% of
cases and the witness was unable to make an identifi cation or rejected the
line - up in 26% of cases (Behrman & Davey, 2001 ). It should be noted that
in the USA there is no minimum number of foils, although use of fi ve is
common.
Devlin (1976) did not identify any fl aw in the procedure of the identity
parade that substantially increases the risk of error. Instead, it was argued that
the only way to reduce the risk of wrongful conviction was to increase the
burden of proof. Devlin noted that this would make it more diffi cult to
convict the guilty as well as the innocent and he recommended that this course
should be followed with restraint. At the time when the Devlin Report was
published the modern era of research on eyewitness testimony had barely
Recent Developments in Eyewitness Identifi cation Procedures
225
begun. Devlin had the foresight to hear evidence from psychologists and is to
be commended for appreciating the contribution that psychological science
had to make.
Theoretical i ssues in e yewitness i dentifi cation
Relative and a bsolute j udgements
A persistent problem in understanding eyewitness identifi cation is to explain
why a sizeable minority of witnesses make mistaken identifi cations, despite
appropriate warnings that the perpetrator may not be present in the line - up.
Wells (1993) demonstrated that at least part of the problem may be attribut-
able to witnesses who make a relative judgement rather than an absolute judge-
ment. When confronted with a line - up a witness may only identify a person if
their resemblance to the culprit exceeds some criterion of recollection (an
absolute judgement). Alternatively, a witness may examine all the members of
a line - up and identify the person who most closely resembles the perpetrator
(a relative judgement). Wells (1993) and Clark & Davey (2005) have shown
that a proportion of witnesses, who would be able to identify the culprit
if present in a line - up, make a mistaken identifi cation of a foil when asked
to make an identifi cation from a line - up that does not include the culprit
even when the option not to identify anybody is explicitly available. These
data suggest that relative identifi cation decisions are a cause of mistaken
identifi cations.
Sequential vs. s imultaneous p resentation
A method of sequential line - up presentation was developed to prevent wit-
nesses from making a relative judgement. In a sequential presentation, photo-
graphs of faces are presented one at a time (Lindsay & Wells, 1985 ). The
witness is not told how many faces will be presented, but must decide as each
face is presented whether or not it is the culprit before the next face is pre-
sented. The line - up administrator should not know the identity of the suspect.
Furthermore, witnesses must not be allowed a second choice or to see a
gain
a face previously presented (Lindsay, Lea & Fulford, 1991 ).
Sequential presentation has been found to reduce the number of mistaken
identifi cations from culprit - absent line - ups, but it also reduces the number of
correct identifi cations from culprit - present line - ups (Steblay, Dysart, Fulero
& Lindsay, 2001 ). Meissner, Tredoux, Parker & Maclin (2005) found that
sequential line - ups induce a more conservative response criterion, but do not
affect discrimination accuracy. In short, both the guilty and the innocent are
less likely to be identifi ed from a sequential line - up. Use of sequential line - ups
instructions in the context of video identifi cation is discussed below.
226
Handbook of Psychology of Investigative Interviewing
Recollection and f amiliarity
It would be benefi cial to have a better understanding of the conditions that
encourage use of a relative judgement. One possibility is that witnesses may
be prone to using relative judgements when the recognition task is diffi cult.
Valentine, Darling & Memon (2007) found that witnesses were more likely
to report using relative judgements when they saw culprit - absent line - ups than
when they saw a culprit - present line - up. If a witness is able to recollect seeing
a culprit at the crime scene, she or he can rely on an absolute decision.
However, in the absence of recognizing the culprit, the witness may be
tempted to rely on a feeling of familiarity (e.g. ‘ number 2 seems familiar ’ ).
Witnesses may be more likely to rely on a relative judgement when responding
on the basis of familiarity alone, i.e., the witness may identify the person who
feels most familiar. A feeling of subjective familiarity can arise from the per-
ceived typicality of a face or similarity to previously seen faces, other than the
culprit (Bartlett, Hurry & Thorley, 1984 ; Valentine, 1991 ).
The distinction between recollection and familiarity has been made in theo-
retical models of recognition memory (e.g. Mandler, 1980 , Jacoby & Dallas,
1981 ; see Yonelinas, 2002 , for a review). A similar distinction has been char-
acterized as that between remembering and (just) knowing that a stimulus has
been seen before (Tulving, 1985 ). There is empirical evidence that the distinc-
tion captures distinct ways of remembering (see Gardiner
& Richardson