continued search for corroborating evidence is warranted. Further, a positive
identifi cation that is not made with high confi dence should suggest to
Obtaining and Interpreting Eyewitness Identifi cation Test Evidence 215
investigators that there is a very real possibility that their suspect is not the
culprit. Third, although a line - up rejection provides a valuable pointer that
the suspect does not match the witness ’ s memory (Wells & Olson, 2002 ),
the confi dence expressed in a line - up rejection does not assist in determining
whether the rejection is likely to be accurate. Similar interpretations of con-
fi dence recorded at the time of the identifi cation should be made by lawyers,
judges and jurors.
Post - i dentifi cation i nfl uences on c onfi dence
In the previous section we emphasized the importance of obtaining a confi -
dence estimate from the witness immediately after the identifi cation decision.
Why is this important? Basically, the objective should be to obtain a confi dence
judgement that provides an independent assessment of the witness ’ s memory
strength rather than one that is shaped by social infl uences emanating from
post - identifi cation test interactions. We are not suggesting here that an imme-
diately provided verbal confi dence assessment guarantees a precise index of the
witness ’ s memory quality. Rather, we are acknowledging the now overwhelm-
ing body of evidence demonstrating the malleability of identifi cation confi -
dence and, hence, the potential unreliability of delayed post
- identifi cation
confi dence assessments.
After making an identifi cation, a witness may receive feedback (explicit or
implicit) from a number of sources. The line - up administrator might clearly
indicate to the witness that he or she has picked the ‘ right guy ’ (e.g., ‘ That ’ s
our man ’ , ‘ Good, you identifi ed the suspect ’ or the simple statement/question
‘ Great! You would testify to that in court, right? ’ ). Or, the administrator ’ s
facial expression or non - verbal demeanour following the witness ’ s decision
might be interpreted as confi rming the choice. There are many non - verbal
signs of acceptance, or positive reaction such as smiles, head nods and other
spontaneous gestures.
Interestingly, disconfi rming feedback following fi ller identifi cations is also
considered a problem. Telling an eyewitness that she or he has identifi ed a
fi ller leads witnesses to ‘ back off ’ from their identifi cation and claim that they
were not as certain as they in fact were. But, research shows that fi ller identi-
fi cations have diagnostic value because they are more frequent when the
suspect is innocent than when the suspect is guilty (Wells & Lindsay, 1980 ;
Clark & Wells, 2008 ). In effect, witnesses who identify a fi ller are saying that
the person they identifi ed looks more like the perpetrator than does the
suspect. If they are fairly certain in this judgement, then it would make sense
to ask them how certain they are before telling them that they identifi ed a
fi ller.
Confi rming or disconfi rming feedback might also emerge if the witness is
placed in a situation where he or she discusses the identifi cation test with
216
Handbook of Psychology of Investigative Interviewing
another witness to the crime. Such feedback or cues from line - up admini-
strators or co - witnesses are known to exert a powerful effect on witnesses ’
subsequent expressions of confi
dence in their identifi cation decisions.
Confi rming feedback infl ates witness confi dence, whereas disconfi rming
feedback has the opposite effect (Luus & Wells, 1994 ; Wells & Bradfi eld,
1998; 1999
; Bradfi eld, Wells
& Olson,
2002 ; Wells, Olson
& Charman,
2003 ; Hafstad, Memon
& Logie,
2004 ). This pattern occurs both for
positive identifi cations and for line - up rejections, and it occurs for witnesses ’
recollection of their confi dence at the time of the identifi cation and at the time
they are asked about it (Semmler, Brewer & Wells 2004 ). The effect is not
dependent on delivery by a
‘ live ’ administrator, occurring also when deli-
vered by a computer or co - witness (Luus & Wells, 1994 ; Semmler et al. ,
2004 ). Moreover, it has even been detected when the line - up administrator
knew the suspect ’ s identity but did not provide any verbal feedback, reinforc-
ing the potential for infl uence via non - verbal cues (Garrioch & Brimacombe,
2001 ).
Given our earlier observations about the persuasiveness of confi dent wit-
nesses and identifi cations, the implications of these confi dence malleability
fi ndings are reasonably obvious. Witnesses who pick the police suspect and/
or make the same choice as a co - witness may, as a result of cues provided by
the line - up administrator or another witness, end up expressing confi dence
levels way above (or below) what they would have reported if there had been
no interaction with the line - up administrator or co - witness. Imagine the likely
difference in the impact on a jury if a witness reports that they are about 70%
certain that they observed the defendant commit the crime vs. reporting abso-
lute certainty in the identifi cation. Clearly, this malleability of confi dence
judgements means that expressions of confi dence obtained from witnesses in
the courtroom are not only uninformative but also potentially highly mislead-
ing. Moreover, it means that witnesses should be asked to indicate their iden-
tifi cation decision confi dence immediately after making the decision and prior
to any social interactions with line - up administrators or police, and that this
confi dence estimate should be recorded and be the confi dence estimate that
is tendered as evidence.
The consistent implementation of this practice for recording and tendering
confi dence evidence would represent a signifi cant breakthrough. Nevertheless,
we should highlight at least one caveat. Although there is now evidence that
mock - jurors downgrade the credibility of witnesses who display confi dence
infl ation (Bradfi eld & McQuiston,
2004 ), this is not a uniform reaction.
Jones, Williams & Brewer (2008) found that while mock - jurors discredited
witnesses who provided unconvincing reasons for their confi dence infl ation,
they were less likely to do so when the witness was able to offer some plausible
insight that apparently justifi ed the infl ation. This fi nding suggests that simply
tendering as evidence a confi dence estimate obtained at the time of the iden-
tifi cation will not always be suffi cient to combat the impact of confi dence
infl ation.
Obtaining and Interpreting Eyewitness Identifi cation Test Evidence 217
Post - i dentifi cation i nfl uences on o ther
w itness j udgements
Another striking fi nding in the eyewitness identifi cation literature is that post -
identifi cation feedback affects not only witnesses ’ recollections of how confi -
dent they were at the time of
the identifi cation but also their perceptions of
both the witnessing and the identifi cation test experience. For example, post -
identifi cation feedback results in witnesses ‘ infl ating ’ their perceptions of the
quality of their view of the event, the amount of attention they were paying
at the time, and the ease and speed with which they had made the identifi ca-
tion (Wells & Bradfi eld, 1998; 1999 ). Just as witness confi dence shapes judge-
ments of witness credibility, so too are these perceptions likely to shape jurors ’
evaluations of the extent to which the witness ’ s identifi cation should be relied
upon. Accordingly, to ensure that what is essentially distorted evidence does
not shape juror judgements, ways of recording such witness perceptions imme-
diately after the identifi cation test (e.g., via recording of the witness – line - up
administrator interaction) need to be encouraged.
Putting these r ecommendations into p ractice
How do recommendations such as those we have outlined in this chapter get
incorporated into practice? There is no magical formula. There are, however,
a variety of possible approaches, any or all of which may be effective given
the right timing. Clearly the DNA exoneration cases in the USA have been
a catalyst for change in that country and elsewhere. In the USA, a number
of these recommendations (e.g., providing unbiased instructions, obtaining
a confi dence measure directly after the identifi cation) have already been
embodied in the National Institute of Justice Guidelines for the Collection
of Eyewitness Evidence (Technical Working Group for Eyewitness Evidence,
1999 ). Some of these recommendations (e.g., unbiased instructions) have
been widely adopted by police jurisdictions in different parts of the world.
Which recommendations are likely to be adopted, and why, is diffi cult to
ascertain, particularly given the complexity of factors that drive organizational
change.
One thing that the authors advocate, however, is working hard to promote
the implications of scientifi c research to relevant practitioner groups such as
police, lawyers and judges. Over many years Gary Wells has conducted scores
of lectures, workshops, etc. for police, lawyers and judges in the USA. More
recently, Neil Brewer has presented numerous lectures and workshops for
judges and magistrates in most Australian legal jurisdictions. While none of
this guarantees any policy or practical change, a common thread noted by both
authors is that their audiences have engaged enthusiastically and intelligently
in such discussions. Assuming that researchers are persistent in their infl uence
218
Handbook of Psychology of Investigative Interviewing
attempts, such responses bode well for the likelihood of effecting change in
identifi cation test procedures.
Acknowledgement
Research supported by Australian Research Council Discovery Grant
DP0556876.
References
Bradfi eld , A. & McQuiston , D. E. ( 2004 ). When does evidence of eyewitness confi -
dence infl ation affect judgments in a criminal trial? Law and Human Behavior ,
28 , 369 – 387 .
Bradfi eld , A. L. & Wells , G. L. ( 2000 ). The perceived validity of eyewitness identifi ca-
tion testimony: A test of the fi ve Biggers criteria . Law and Human Behavior , 24 ,
581 – 594 .
Bradfi eld , A. L. , Wells , G. L. & Olson , E. A. ( 2002 ). The damaging effect of confi rm-
ing feedback on the relation between eyewitness certainty and identifi cation
accuracy . Journal of Applied Psychology , 87 , 112 – 120 .
Brewer , N. ( 2006 ). Uses and abuses of eyewitness identifi cation confi dence . Legal and
Criminological Psychology , 11 , 3 – 4 .
Brewer , N. & Burke , A. ( 2002 ). Effects of testimonial inconsistencies and eyewitness
confi dence on mock - juror judgments . Law and Human Behavior , 26 , 353 – 364 .
Brewer , N. & Wells , G. L. ( 2006 ). The confi dence – accuracy relationship in eyewitness
identifi cation: Effects of line - up instructions, foil similarity and target - absent base
rates . Journal of Experimental Psychology: Applied , 12 , 11 – 30 .
Brewer , N. , Weber , N. & Semmler , C. ( 2005 ). Eyewitness identifi cation . In N. Brewer
& K. D. Williams (Eds.), Psychology and law: An empirical perspective (pp. 177 –
221 ). New York : Guilford Press.
Clark , S. E. ( 2005 ). A re - examination of the effects of biased line - up instructions in
eyewitness identifi cation . Law and Human Behavior , 29 , 395 – 424 .
Clark , S. E. & Wells , G. L. ( 2008 ). On the diagnosticity of multiple - witness identifi ca-
tions . Law and Human Behavior , 32 , 406 – 422 .
Cutler , B. L. & Penrod , S. D. ( 1995 ). Mistaken identifi cation: The eyewitness, psychology,
and the law . New York : Cambridge University Press .
Cutler , B. L. , Penrod , S. D. & Stuve , T. E. ( 1988 ). Jury decision making in eyewitness
identifi cation cases . Law and Human Behavior , 12 , 41 – 56 .
Douglass , A. B. , Smith , C. & Fraser - Thill , R. ( 2005 ). A problem with double - blind
photospread procedures: Photospread administrators use one eyewitness ’ s confi -
dence to infl uence the identifi cation of another witness . Law and Human Behavior ,
29 , 543 – 562 .
Garrioch , L. & Brimacombe , C. ( 2001 ). Line - up administrators’ expectations: Their
impact on eyewitness confi dence . Law & Human Behavior , 25 , 299 – 314 .
Hafstad , G. S. , Memon , A. & Logie , R. ( 2004 ). Post - identifi cation feedback, confi -
dence and recollections of witnessing conditions in child witnesses
.
Applied
Cognitive Psychology , 18 , 901 – 912 .
Obtaining and Interpreting Eyewitness Identifi cation Test Evidence 219
Harris , M. J. & Rosenthal , R. ( 1985 ). Mediation of interpersonal expectancy effects:
31 meta - analyses . Psychological Bulletin , 97 , 363 – 386 .
Innocence Project
(
2009 ).
Innocence project .
http://www.innocenceproject.org/
about/index.php . Accessed 23 March 2009.
Jones , E. E. , Williams , K. D. & Brewer , N. ( 2008 ). ‘ I had a confi dence epiphany! ’ :
Obstacles to combating post - identifi cation confi dence infl ation . Law and Human
Behavior , 32 , 164 – 176 .
Juslin , P. , Olsson , N. & Winman , A. ( 1996 ). Calibration and diagnosticity of confi -
dence in eyewitness identifi cation: Comments on what can be inferred from the
low confi dence - accuracy correlation . Journal of Experimental Psychology: Learning,
Memory, and Cognition , 22 , 1304 – 1316 .
Keast , A. , Brewer , N. & Wells , G. L. ( 2007 ). Children ’ s metacognitive judgments in
an eyewitness identifi cation task . Journal of Experimental Child Psychology , 97 ,
286 – 314 .
Lindsay , D. S. , Read , J. D. & Sharma , K. ( 1998 ). Accuracy and confi dence in person
identifi cation: The relationship is strong when witnessing conditions vary widely .
Psychological Science , 9 , 215 – 218 .
Lindsay ,
R. C. L.
,
Wells ,
G. L.
& O ’ Connor ,
F. J.
(
1989 ).
Mock - juror belief of
accurate and inaccurate eyewitnesses
.
Law and Human Behavior ,
13 ,
333 –
339 .
Luus , C. & Wells , G. L. ( 1994 ). The malleability of eyewitness confi dence: Co - witness
and perseverance effects . Journal of Applied Psychology , 79 , 714 – 723 .
Malpass , R. S. & Devine , P. G. ( 1981 ). Eyewitness identifi cation: Line - up instruc-
tions and the absence of the offender . Journal of Applied Psychology , 66 , 482 –
489 .
Meissner , C. A. & Brigham , J. C. ( 2001 ). A meta - analysis of the verbal overshadowing
effect in face identifi cation . Applied Cognitive Psychology , 16 , 911 – 928 .
Memon , A. , Gabbert , F. & Hope , L. ( 2004 ). The ageing eyewitness . In J. Adler (Ed.),
Forensic psychology: Debates, concepts and practice . Cullompton : Willan Publishing .
Morrell , H. E. R. , Gaitan , S. & Wixted , J. T. ( 2002 ). On the nature of the decision
axis in signal
- detection - based models of recognition memory
.
Journal of
Experimental Psychology: Learning, Memory and Cognition , 28 , 1095 – 1110 .
Pike , G. , Brace , N. & Kynan , S. ( 2002 ). The visual identifi cation of suspects: Procedures
and practice . Briefi ng Note 2/02. London : Home Offi ce .
Potter , R. & Brewer , N. ( 1999 ). Perceptions of witness behaviour – accuracy relation-
ships held by police, lawyers and jurors
.
Psychiatry, Psychology and Law ,
6 ,
97 – 103 .
Schooler ,
J. W.
& Engstler - Schooler ,
T. Y.
(
1990 ).
Verbal overshadowing of
visual memories: Some things are better left unsaid
.
Cognitive Psychology ,
22 ,
36 – 71 .
Semmler , C. , Brewer , N. & Wells , G. L. ( 2004 ). Effects of postidentifi cation feedback
on eyewitness identifi cation and nonidentifi cation confi dence . Journal of Applied
Psychology , 89 , 334 – 346 .
Sporer , S. L. , Penrod , S. D. , Read , J. D. & Cutler , B. L. ( 1995 ). Choosing, confi dence,
and accuracy: A meta - analysis of the confi dence - accuracy relation in eyewitness
identifi cation studies . Psychological Bulletin , 118 , 315 – 327 .
Handbook of Psychology of Investigative Interviewing: Current Developments and Future Directions Page 42