Content uploaded by Roy S. Malpass
Author content
All content in this area was uploaded by Roy S. Malpass on Mar 01, 2014
Content may be subject to copyright.
3
Chapter 1
The Need for Expert Psychological Testimony
on Eyewitness Identifi cation
Roy S. Malpass, Stephen J. Ross, Christian A.
Meissner and Jessica L. Marcon
The evidence requirements for eyewitness identifi cation were improved in the
middle of the 19th century by a Middlesex magistrates’ court in the United
Kingdom (Malpass, Tredoux, & McQuiston-Surrett, 2007 ), which established
lineups as a means of protecting innocent suspects from false identifi cation.
Members of the many groups involved in the criminal justice system have
recognized the fragility of eyewitness identifi cation and a troublingly
large possibility of error that sends innocent people to jail. Although the
problem has appeared to be resistant to reduction, it has not been diffi cult
to document.
Evidence of a need for development of diagnostic procedures for obtain-
ing and evaluating eyewitness and other memory-based evidence has been
apparent for many decades. Borchard ( 1932 ) reviewed 65 criminal cases
involving persons known to be completely innocent. He began with this
analysis: “Perhaps the major source of these tragic errors is an identifi cation
of the accused by the victim of a crime of violence. This mistake was practi-
cally alone responsible for twenty-nine of these convictions.” He continues
“How valueless are these identifi cations by the victim of a crime is indicated
by the fact that in eight of these cases, the wrongfully accused person and the
really guilty criminal bore not the slightest resemblance to each other, whereas
in twelve other cases, the resemblance, while fair, was still not at all close.
In only two cases can the resemblance be called striking” (p. xiii). Borchard
Cutler Chapters.indd 3Cutler Chapters.indd 3 4/30/2009 8:21:53 PM4/30/2009 8:21:53 PM
4 Expert Testimony on the Psychology of Eyewitness Identifi cation
made a series of recommendations, which had little detectable impact on the
legal system.
Brandon and Davies ( 1973 ) similarly reviewed 70 cases from the United
Kingdom. As to causes of wrongful convictions, they write: “Since eyewitness
identifi cation is a very common form of evidence in criminal cases, it is per-
haps not surprising that a large proportion of the mistakes we have come
across occur in this fi eld. Nevertheless, a greater number of mistakes seem to
occur in this fi eld even than one would expect. Of the cases we have examined
of people who have subsequently been pardoned, or whose convictions have
been quashed, or sentences remitted, a remarkably high proportion have
involved misidentifi cation” (p. 24).
Almost 45 years after Borchard ( 1932 ) and shortly after Brandon and
Davies ( 1973 ) — and again in the United Kingdom — Devlin ( 1976 ) was asked
to lead an inquiry into eyewitness identifi cation after two cases of wrongful
conviction came to the attention of the public. Devlin made a detailed series
of recommendations; however, there was little effect leading to immediate
changes in the administration of investigations and the development of
eyewitness evidence. Devlin’s impact was subtle and took time; some of his
impact may be seen in the U.K. National Court of Appeal decision in
R v. Turnbull ( 1976 ), but substantial changes were not visible until the Police
and Criminal Evidence (PACE) Act of 1984 and its subsequent updates. The
organization of law enforcement and its administration into a highly
centralized entity in the United Kingdom likely lends itself to more effective
centralized research and development processes, compared with the highly
fractionated criminal justice system in the United States.
Radelet, Bedau, and Putnam ( 1992 ) reviewed the magnitude of wrongful
convictions in 400 capital cases in the United States. They assess the causes as
follows: “As for the causes of the errors, our research has shown that the two
most frequent are perjury by prosecution witnesses and mistaken eyewitness
testimony” (p. 18). Shortly afterward, Huff, Ratner, and Sagarin ( 1996 )
reviewed more than 200 cases of wrongful conviction in the United States and
articulated the distribution of errors contributing to such instances. Out of
205 cases in their database, the authors cite eyewitness misidentifi cation as a
major contributor to 100 of these — just under 50 % of the cases. They write:
“We believe that the single most important factor leading to wrongful convic-
tion in the United States and England is eyewitness misidentifi cation . . . This
is shown not only in our database but in the responses to our questionnaire
[sent to attorneys and judges], where nearly 8 out of 10 ranked witness error
(primarily witness misidentifi cation, but also including some less frequent
types of witness error) as by far the most frequent type of error leading to false
convictions.”
At about the same time, events began to move forward on two fronts. The
need for a focused response was broadly recognized among eyewitness
researchers, and a policy paper was composed and published by the American
Cutler Chapters.indd 4Cutler Chapters.indd 4 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 5
Psychology-Law Society in the Society’s journal Law and Human Behavior.
This paper (Wells et al., 1998 ) contained a focused set of recommendations
based on a review of fi ndings of well-established lines of scientifi c investiga-
tion. The second front was manifested in a report from a study commissioned
by the National Institute of Justice on the exoneration of wrongfully con-
victed persons via the then-new technology of DNA analysis (Connors,
Lundregan, Miller, & McEwen, 1996 ). The results of the Connors and
colleagues’ ( 1996 ) analysis was examined by Wells and associates ( 1998 ) to
show that, of the many causes associated with wrongful conviction, faulty
eyewitness identifi cation was primary among them. This outcome was
recognized by the National Institute of Justice, which subsequently convened
the Technical Working Group on Eyewitness Evidence that completed their
report in 1999 . The Technical Working Group was composed of police inves-
tigators, prosecutors, defense attorneys, and scientists, and was tasked to
develop a guide for law enforcement representing best practices in the devel-
opment of eyewitness identifi cation evidence and the administration of eye-
witness identifi cation procedures by law enforcement (Technical Working
Group on Eyewitness Evidence, 1999 ).
The most intensive and effective review of wrongful convictions has
been carried out by the Innocence Project, created by Barry C. Scheck and
Peter J. Neufeld and affi liated with the Benjamin N. Cardozo School of Law at
Yeshiva University, New York City. As of this writing, 225 persons have been
exonerated through the efforts of the Innocence Project and released from
incarceration as innocent persons. Their assessment of the causes of wrongful
conviction contains the following statement: “Eyewitness misidentifi cation
is the single greatest cause of wrongful convictions nationwide, playing a
role in more than 75 % of convictions overturned through DNA testing.”
And “In case after case, DNA has proven what scientists already know —
that eyewitness i dentifi cation is frequently inaccurate. In the wrongful
convictions caused by eyewitness misidentifi cation, the circumstances varied,
but judges and juries all relied on testimony that could have been more
accurate if reforms proven by science had been implemented” (Innocence
Project, 2008 ).
The point not to miss is this: Failures in the development of diagnostic
procedures for obtaining eyewitness identifi cation evidence, failures in the
administration of eyewitness identifi cation procedures, and failures in the eval-
uation of eyewitness testimony by police, attorneys, jurors, judges, the public,
and the American criminal justice system, considered as a fact-fi nding entity,
have been documented repeatedly over at least the last 75 years, with remark-
ably little effect (in the United States) on the routine operation of these systems
of justice. Yet, a scientifi c approach to understanding the causes of eyewitness
misidentifi cation has been undertaken, and a body of knowledge and the means
of expanding it to benefi t society in general and the criminal justice system in
particular is available to law enforcement entities. This chapter explores the
Cutler Chapters.indd 5Cutler Chapters.indd 5 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
6 Expert Testimony on the Psychology of Eyewitness Identifi cation
underlying research base on eyewitness identifi cation, including the scientifi c
basis upon which expert testimony relies. We begin by considering confl icting
knowledge systems employed in the criminal justice system.
Confl icting Knowledge Systems: Customary Versus
Scientifi c Knowledge
Imagine that two societal institutions are concerned with the same domain of
knowledge, but they do not share information about what they know, what
they do not know, and how they use the knowledge they have.
Imagine that the fi rst of these institutions makes judgments about whether
people will lose their freedom, live, or die based in whole or in part on accusa-
tions made by other persons based on their memory, judgment, and response
to various events in their cognitive and social environment. Participants in
this process make laws governing the process, measure the process against
constitutional requirements, administer the process of bringing accused per-
sons into the process, develop evidence for and against their guilt, and judge
the validity of the evidence. Imagine further that these participants in the pro-
cess — through which life or death decisions are made — are informed primar-
ily by customary knowledge and personal experience.
Imagine that the second of these institutions uses scientifi c techniques of
investigation to study the processes through which persons come to have the
knowledge — beliefs, perception, memory, etc. — that forms the basis of their
accusations that a particular individual acted in a certain way or committed
specifi c acts at a specifi c time and place. Additionally, this institution studies
the ways in which events subsequent to observation of an event may change
an individual’s memory for events and/or persons. They further study the
ways in which errors of memory and reports based on memory contribute to
the judgment processes — and errors — of the fi rst institution (including the
ways in which memory is contaminated in the process of investigation and
prosecution). These matters, known to the second institution through scien-
tifi c study, are largely unaddressed in the procedural manuals, training, and
oversight processes of the fi rst institution.
We know that customary knowledge is commonly used in many areas of
our society and in a number of domains of social life. For example, individu-
als treat themselves (and sometimes others) for various ailments using folk
medicines that have never seen scientifi c evaluation, and modern govern-
ments (e.g., Canada and the United States) have made provision for certain
indigenous groups to exercise customary legal and governance processes as
part of their cultural and tribal autonomy. But we also know that folk medi-
cines and treatment regimes have been supplanted and regulated in the light
of science-based medicine and surgical practice, that modern building codes
supplant customary principles of construction, and that modern property law
has supplanted folk concepts of ownership (Malpass, 1999 ).
Cutler Chapters.indd 6Cutler Chapters.indd 6 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 7
Science is a means of evaluating claims to knowledge based on open, pub-
lic, explicit procedure. Customary practices are not amenable to such evalua-
tions. Modern knowledge based on scientifi c analysis has made important
contributions to many areas of society, and this process continues in many
areas of criminal investigation and legal process — however, the degree of
penetration has been quite modest. We will discuss the contrast between
customary and scientifi c knowledge later, but for now it might be surprising
that, in a time when scientifi c contributions to the improvement of life and
knowledge have been strong in so many important areas of public and private
life, there continues to be vigorous resistance to the contribution of science-
based information to criminal justice processes and especially to the delibera-
tions of those every-day people — lay jurors — who make life and death
decisions about persons accused of crimes.
To some extent, this should not be surprising, since an adversarial
approach is hard-wired into our system of justice, such that both prosecution
and defense are devoted to making the best case possible (within certain broad
limits) for their “side.” It is expensive to replace customary knowledge and
traditional approaches with scientifi cally based information and procedures.
And a science-based approach requires knowledge beyond the ken of many
practitioners in the legal system, from police and attorneys to judges and leg-
islators. Progress on this front has been slow. It is in the adversarial interest of
prosecutors in criminal cases (although not necessarily in the interest of jus-
tice) to keep modern scientifi c analysis — as contrasted with common sense–
based “common knowledge” — out of the legal system. Although one can
argue for many interpretations of evidence based on customary knowledge,
interpretations based on a scientifi c footing are less easy to mold to the pur-
poses of a particular case.
It is reasonable to inquire into the critical differences between customary
knowledge and science. To start with, scientifi c knowledge is designed from
the ground up to be questioned, challenged, evaluated as to its validity and
reliability, and ultimately improved. Science provides criteria against which
accuracy and adequacy can be evaluated. Apart from what people generally
believe, the law has no way of applying validity tests to substantive knowledge
other than by asking experts.
Science-based improvements can be monitored according to procedures
and evaluative criteria that are open, accessible to anyone with the background
to understand them, and which become part of the public domain and dis-
course. Customary knowledge, on the other hand, has merely the force of
common belief and is a standard of society by default — rather than evaluated
and improved by rational methods. Evaluation and development of custom-
ary knowledge and practice more often occur against criteria of general agree-
ment or political popularity.
Customary knowledge has no explicit origins, it has no records of the
conditions on which beliefs are based that might facilitate replication, and it
provides no way for interested parties to test the reliability and generalizability
Cutler Chapters.indd 7Cutler Chapters.indd 7 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
8 Expert Testimony on the Psychology of Eyewitness Identifi cation
of the resulting conclusions beyond specifi c forms of personal experience.
Rather customary knowledge relies on the perceptions and memories of
unknown people under unexamined conditions and in unknowable circum-
stances; thus it is for these reasons mostly incapable of being evaluated for
validity in anything but an impressionistic way.
Science has made great advances specifi cally because the history of scien-
tifi c fi ndings is publicly available. The records of this history are explicit,
quantitative, and tied to procedures that can be repeated for contemporary
evaluation. Reliance on memory is minimized because the publication stan-
dards of academic journals and other sources in which scientifi c studies are
published require records that can be produced for inspection. Detailed infor-
mation is available about the individuals involved, their scientifi c history and
credentials, their institutional location, and the methods and procedures on
which observations are based. Records of the original observations frequently
are available for scrutiny and reanalysis.
A scientifi c approach to the collection and evaluation of eyewitness evi-
dence is far more likely to result in improved effectiveness, and it is more
likely to benefi t from both new science and the continuing experience of the
criminal justice system with obtaining and evaluating evidence based on eye-
witness testimony. A systematic eye to improving the quality of the informa-
tion contributed to criminal investigation by eyewitnesses and the way it is
used by the criminal justice system can bring important benefi ts. Application
of a scientifi c approach is likely to provide greater benefi ts the closer it is to the
initial investigation. Placed within the investigation process in law enforce-
ment organizations, science provides a powerful set of tools. Brought in after
an investigation has run its course and a prosecution is under way, its power
and scope is greatly reduced. In lieu of its presence in the initial investigation
and the framing of the prosecutions’ case, scientifi cally based expert testimony
is an important way to bring science into the process.
Arguments regarding expert testimony and its legitimacy for the court-
room are endemic to the adversarial system. Many debates have been under-
taken over whether eyewitness researchers should testify as experts in criminal
trials and the moral and ethical dilemmas that such testimony elicits (i.e.,
Clifford, 1997 ; Ebbesen & Konečni, 1996 ; Elliot, 1993 ; Loftus, 1983 ; McCloskey
& Egeth, 1983 ; McCullough, 2002 ; Wells, 1993 ; Yarmey, 2001 ). It is not our
goal to rehash these arguments and determine the answer to the question of
whether we should be testifying. This issue is a topic for the courts to decide,
and we feel that it should remain so. Instead, we focus on the arguments that
arise from evaluating the utility of the research on eyewitness identifi cation.
Specifi cally, these arguments are related to whether the testimony is needed
and whether the research on which the testimony is based is scientifi cally
valid.
We have already documented that a need exists for understanding
the factors that infl uence eyewitness identifi cation errors, because these
errors are a principal cause of wrongful convictions of innocent individuals.
Cutler Chapters.indd 8Cutler Chapters.indd 8 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 9
However, even though overwhelming evidence suggests that eyewitness
identifi cations are fallible, arguments have arisen as to whether it is necessary
for social scientists to testify about these issues in the courts. A common argu-
ment against the admissibility of expert testimony is that jurors already
know about the fallibility of eyewitness memory and, therefore, expert testi-
mony on these issues is not needed. We offer the following as a contemporary
illustration. From WJBC, AM1230, Radio Bloomington, the Voice of McLean
County (Bloomington, Illinois) comes this news report, dated 04/24/2008:
A McLean County judge will not let an expert about witness
identifi cation testify during next month’s rape trial for former
Bloomington Police Sergeant Jeff Pelo. Pelo is accused of sexually
assaulting four women, dating back to 2002. Judge Bob Freitag says
there’s no compelling reason to allow an Ohio expert to testify
about factors, such as memory, that can affect eyewitness identifi ca-
tion. Prosecutor Mark Messman had argued putting such an expert
on the stand to sort out what are common sense issues would only
make for a “battle of experts.” (WJBC, 2008 ; italics added)
Read and Desmarais (Chapter 6) dissect this “common sense” argument
in detail, and therefore we will not take too much space to discuss their thesis.
In short, they present evidence from multiple studies evaluating layperson
knowledge of factors affecting eyewitness identifi cations that have shown that
jurors do not understand many of the variables that infl uence eyewitness
identifi cations. In the few situations in which jurors do understand, they do
not know how to apply their understanding to the interpretation of evidence.
Laymen may have some forms of knowledge about eyewitness matters (e.g.,
that certain instructions infl uence eyewitness identifi cation), but they are
much less likely to know what moderates that effect, such as not getting a
good opportunity to view the face at the witnessing event, thus making a false
identifi cation more likely if an unbiased instruction is not used. In general,
laymen (jurors) are unlikely to have discovered for themselves the fruits of the
last 50 years of scientifi c research on eyewitness identifi cation and memory.
In addition to the “common sense” approach, critics of expert testimony
on eyewitness issues have argued that the legal system provides effective safe-
guards against erroneous eyewitness identifi cations leading to wrongful
convictions. Expert testimony is therefore unnecessary. These safeguards
include the presence of counsel during identifi cation procedures, the use of
voir dire to discover jurors who may be unwilling or unable to carefully evalu-
ate eyewitness evidence, motions to suppress identifi cations, cross-examina-
tion of eyewitnesses and procedure administrators during trial, and judicial
instructions. Devenport, Kimbrough and Cutler (Chapter 3) present research
that has evaluated the effectiveness of these safeguards and the defense of
“common sense” criticism; we direct the reader to their chapter for a detailed
evaluation of the validity of this argument. Briefl y, their research suggests that
these safeguards are relatively ineffective for preventing eyewitness errors
Cutler Chapters.indd 9Cutler Chapters.indd 9 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
10 Expert Testimony on the Psychology of Eyewitness Identifi cation
from being presented at trial and infl uencing juror judgments. A closer exam-
ination of their research indicates that judges and attorneys (similar to jurors)
are generally insensitive to the factors that infl uence eyewitness accuracy,
thereby limiting the effectiveness of these safeguards. It is evident that layper-
son, attorney, and judicial insensitivity to the factors that infl uence eyewitness
memory present a need for expert testimony to serve as a knowledge transfer
approach.
What can be said, though, regarding the scientifi c merit of the research
upon which expert eyewitness testimony is based? Other authors in this vol-
ume discuss the ecological validity and generalizability of eyewitness research
in detail (Flowe, Finklea, & Ebbesen, in Chapter 9; Bailey & Mecklenburg, in
Chapter 10). We, therefore, focus our efforts on evaluating the methodologies
of the eyewitness research.
Is the Study of Eyewitness Memory and Identifi cation a
Field of Science?
The study of eyewitness identifi cation emerged from basic research in experi-
mental psychology, particularly with regard to the study of human memory,
which was an active fi eld of scientifi c study in the last half of the 19th century.
At the beginning of the 20th century, the fi rst laboratory containing a research
program on eyewitness identifi cation was established in the Psychology
Department at Harvard University by Prof. Hugo Münsterberg. With the
birth of what has become known as “cognitive psychology” in the 1960s,
research on perception, memory, and decision-making processes has greatly
accelerated. During this period, research specifi cally on the memory of eye-
witnesses was vastly expanded, and it has been the subject of inquiry by psy-
chologists from many diverse areas of psychological science, including
cognitive, social, and developmental psychology.
Experimental psychology and the applied fi eld of eyewitness identifi ca-
tion and memory are widely recognized as fi elds of scientifi c study, embody-
ing the techniques, methodologies, and standards that defi ne science in
relation to other forms of knowledge. The fi eld is characterized by the devel-
opment and evaluation of explicit theory and the use of controlled studies,
with experimental methods and research designs capable of producing clearly
interpretable results and theoretical advances. Further, researchers seek to
generalize their fi ndings to encompass the real-world constraints that wit-
nesses may be exposed to.
In this section, we describe the organizational structure of scientifi c psy-
chology, placing the study of eyewitness memory within this system. We then
briefl y describe the methodologies that make the study of eyewitness memory
a scientifi c endeavor, and we discuss the concept of general agreement within
the scientifi c community. Finally, we highlight the reliable phenomena that
scientists have discovered with regard to eyewitness memory.
Cutler Chapters.indd 10Cutler Chapters.indd 10 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 11
The Organizational Structure of Scientifi c Psychology
The International Congress of Scientifi c Unions (ICSU) is the umbrella
international organization of all scientifi c fi elds in the world. Representing
psychology in this world-wide scientifi c organization is the International
Union of Psychological Sciences (IUPsyS). Members of IUPsyS are national
psychological associations rather than individual scientists. Both the American
Psychological Association (APA) and the Association for Psychological
Science (APS) are members of IUPsyS, and many other specialty societies are
affi liated. One of the oldest international organizations in which individual
psychologists can be members is the International Association for Applied
Psychology (IAAP), founded in the 1920s in Europe.
The major umbrella organization for the sciences in the United States is
the American Association for the Advancement of Science (AAAS). Serving
more than 10 million members, AAAS is the world’s largest federation of sci-
entifi c and engineering societies. The premiere scientifi c publication in the
United States is Science , an offi cial publication of the AAAS. Researchers
studying both the basic and applied aspects of memory have published their
fi ndings in this journal (e.g., Loftus, 1979 ).
Among the 271 societies and academies of science affi liated with AAAS
are the two most prominent associations in the fi eld of scientifi c psychology,
namely the APS and the APA. Researchers who conduct studies on eyewitness
memory are members of these organizations, as well as of several specialized
divisions or sections of the just-noted societies devoted to psychology and
law, including the American Psychology-Law Society (AP-LS, Division 41 of
APA) and the Division of Psychology and Law in the IAAP (Division 10).
Many scientifi c specialty organizations in American psychology, including the
Psychonomic Society, the Society for Research in Child Development (SRCD),
the Society of Personality and Social Psychology (SPSP), and the Society for
Applied Research in Memory and Cognition (SARMAC) include eyewitness
scientists among their members and publish research on eyewitness identifi -
cation and memory in their journals.
The Scientifi c Methodologies of Eyewitness Research
Science is generally characterized by a form of knowledge development based
on empirical observation. Theories are tested and modifi ed through observa-
tion of events in the environment that are intersubjective in nature — that is,
different observers can agree on whether an event occurred and about the
attributes of the event. Like all sciences (including the “hard sciences” such
as engineering, physics, chemistry, and biology), psychologists rely upon
basic principles of scientifi c inquiry that ensure the reliability and validity
of their fi ndings and that alleviate bias or error resulting from the intersubjec-
tive characteristics of observation. These methodological principles involve
(a) strict adherence to the scientifi c method, with a focus on the concept of
Cutler Chapters.indd 11Cutler Chapters.indd 11 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
12 Expert Testimony on the Psychology of Eyewitness Identifi cation
falsifi ability; (b) experimental design that ensures both internal and external
validity; (c) reliance upon statistical inference and the provision of error rates
associated with each hypothesis test; and (d) a peer review process that pro-
vides quality control over studies that are ultimately published. An important
publication (Hedges, 1987 ) favorably compares the reliability and replicabil-
ity of fi ndings in scientifi c psychology with the physical sciences. Specifi cally,
Hedges ( 1987 ) evaluated the statistical consistency of research in particle
physics with that of psychology and found that the results of research in par-
ticle physics were no more consistent than the results found in psychological
research. In short, the results in psychology were just as reliable and replicable
as those in the “hard science” of particle physics.
The Scientifi c Method
All scientifi c endeavors, regardless of discipline, must adhere to a common
method in which theoretical inquiry allows one to generate specifi c hypothe-
ses that are objective and clearly testable. Such hypotheses are tested based
upon commonly accepted paradigms or measurement standards within a
given fi eld, and the results of these tests permit scientists to either support or
refute a specifi c theoretical position. A critical requirement of this method of
inquiry is commonly referred to as falsifi cation — the notion that it must be
possible to disprove any given theoretical proposition through reliable meth-
ods of assessment. Thus, the scientifi c method provides a manner in which
scientists may test, or falsify, a given theoretical position.
Experimental Design That Ensures Both Internal and
External Validity
Hypotheses may be tested in a number of ways via the scientifi c method,
including direct observation of behavior ( observational designs ), assessing the
relative association between two variables of interest ( correlational designs ),
identifying groups of individuals that may differ in the hypothesized manner
( quasi-experimental designs ), or experimentally inducing the hypothesized
differences in groups of individuals that have been randomly assigned to the
relevant conditions (true experimental designs ). Although each of these
approaches is used across the various sub-disciplines of psychological science,
it is important to note that these approaches differ in the extent to which
causal inferences may be generated and strong internal validity may be estab-
lished. Specifi cally, true experimental designs provide the most robust test
regarding the direct effect of one variable on another. These designs involve
the random assignment of participants to “conditions” (or groups) that
systematically vary along a specifi ed dimension relating to the proposed
hypothesis. Such an experimental treatment of the hypothesis allows scientists
to infer that any differences between conditions are the result of the proposed
hypothesis (because other potential factors are controlled through aspects
Cutler Chapters.indd 12Cutler Chapters.indd 12 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 13
of the design). Although psychologists studying eyewitness memory have
relied upon the variety of designs just described, the most robust fi ndings
stem from those studies employing a true experimental design that ensures
strong internal validity.
One critique often sounded against the domain of eyewitness psychology
is that the experiments conducted fail to simulate the reality of a crime or the
“real world.” In scientifi c terms, this critique centers on the degree of ecologi-
cal validity demonstrated by the science — or the extent to which the fi ndings
of a given study can generalize to the population from which the sample was
generated or the real-world conditions in which the phenomenon might be
observed. For example, one might inquire whether university students who
participate in experimental research differ from those individuals who might
fi nd themselves witnesses to a real crime, or whether a laboratory demonstra-
tion of a given effect might generalize to perception and memory in everyday
life. As mentioned previously, other contributors address this ecological valid-
ity concern later in this volume (Flowe and colleagues, in Chapter 9; Bailey &
Mecklenburg, in Chapter 10), therefore we will not go into great detail in
discussing it. However, we are compelled to mention that there appear to be
very few, if any, discontinuities in the processes that infl uence individuals in
experimental research and the real world. Perception and memory processes
do not work in one way under one circumstance and in quite another way in
a different situation. Most events effecting memory and witness testimony are
located on an ordered continuum, and the functional relationship is a mono-
tonic function: As the causative factor increases, so does the result. With some
well-known exceptions (e.g., stress and arousal, Deffenbacher, Bornstein, &
Penrod, 2006 ), this appears to be true across a wide range of factors, and even
what were thought to be exceptions recently have appeared to be less
exceptional.
One approach taken by scientists to generalize their results involves sys-
tematically examining a phenomenon across a variety of conditions and
methodological parameters. Although scientists may begin to explore a phe-
nomenon by employing controlled laboratory experiments, they will often
seek to generalize the observed phenomenon by taking their tests of the effect
out of the laboratory and into the real world. Eyewitness researchers often
utilize experimental paradigms that more appropriately replicate the condi-
tions of a real-world event, such as unintentional encoding conditions and
signifi cant delays between the presentation of the event and the administra-
tion of a lineup. Researchers may also conduct archival analyses of real-world
cases to determine if a given phenomenon has any effect on real witnesses.
One example of this approach can be seen in studies on the cross-race
effect in memory for faces (i.e., the fi nding that memory for faces of one’s own
race is superior to memory for faces of another, less familiar race). The fi rst
studies examining the cross-race effect involved a laboratory experiment in
which White and Black university participants viewed a sequence of slides
depicting photographs of same- and other-race faces (Malpass & Kravitz, 1969 ).
Cutler Chapters.indd 13Cutler Chapters.indd 13 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
14 Expert Testimony on the Psychology of Eyewitness Identifi cation
Although this phenomenon has been replicated in the laboratory many times
over the past 40 years (see Meissner & Brigham, 2001 ), researchers
have also demonstrated that the effect is not unique to individuals of certain
racial or ethnic backgrounds (or to university participants), and that the
phenomenon is observed both in “eyewitness” paradigms that capture many
real-world elements and archival analyses of true witness identifi cations.
Statistical Inference and Error Rates
As described earlier, the process of scientifi c inquiry involves the assessment
of relationships between variables and the testing of differences between
groups of individuals who are randomly assigned to experimentally differen-
tiated conditions. The process of scientifi c work and the evaluation of theories
and hypotheses inherently involve the calculation of error rates associated
with the work. In fact, most areas of psychological science use inferential sta-
tistical analysis, in which the calculation of error rates is a basic part of the
analysis and interpretation of research fi ndings. Generally, to be considered
statistically reliable and reported in scientifi c journals, observed fi ndings must
rise above levels of error by specifi c amounts calculated from data and dic-
tated by quantitative analytic techniques developed and accepted by research-
ers over a period of more than a century.
The Process of Peer Review
The International Union (IUPsyS), IAAP, APA, APS, and various specialty
organizations (such as the Psychonomic Society, SARMAC, and AP-LS) pub-
lish scientifi c journals in psychology, many of which include scientifi c work
on memory and eyewitness identifi cation. All of these journals engage in a
process of peer review that is devoted to evaluating the validity of those scien-
tifi c studies submitted for consideration. The peer review process is a method
of quality control that ensures the validity and reliability of experimental
research. Papers submitted for publication are reviewed by leading scientists
(experts) in the relevant area. These experts provide their critique of the stud-
ies in an anonymous fashion, including such aspects as the quality of theoreti-
cal contribution, the validity and appropriateness of the scientifi c methods
and procedures employed, and the appropriateness of the data analysis and
interpretations. The editor who solicits these reviews is then responsible for
permitting revisions to the work by the authors (if such revisions would pro-
mote the scientifi c value of the study) or rejecting the manuscript for publica-
tion (if fatal fl aws in the study are apparent to the reviewers, such that the
study contributes no scientifi c value to knowledge development). Journals
using the peer review system generally accept only a small percentage of the
manuscripts that have been submitted for publication — thus, studies pub-
lished in these outlets have passed a rigorous test and are generally considered
worthy of consideration by the greater scientifi c community.
Cutler Chapters.indd 14Cutler Chapters.indd 14 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 15
Determining General Acceptance Within the
Scientifi c Community
As with other scientifi c communities, there is general agreement about the
scientifi c fi ndings of the eyewitness community (see Hosch, Jolly, Schmersal, &
Smith, Chapter 7). General agreement can be evidenced objectively in many
ways, including qualitative and quantitative (meta-analytic) reviews of the lit-
erature, representation in primary texts representing the science, through
documents that chronicle the fi ndings of consensus panels of scientifi c experts,
or more directly through surveys of scientifi c experts. A review of these areas
suggests that it would be very diffi cult to sustain the position that many of the
fi ndings in research on eyewitness memory lack general agreement within the
scientifi c community.
Qualitative and Quantitative (Meta-analytic) Reviews of the
Literature
Comprehensive reviews of the nature, content, and contemporary achieve-
ments of scientifi c fi elds are important indicators of their maturation and
standing. Examples are the recent article on eyewitness testimony in the
Annual Review of Psychology 1 (Wells & Olson, 2003 ) and the Psychology and
Law section in the Encyclopedia of Applied Psychology : 13 articles on psychol-
ogy and law in general and two on eyewitness memory and identifi cation in
particular (Spielberg, 2004 ). More recent publications, the Encyclopedia of
Psychology & Law (Cutler, 2007 ) and the two-volume Handbook of Eyewitness
Psychology (Toglia, Read, Ross, & Lindsay, 2006; Lindsay, Ross, Read, &
Toglia, 2007 ) provide up-to-date treatments of many facets of the fi eld.
In many areas of eyewitness identifi cation, general agreement on a given
phenomenon is bolstered via the fi ndings of meta-analytic reviews of the
research literature. A meta-analysis is a synthesis of all obtainable data
collected in a specifi ed topical area. The benefi ts of a meta-analysis are
that greater statistical power can be obtained by combining data from many
studies. Hypotheses not originally specifi ed in the research, such as publica-
tion bias or gender effects, can be examined. And fi nally, directions for
future research can be reported when applicable. Like most scientifi c
research, meta-analyses are subject to peer review prior to publication.
Important examples of such analyses include topics such as the cross-race
effect (Meissner & Brigham, 2001 ), the confi dence–accuracy relationship
(Bothwell, Deffenbacher, & Brigham, 1987 ; Cutler & Penrod, 1989 ; Sporer,
Penrod, Read, & Cutler, 1995 ), factors infl uencing face identifi cation (Shapiro
& Penrod, 1986 ), the description–identifi cation relationship (Meissner &
Brigham, 2001 ; Meissner, Sporer, & Susa, 2008 ), the effects of sequential
versus simultaneous lineups and showups (Steblay, Dysart, Fulero, & Lindsay,
2001 ; McQuiston-Surrett, Malpass, & Tredoux, 2006 ), the infl uence of lineup
instructions (Steblay, 1997 ; Clark, 2005 ), the mug-shot exposure effect
Cutler Chapters.indd 15Cutler Chapters.indd 15 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
16 Expert Testimony on the Psychology of Eyewitness Identifi cation
(Deffenbacher, Bornstein, & Penrod, 2006 ), the weapon-focus effect (Steblay,
1992 ), the infl uence of emotion/anxiety (Deffenbacher, Bornstein, Penrod, &
McGorty, 2004 ), the effects of hypnosis on recall (Steblay & Bothwell, 1994 ),
and post-identifi cation feedback (Douglass & Steblay, 2006 ).
Representation in Introductory Textbooks
Introductory textbooks are the most concise and authoritative presentations
of the core concepts and fi ndings of a fi eld. They are aggressively peer-
reviewed and represent a range of professionally acceptable emphases or ver-
sions of the nature of the fi eld. Eyewitness research concepts are represented
in psychology textbooks at a high rate. Hosch and associates (Chapter 7)
provide a thorough evaluation of the extent to which phenomena related to
eyewitness processes are presented in commonly adopted textbooks across
various areas of psychology. In short, they found that almost every textbook
contained a presentation of research related to eyewitness issues, with some
devoting entire chapters to the discussion of these issues.
Consensus Panels of Scientifi c Experts
Groups of experts are frequently brought together to assess the foundations of
a given area of scientifi c inquiry. Such consensus panels have been assembled
in the fi eld of eyewitness psychology over the past decade. For example, the
AP-LS sponsored a review of the scientifi c work on eyewitness identifi cation
with the purpose of proposing changes in law enforcement policy. This “white
paper,” entitled Eyewitness Identifi cation Procedures: Recommendations for
Lineups and Photospreads , was authored by Wells and colleagues ( 1998 ). It
represented the study of eyewitness identifi cation as a stable body of research
as of that date, and was authored by six prominent scientists in the fi eld.
A second consensus panel document on eyewitness identifi cation research
involved the publication of Eyewitness Evidence: A Guide for Law Enforcement
by the Technical Working Group on Eyewitness Evidence, under the direc-
tion of the Attorney General of the United States (see Wells et al., 2000 ). This
Technical Working Group consisted of a select group of psychological
researchers, attorneys, and representatives of law enforcement, and its fi nd-
ings and recommendations were published in 1999, with a training manual
following in 2003. The guide and training manual were disseminated to all
law enforcement agencies in the United States.
Surveys of Scientifi c Experts
Periodically, researchers and experts in the eyewitness fi eld are administered
questionnaires to determine their level of agreement on specifi c eyewitness
topics. The results of these questionnaires indicate that a high level of consen-
sus exists regarding those issues relevant to the fi eld (Kassin, Ellsworth, &
Cutler Chapters.indd 16Cutler Chapters.indd 16 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 17
Smith, 1989 ; Kassin, Tubb, Hosch, & Memon, 2001 ). For example, the most
recent survey conducted by Kassin and colleagues indicated signifi cant agree-
ment (over 90 % of experts surveyed) on the effects of cross-racial identifi ca-
tion, alcohol intoxication, hypnotic suggestibility, attitudes and expectations,
child suggestibility, post-event information, mug-shot exposure, confi dence
malleability, lineup instructions, and leading interview questions. This matter
is reviewed at length by Hosch and colleagues later in this volume.
What Can Research Tell Us About the Factors That Infl uence
Eyewitness Identifi cation?
As described earlier, psychologists have developed a scientifi c understanding
of factors that infl uence human perception, memory, and judgment that are
relevant to the evaluation of eyewitness evidence. Numerous reviews are avail-
able to the interested reader that document, in detail, the specifi c fi ndings and
scientifi c studies that serve as the scientifi c basis of this literature (see Brewer
& Williams, 2005 ; Lindsay et al., 2007 ; Neuschatz & Cutler, 2008 ; Toglia et al.,
2006). This corpus of research has been published in many of the most highly
regarded, peer-reviewed journals in the fi elds of experimental, cognitive,
social, developmental, and applied psychology. Although it is not our charge
to systematically review the scientifi c research on eyewitness identifi cation, we
review more generally the basic scientifi c knowledge evidenced by these
studies.
Following a timeline beginning with a criminal event, and leading to the
investigation of this crime and, ultimately, to the trial of an accused perpetra-
tor, accurate or inaccurate identifi cations derive principally from the
following:
• The cognitive structures (encoding capabilities) , fears , or values that a
witness brings to the witnessing situation. Witnesses who are cognitively
impaired through their affective response to a situation, or those
distracted from observing relevant information via a focus on other
salient aspects of the event (e.g., the weapon used to threaten them),
are less likely to render an accurate identifi cation. Certain individuals
who are cognitively impaired or those with limited cognitive capability
either through alcohol/drug inducement or more natural
developmental processes are similarly less accurate in their
identifi cations.
• The opportunities and constraints afforded the witness at the witnessing
situation. The greater the opportunity a witness has to observe an
event or a perpetrator, and the clearer the resolution and acuity
associated with this observation, the more accurate the witness’s
memory is likely to be. Factors such as distance, lighting, and time to
view have all been shown to infl uence the accuracy of witness
identifi cation.
Cutler Chapters.indd 17Cutler Chapters.indd 17 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
18 Expert Testimony on the Psychology of Eyewitness Identifi cation
• The information a witness encodes about the offender from the initial
viewing of the event. As we’ve noted earlier, the constraints on the
information a witness encodes into memory depend upon both
attributes of the person and attributes of the environment. But
whatever the constraints on encoding, the information obtained at
the original event is the base from which all subsequent memory
(recall, recognition, identifi cation) processes work. If the initial
encoding is weak, then subsequent memory-contaminating events
are likely to have greater effects than if the original memory is
strong.
• The history of the encoded information from the initial viewing to the
time of the identifi cation request. Witnesses are susceptible to the
infl uence of both suggestive questioning by investigators, as well as to
social information that may be shared by other witnesses. They may be
exposed to images of other persons or faces represented as being the
offender in the media or as part of the investigation process. In
addition, witnesses may forget information as the time between their
viewing of the event and their attempt at an identifi cation increases.
• The circumstances surrounding the identifi cation (including the
identifi cation procedures , the quality of the lineup that is presented , the
witness’s social motives and values , etc.). Suggestive lineup identifi cation
procedures can lead to misidentifi cation, including the exposure of a
witness to multiple images of the same target person, poorly
constructed lineups in which the target person is perceptually salient
to the witness, and lineup instructions that induce choosing on the
part of the witness.
• The witness’s expectations about testimony and the task of giving
evidence , including information derived from interactions with law
enforcement or other witnesses. Witnesses may become overconfi dent in
their testimony in the courtroom if they are provided positive
feedback at the time of their identifi cation or if they subsequently draw
conclusions about their likely accuracy based on prosecution of the
identifi ed person.
The accuracy of a witness’s identifi cation(s) and identifi cation testimony
is a function of all these factors. Many of these are attributes of human per-
ception, memory, and judgment processes, and these psychological processes
have been subject to both general models of cognition (see Lane & Meissner,
2008 ; Turtle, Read, Lindsay, & Brimacombe, 2008 ) and more specifi c models
of eyewitness identifi cation performance (Clark, 2008 ). Although this corpus
of research will continue to mature and expand its knowledge base, it is with-
out controversy that our current scientifi c understanding of eyewitness mem-
ory is beyond the ken of lay and customary knowledge (see Read & Desmarais,
Chapter 6). It is critical, then, that we seek to transfer this knowledge to those
who rely upon eyewitness evidence and make decisions within our criminal
justice system.
Cutler Chapters.indd 18Cutler Chapters.indd 18 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 19
Knowledge Transfer
The transfer of scientifi c knowledge from research laboratories to criminal
justice practitioners can be accomplished in many ways. Legislative action, as
well as training for law enforcement investigators, attorneys (in law school
and beyond), and judges are all mechanisms for changing the criminal justice
system. But any training regimen has its problems. For example, for attorneys
and judges, the path to the law is not often through the sciences, a fact that
creates a training-readiness question. Similarly, for law enforcement investi-
gators, educational standards are uneven, frequently low, and infrequently
based in science. Legal reform, in the sense of requiring new standards of evi-
dence and standards of oversight of investigation processes, are cumbersome
and have low levels of probable success.
Training for law enforcement personnel has an effect more proximal to
the source of errors leading to wrongful conviction, and this seems to be a
promising knowledge transfer path. However, law enforcement as a societal
institution is severely fractionated in the United States (thousands of indi-
vidual and overlapping jurisdictions), as are training standards and facilities.
Budgets for training are commonly thin, and professionalism and educational
standards vary widely across individual law enforcement units. Law enforce-
ment has very limited research and development capabilities, especially in the
areas of the social and behavioral sciences that focus on such a large range of
issues in investigation, use of witnesses, and other areas of legal decision mak-
ing (Geller, 1997 ). The contributions of university-based research and educa-
tional programs, therefore, are correspondingly more important, and the few
genuinely collaborative research programs between law enforcement and uni-
versity-based researchers have unique value.
Expert testimony, provided to the fi nders of fact during trial, is another
knowledge transfer point. It is particularistic in the sense that it is applied one
case at a time. However, this volume and the cumulative base of scientifi c
study of eyewitness identifi cation exemplify the fact that, as in the laboratory,
evidence gathered one instance at a time can be productively examined as a
means of understanding a phenomenon. Expert testimony has become a fi eld
for scientifi c scrutiny while at the same time it confronts the problems associ-
ated with making a case for scientifi c knowledge to persons who overwhelm-
ingly bring customary knowledge to the courthouse. The many questions of
its importance and effectiveness, as well as the size and location of discrepan-
cies between scientifi c and customary knowledge, are the subject of detailed
treatment in the remaining chapters of this volume.
Conceptualizations of Expert Testimony
Ultimate Opinion Testimony
Offering an opinion about whether a witness is correct or incorrect in
his identifi cation of a defendant is a form of ultimate opinion testimony .
Cutler Chapters.indd 19Cutler Chapters.indd 19 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
20 Expert Testimony on the Psychology of Eyewitness Identifi cation
Ultimate opinion testimony is beyond our reach for a number of reasons,
beginning with the fact that no one can retrieve precise information on the
witnesses’ personal state, attentional deployment, memory capability, and
similar factors that were present at the time of the offense. The science, based
on controlled studies, tells us that these factors have effects on the encoding of
information into memory, but we cannot retrospectively measure the degree
to which they were present at any particular time. We can advise the fi nders of
fact about indicators of the strength of these factors, but we ourselves cannot
form a science-based opinion of the degree of their presence in a given case.
However, we can say that, for example, when a weapon was visible in a crime,
controlled studies show that effects occur on person recognition and evidence
suggests that these effects result from attention distraction resulting from the
presence of the weapon. We cannot quantify the degree of attention distrac-
tion; we can, however, inform the fi nder of fact that the presence of a weapon
is known to reduce identifi cation accuracy, through the mechanism of atten-
tion deployment. This, of course, is only one of many factors that may poten-
tially interfere with the process of acquiring information about the appearance
of the offender at the crime scene, and this imposes a limit on the quality of
the subsequent memory of the offender’s image for use in other processes
(e.g., judgment of whether a photograph is or is not the same person who is in
their memory). Obviously, if his image is not well established in the witness’
memory, the later identifi cation task will be more diffi cult for the witness and
the result will be less reliable as evidence. These are things we can share with
the fi nders of fact.
Likewise, we may know what memory-contaminating events may have
occurred in the witness’ experience, such as making and subsequently study-
ing a composite image intended to represent the offender. But we cannot
quantify exactly the magnitude of the effect that this would have on the wit-
ness. We can come very close to specifying the average of a group of people
exposed to a similar condition, but it is a characteristic of any science based in
statistical analysis and statements of group averages that individual observa-
tions (in this case, responses of individual research participants) vary from the
group average in a direction and to a degree that cannot be specifi ed for an
individual, whether they are in the laboratory or in the witness box. We know
the average effect, and we can share that information with the fi nders of fact.
But we stop short of ultimate opinion testimony because the specifi c quantity
of events that we and the fi nders of fact are certain to have occurred cannot be
quantitatively specifi ed on an individual basis with great precision beyond
that of the group average.
Similarly, we cannot know exactly what the witness brings to the identifi -
cation task: what self instructions were present, how the witness values
accurate identifi cation of offenders in relation to avoidance of false identifi ca-
tion of innocent persons (Malpass, 2006 ), how instructions were weighed,
and with what retrieval and identifi cation decision strategies the witness
approached the task. What does the witness believe is her task in the
Cutler Chapters.indd 20Cutler Chapters.indd 20 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 21
identifi cation procedure, apart from the instructions given, which may or
may not be believed? From early studies (Malpass & Devine, 1981 ), we know
that research participants mis-remember having been given biased lineup
instructions more than four times as frequently than mis-remembering hav-
ing been given unbiased instructions. Does the witness believe that arresting
someone who has a reasonably high probability of being the offender is of
primary importance? Does the witness consider the importance of not identi-
fying an innocent person and its consequences (that an innocent person may
be imprisoned, which results in the guilty person going unpunished and free
to commit more crimes)? Does the witness believe it is almost certain that the
police have caught the right person, and that the identifi cation is confi rma-
tory rather than independent information? We can determine, in many cases,
for example, what instructions were given by law enforcement, and we have a
very good idea of their effects (Clark, 2005 ; Steblay, 1997 ). We can often
determine how the identifi cation procedure was administered and whether
the lineup, for example, was fair (Malpass et al., 2007 ; Eyewitness Identifi cation
Research Laboratory, 2008 ), and we can inform the fi nders of fact about what
we know of the effects on identifi cation of what was, in fact, done. But again,
we can speak of the effects on research participants generally but not claim to
know exactly (quantitatively) how these factors affected the accuracy of a
specifi c witness’ identifi cation.
Expert As Educator
The evaluation of witness accuracy and the weight that eyewitness evidence is
to be given in arriving at a verdict are within the task of the fi nder(s) of fact,
and the responsibility of experts does not extend to offering conclusions.
Doing so is widely thought of as invading the province of the jury. However,
on the experts’ side of that line is plenty of room for other functions. Although
it is within the task of the fi nder of fact to make judgments on these matters,
the contribution our science can make is to alert them to the existence of
factors known to have effects on identifi cation, and to indicate how strong
those effects have been shown to be. In many cases, this will bring factors
known to be important to the attention of the fi nders of fact. They may not
have thought of these if not reminded of them, or they may not know of them
at all. The extent to which the fi nders of fact know what scientists know about
eyewitness identifi cation is discussed by Read and Desmarais later in this
volume (Chapter 6).
“Educator” is an obvious way to think about the role of an expert witness.
This role can be expressed in a range of knowledge-transfer contexts, as dis-
cussed earlier. In the context of the ultimate opinion testimony discussion, the
expert’s role as educator is explicit. Clearly, there are bits of factual knowledge
that it would be appropriate for the fi nders of fact to learn. These range from
the very concrete (e.g., how long a look at a face is suffi cient for a good mem-
ory image under good viewing conditions) to very abstract (e.g., that memory
Cutler Chapters.indd 21Cutler Chapters.indd 21 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
22 Expert Testimony on the Psychology of Eyewitness Identifi cation
involves integrative processes that take place outside of awareness). These are
part of an effort to educate the fi nders of fact and the court more generally
about what psychological science knows, and how it might be applied to their
evaluation of the eyewitness evidence in the case at hand. However, there are
integrative functions that the expert witness can perform that are not simple
transfers of knowledge about the effects of a list of factors.
Wagenaar ( 1988 ) suggests (p. 149) a Bayesian approach to the expert’s
role, and uncovers two quite different orientations to the expert’s testimony.
The fi rst is the estimation of the probability of guilt given the evidence. This
framing of expert testimony is perilously close to giving an ultimate opinion
about guilt or innocence, given the evidence in the case. The second approach
reveals a very different and interesting use of an eyewitness scientist’s exper-
tise: to provide an estimate of the probability of the evidence given innocence.
This is based on the principle of diagnosticity: if the probability of the evidence
given innocence is as high as the probability of the evidence given guilt then
the evidence is not diagnostic either of guilt or of innocence, and guilt and
innocence cannot be distinguished based on the evidence. The evidence is
diagnostic of guilt only when its probability given innocence is substantially
lower than its probability given guilt. For example, if, prior to making an
identifi cation, the witness who saw the offender briefl y under poor conditions
viewed the defendant on television as someone being sought in the case, saw
the defendant in the police station prior to the identifi cation procedure, saw
the defendant’s photo in a mug book, saw a lineup in which no other photos
resembled either the description given of the offender or the defendant and,
without further admonitions or instructions, was told that the police have
caught the offender and need the witness to pick him out of a lineup, then one
can argue that the witness would not need to have seen the defendant commit
the crime in order to “remember” his face and choose to select him from a
lineup. The probability of the lineup choice is as likely to result from the
events prior to the identifi cation procedure as it is from having seen the defen-
dant commit the crime — perhaps even more likely because, as a result of those
events, the witness is likely to have developed a strong memory image of the
defendant. Thus, the suggestions in the administration of the identifi cation
process come perilously close to forcing an identifi cation of someone from the
lineup: the most familiar and distinctive person. This use of a scientist’s exper-
tise goes beyond a recitation of factors affecting identifi cation, and allows for
an integrative approach to informing the fi nders of fact about the scientifi c
evaluation of the evidence without offering an ultimate opinion about the
accuracy of the witness.
Conclusion
It is an extremely unusual law enforcement organization that has research
|and evaluation components that work on improving the process of evidence
Cutler Chapters.indd 22Cutler Chapters.indd 22 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 23
collection and preservation. The techniques used are to a large extent
“customary” in that they have been developed over a period of time by law
enforcement offi cers, without benefi t of scientifi c consultation. This is less
true of scientifi c procedures that are, without question, inaccessible to
persons without specifi c training.
DNA is a good example. Biological materials are harvested by trained
technicians and brought to other, more specifi cally trained and certifi ed tech-
nicians who analyze the materials by procedures developed by scientists and
approved by the courts. Eyewitness evidence, on the other hand, is collected
by people who are not trained as technicians, and the material harvested is not
delivered for evaluation to certifi ed technicians who use techniques that have
been developed by scientists and approved by the courts. In contrast, reports
based on eyewitness evidence are not constructed by specialists qualifi ed to
evaluate the signifi cance of the fi ndings of the identifi cation procedure, and
they are not contributed to the evidence fi le, as they are for even the most
troublesome areas of forensic evidence, such as bite marks or hair samples.
Procedures that are customary, and that have been developed by non-
technicians, non-scientists, without a research base, and not calibrated to
continuing scientifi c fi ndings have not met the burden of showing that the
procedures used to develop the evidence being offered to the court are valid.
When a scientifi c fi eld of study 100 years old develops theories and fi ndings
that have only rarely been used to inform evidence gathering, preservation,
and interpretation processes in law enforcement, it is improper to grant
implicit validity to customary procedures without consideration of the corpus
of scientifi c work, or its absence.
Because of the great power of systematic study that it would bring, there
is almost no rational way to argue that scientifi c approaches to the evaluation
of eyewitness identifi cation procedures used in investigations would not be
helpful. And, in the face of decades of documentation of wrongful conviction
based on erroneous eyewitness testimony, it is scandalous that law enforce-
ment has not taken on this problem itself. Although some U.S. jurisdictions
have attempted to implement modifi cations in identifi cation procedures
based on scientifi c evidence, the evidence on which such change was based
was developed in academic research laboratories, not as the product of law
enforcement agency studies of the effectiveness of their own procedures. We
grant that the severe fractionation of law enforcement and legal jurisdictions
in the United States works against the development of an effective research
and development arm of law enforcement. When we look to nations where
law enforcement is far more integrated and centralized, as in the United
Kingdom, we fi nd that law enforcement does in fact undertake research and
development activities with regard to eyewitness identifi cation (and many
other topics), and they work closely with members of the academic commu-
nity and with members of law enforcement who have scientifi c training. With
fractionation in the United States, much of the research and development
activity has fallen to academic researchers who attempt to fi ll the void.
Cutler Chapters.indd 23Cutler Chapters.indd 23 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
24 Expert Testimony on the Psychology of Eyewitness Identifi cation
It is unfortunate that so few collaborative projects exist. We acknowledge
that since the advent of DNA exonerations more collaborative projects have
sprung up, but these too are fractionated, and the projects are not integrated
into the everyday practice of evidence collection on eyewitness identifi cation
and systematic evaluation as an ongoing activity. It would be benefi cial if the
task of doing the research could be placed inside the police station.
From our position outside of the customs and traditions of law enforce-
ment and the courts, we are not bound to accept precedent as important. Our
premise is that the legal system is wrong and wrong-headed about the role of
scientifi c study of eyewitness identifi cation in the criminal justice system. If
eyewitness research is rightfully criticized for taking place outside of it, that is
because law enforcement agencies and the courts have not done the work
themselves. No effective research and development organization embedded
within the fractionated criminal justice system in the United States has evalu-
ated the questions addressed by academic research. The way in which law
enforcement holds onto customary methods and procedures (i.e., eyewitness
identifi cation, interviewing and interrogation, detection of deception) is
based on a commitment to customary knowledge and a rejection of scientifi c
knowledge. Rather than bring scientifi c knowledge into the house and work
with it to evaluate and improve techniques that are demonstrably fl awed, the
U.S. criminal justice system has abrogated this work to academic researchers
instead.
Without these activities becoming a routine part of law enforcement,
there remain other mechanisms for making the knowledge gained through
psychological science, and administered from outside the criminal justice sys-
tem, available to law enforcement and the courts. Expert testimony is but one
of these, and it is the mechanism examined in this book.
References
Borchard, E. M. (1932). Convicting the innocent . New Haven: Yale University Press.
Bothwell, R. K., Deffenbacher, K. A., Brigham, J. C. (1987). Correlation of
eyewitness accuracy and confi dence: Optimality hypothesis revisited. Journal
of Applied Psychology , 72 , 691–695.
Brandon, R., & Davies, C. (1973). Wrongful imprisonment: Mistaken convictions
and their consequences. Hamden, CT: Archon.
Brewer, N., & Williams, K. D. (Eds.). (2005). Psychology and law: An empirical
perspective . New York: The Guilford Press.
Clark, S. E. (2005). A re-examination of the effects of biased lineup instructions in
eyewitness identifi cation. Law and Human Behavior , 29 , 395–424.
Clark, S. E. (2008). The importance (and necessity) of computational modeling
for eyewitness identifi cation research. Applied Cognitive Psychology , 22 ,
803–813.
Clifford, B.R. (1997). A commentary on Ebbesen and Konečni’s “Eyewitness
memory research: Probative v. prejudicial value.” Expert Evidence , 5 , 140–143.
Cutler Chapters.indd 24Cutler Chapters.indd 24 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
The Need for Expert Testimony 25
Connors, E., Lundregan, T., Miller, N., & McEwen, T. (1996). Convicted by juries,
exonerated by science: Case studies in the use of DNA evidence to establish
innocence after trial. Washington, DC: National Institute of Justice.
Cutler, B. L. (Ed.) (2007). Encyclopedia of Psychology & Law . Sage Publications.
Cutler, B. L., & Penrod, S. D. (1989). Forensically relevant moderators of the
relation between eyewitness identifi cation accuracy and confi dence. Journal of
Applied Psychology , 74 , 650–652.
Deffenbacher, K. A., Bornstein, B. H., & Penrod, S. D. (2006). Mugshot exposure
effects: Retroactive interference, mugshot commitment, source confusion,
and unconscious transference. Law and Human Behavior , 30 , 287–307.
Deffenbacher, K. A., Bornstein, B. H., Penrod, S. D., & McGorty, E. K. (2004).
A meta-analytic review of the effects of high stress on eyewitness memory.
Law and Human Behavior , 28 , 687–706.
Devlin, Hon. Lord Patrick. (1976). Report to the Secretary of State for the Home
Department of the Departmental Committee on Evidence of Identifi cation in
Criminal Cases . London: Her Majesty’s Stationery Offi ce.
Douglass, A. B., & Steblay, N. (2006). Memory distortion in eyewitnesses:
A meta-analysis of the post-identifi cation feedback effect. Applied Cognitive
Psychology , 20 , 859–869.
Ebbesen, E. B., & Konečni, V. J. (1996). Eyewitness memory research: Probative
v. prejudicial value. Expert Evidence: The International Digest of Human
Behaviour Science and the Law , 5 , 2–28.
Elliott, R., (1993). Expert testimony about eyewitness identifi cation: A critique.
Law and Human Behavior , 17 , 423–437.
Eyewitness Identifi cation Research Laboratory (2008). Consultation and expert
testimony . Retrieved from http://eyewitness.utep.edu/consult01.html.
Geller, W. A. (1997). Suppose we were really serious about police departments
becoming ‘‘learning organizations’’? National Institute of Justice Journal , 234 , 2–8.
Hedges, L. V. (1987). How hard is hard science, how soft is soft science?: The
empirical cumulativeness of research. American Psychologist , 42(2) , 443–455.
Huff, C. R., Rattner, A., & Sagarin, E. (1996). Convicted but innocent: Wrongful
conviction and public policy . Thousand Oaks: Sage.
Innocence Project. (2008). Understand the causes: Eyewitness misidentifi cation .
Retrieved November 24, 2008 from http://www.innocenceproject.org/
understand/Eyewitness-Misidentifi cation.php.
Kassin, S. M., Ellsworth, P. C., & Smith, V. L. (1989). The “general acceptance” of
psychological research on eyewitness testimony. American Psychologist , 44 ,
1089–1098.
Kassin, S. M., Tubb, V. A., Hosch, H. M., & Memon, A. (2001). On the “general
acceptance” of eyewitness testimony research: A new survey of the experts.
American Psychologist , 56 , 405–416.
Lane, S. M., & Meissner, C. A. (2008). A ‘middle road’ approach to bridging the
basic-applied gap in eyewitness identifi cation research. Applied Cognitive
Psychology , 22 , 779–787.
Lindsay, R. C. L., Ross, D. F., Read, J. D., & Toglia, M. P. (Eds.). (2007). The
handbook of eyewitness psychology, Vol. II: Memory for people . Mahwah, NJ:
Lawrence Erlbaum.
Loftus, E. F. (1979). Eyewitness reliability. Science , 205 , 386–387.
Loftus, E. F. (1983). Silence is not golden. American Psychologist , 38 , 564–572.
Cutler Chapters.indd 25Cutler Chapters.indd 25 4/30/2009 8:21:54 PM4/30/2009 8:21:54 PM
26 Expert Testimony on the Psychology of Eyewitness Identifi cation
Malpass, R.S. (1999). Subjective culture and the law. In J. Adamopoulos & Y.
Kashima (Eds.) Social Psychology and Cultural Context . Newberry Park, CA:
Sage Publications.
Malpass, R. S. (2006). A policy evaluation of simultaneous and sequential lineups.
Psychology, Public Policy and Law , 12(4) , 394–418.
Malpass, R. S., & Devine, P. G. (1981). Eyewitness identifi cation: Lineup
instructions and the absence of the offender. Journal of Applied Psychology , 66 ,
482–489.
Malpass, R. S., & Kravitz, J. (1969). Recognition for faces of own and other race.
Journal of Personality and Social Psychology , 13 , 330–334.
Malpass, R. S., Tredoux, C. G., & McQuiston-Surrett, D. E. (2007). Lineup
construction and lineup fairness. In R. C. L. Lindsay, D. F. Ross, J. D. Read, &
M. P. Toglia (Eds.), The handbook of eyewitness psychology, Vol. II: Memory for
people . (pp. 87–100). Mahwah, NJ: Lawrence Erlbaum.
McCloskey, M., & Egeth, H. E. (1983). Eyewitness identifi cation: What can a
psychologist tell a jury? American Psychologist , 38 , 550–563.
McCullough, M. L. (2002). Do not discount lay opinion. American Psychologist , 57 ,
376–377.
McQuiston-Surrett, D., Malpass, R. S., & Tredoux, C. G. (2006). Sequential vs.
simultaneous lineups: A review of methods, data, and theory. Psychology,
Public Policy, and Law , 12 , 137–169.
Meissner, C. A., Sporer, S. L., & Susa, K. J. (2008). A theoretical review and meta-
analysis of the description-identifi cation relationship in memory for faces.
European Journal of Cognitive Psychology , 20(3) , 414–455.
Meissner, C. M., & Brigham, J. C. (2001). A meta-analysis of the verbal
overshadowing effect in face identifi cation. Applied Cognitive Psychology , 15 ,
603–616.
Neuschatz, J. S., & Cutler, B. L. (2008). Eyewitness identifi cation. In H. L.
Roediger, III (Ed.), Cognitive psychology of memory (Vol. 2 of Learning and
memory: A comprehensive reference , J. Byrne, Series Ed., pp. 845–865 ) . Oxford:
Elsevier.
Police and Criminal Evidence Act 1984 (s.60(1)(a)) Code of Practice (D) on
identifi cation of persons. Section 2: Identifi cation by witnesses. London:
Home Offi ce Directorate.
R v. Turnbull , 63 Cr. App. R. 132 (1976).
Radelet, M. L., Bedau, H. A., & Putnam, C. E. (1992). In spite of innocence:
Erroneous convictions in capital cases . Boston: Northeastern University Press.
Shapiro, P., & Penrod, S. (1986). Meta-analysis of facial identifi cation studies.
Psychological Bulletin , 100 , 139–156.
Spielberg, C. D. (Ed.). (2004). Encyclopedia of applied psychology . New York:
Elsevier Academic Press.
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.
Steblay, N. (1997). Social infl uence in eyewitness recall: A meta-analytic review of
lineup instruction effects. Law and Human Behavior , 21 , 283–298.
Steblay, N. Dysart, J., Fulero, S. Lindsay, R. C. L. (2003). Eyewitness accuracy rates
in police showup and lineup presentations: A meta-analytic comparison. Law
and Human Behavior , 27 , 523–540.
Cutler Chapters.indd 26Cutler Chapters.indd 26 4/30/2009 8:21:55 PM4/30/2009 8:21:55 PM
