New Perspectives in Policing

augus t 2014
VE

RI

TAS

HARVARD Kennedy School
Program in Criminal Justice
Policy and Management

National Institute of Justice

Policing and Wrongful Convictions
Anthony W. Batts, Maddy deLone and Darrel W. Stephens

Introduction

Executive Session on Policing and
Public Safety

Few events subject the criminal justice system to

This is one in a series of papers that will be published as a result of the Executive Session on
Policing and Public Safety.

officials, the media and the general public as the

Harvard’s Executive Sessions are a convening
of individuals of independent standing who take
joint responsibility for rethinking and improving
society’s responses to an issue. Members are
selected based on their experiences, their reputation for thoughtfulness and their potential for
helping to disseminate the work of the Session.
In the early 1980s, an Executive Session on Policing
helped resolve many law enforcement issues of
the day. It produced a number of papers and
concepts that revolutionized policing. Thirty years
later, law enforcement has changed and NIJ and
the Harvard Kennedy School are again collaborating to help resolve law enforcement issues of
the day.
Learn more about the Executive Session on
Policing and Public Safety at:
www.NIJ.gov, keywords “Executive Session
Policing”
www.hks.harvard.edu, keywords “Executive
Session Policing”

as intense scrutiny from policymakers, elected
exoneration of a wrongfully convicted defendant. The multimillion-dollar settlement recently
announced by New York City in the notorious
case of the Central Park Five has once again
brought the injustice of wrongful convictions
and the corollary injustice of failing to convict
the real assailant to the forefront of the national
consciousness. (In that case, five black and Latino
teenagers were convicted on the basis of false
confessions of the brutal — and widely publicized —
rape of a young woman jogging in Central Park.
All were tried as adults and sentenced to lengthy
prison terms, then exonerated years later when
the real perpetrator confessed to the crime — a
confession corroborated by DNA evidence.) As
this and other high-profile wrongful convictions
continue to spark controversy, a dispassionate,
thoughtful examination of the systemic causes
of wrongful convictions and their potential solutions can benefit all components of the criminal
justice system and the community at large.

2 | New Perspectives in Policing

Given their function as gatekeepers to the crimi-

assist with post-conviction investigations and

nal justice system and as the initial point of

work with other stakeholders to uncover past

contact with offenders, accused offenders, vic-

errors. The conclusion explains why adopting

tims and ordinary citizens, police agencies are

these recommendations will provide both justice

in the most advantageous position to conduct

and public safety advantages: (1) they will mini-

such an examination. This paper calls for strong

mize the number of innocent individuals who

leadership from police agencies to lead reviews of

are wrongly convicted and (2) they will prevent

wrongful convictions that can be learning experi-

police from overlooking the actual perpetrator

ences for all components of the criminal justice

early in an investigation because of misleading

system. The lessons learned from these reviews

evidence that points to the wrong person, thereby

can lead to the implementation of changes in

increasing the number of perpetrators correctly

practice across criminal justice agencies that

identified in the initial stage.

will improve justice without compromising safety.
The recommendations are comprehensive, farSection I reviews what has been learned from

reaching, and necessitate strong leadership from

studying exonerations over the past 20 years,

police agencies, both to implement and to pro-

including who the exonerees are, the factors that

vide the guidance required for the entire criminal

contribute to wrongful convictions, and the best

justice community to adopt a more systemwide

practices that address these factors. Section II

approach to learning from wrongful convictions

looks at the role cognitive biases play in wrongful

and to improve the system as a whole.

convictions. Section III presents several recomthe likelihood of wrongful convictions and that

I. What Have We Learned From
Exonerations?

are important for police departments to adopt.

After getting into a car accident with his neigh-

These recommendations include best practices

bor on November 21, 1981, Henry James, then 20

that address known contributing factors, includ-

years old, went to inform his neighbor’s wife that

ing eyewitness identification, false confessions,

her husband had been arrested. The next morning

informant testimony and evidence preserva-

at 6 a.m., the neighbor’s wife was awakened by an

tion. This section also suggests an approach for

intruder who raped her at knifepoint and then left

learning from error and viewing mistakes as

the home. She immediately called the police and

systemwide weaknesses. Section IV provides

told them that the perpetrator was a black male

suggestions for how police can cooperate and

whom she did not know.

mendations that research shows will minimize

Cite this paper as: Batts, Anthony, Maddy deLone and Darrel Stephens,
Policing and Wrongful Convictions. New Perspectives in Policing Bulletin.
Washington, D.C.: U.S. Department of Justice, National Institute of Justice,
2014. NCJ 246328.

During a search of the community the following day,
an officer saw James and thought he fit the description of the perpetrator. The investigator on the case
placed James’ picture in a photo book with 75 other

Policing and Wrongful Convictions | 3

black males, and the victim identified James from his

progresses, it is inevitable that many more wrong-

photo. James was later placed in a live lineup with

ful convictions will be uncovered.

individuals who failed to match the description provided by the victim and was again identified by the

Some ask how many innocent people have been

victim. Biological evidence was taken from both the

convicted. The truth is that we will never know, but

victim and James, and no matches were confirmed.

academics have looked at the question in discrete

At trial, James’ stepfather testified that James was

crime categories. In a study that reviewed defen-

sleeping in the bed beside him on the morning of the

dants sentenced to death for murders involving

attack and woke up for work at 6:30 a.m. Another

rape between 1982 and 1989, Risinger (2008) esti-

witness testified that he saw James at a gas station

mated that between 3 and 5 percent were innocent.

near the time the crime occurred and gave him a ride

Gross (2008) also found that 2.5 percent of all pris-

to work. Despite the overwhelming evidence pointing

oners sentenced to death between 1973 and 1989

to his innocence, James was convicted of rape and

were eventually exonerated. According to the

sentenced to life in prison. He spent almost 30 years

Bureau of Justice Statistics (Carson and Golinelli,

in prison before DNA evidence excluded James as the

2013), approximately 1.57 million prisoners were

perpetrator and he was exonerated.

under the jurisdiction of state and federal correctional authorities at year-end 2012. If as few as 1

Henry James’ case is not an isolated event in the

percent are innocent (a low estimate in light of the

criminal justice system, and the hundreds of

aforementioned studies), that would mean 15,700

wrongly convicted individuals released from prison

innocent men and women are being kept behind

over the past 20 years provide indisputable proof

bars. Regardless of how many there are, prevent-

that wrongful convictions can and do occur. Since

ing additional cases should be a public safety and

1989, there have been over 1,300 exonerations in the

justice priority.

United States — more than 300 through newly discovered DNA evidence (Innocence Project, 2014a)

The wrongly convicted are not a homogeneous

and about 1,000 additional exonerations through

group. A 2012 study at Michigan Law School and

other types of evidence (National Registry of

Northwestern Law School (Gross and Shaffer, 2012)

Exonerations, 2014). This number is not insignifi-

reviewed 873 exonerations and found that a major-

cant; however, what is even more important is that

ity of the exonerees were men, 50 percent were black,

these are only the number of wrongly convicted

38 percent were white, 11 percent were Hispanic

persons that we know about. Because of lim-

and 2 percent were Native American or Asian. The

ited available evidence, a large backlog of actual

exonerations studied span the United States, with

innocence cases, and a restricted number of indi-

large concentrations in Illinois, New York, Texas

viduals who can allocate sufficient time to look into

and California. This distribution is likely shaped

these claims, we are discovering and examining

by the uneven level of effort across states to exam-

only some of the wrongful convictions. As work

ine wrongful convictions as well as differences
in underlying rates of crime and convictions that

4 | New Perspectives in Policing

might distinguish more urban states. The types

an organizational accident model and testing ini-

of crimes that individuals are falsely accused of

tial assumptions in an investigation. Finally, we

committing also vary. Most of the exonerations

provide suggestions on how police can assist in

involve rape and murder cases, although about 20

post-conviction investigations.

percent involve robberies or other crimes (Gross
and Shaffer, 2012).

Adopting these recommendations and developing a more rigorous process for investigations will

Although exonerees are diverse in many respects

provide several benefits. First, they will minimize

and their individual stories need to be high-

the number of innocent individuals who are

lighted, we have learned much from studying the

wrongly convicted. Second, as many wrongful

collection of cases as a whole. Misidentification

conviction cases illustrate, when an individual is

by eyewitnesses, false confessions, erroneous

erroneously convicted of a crime, the true perpe-

scientific evidence, untrue informant testimony,

trator often goes unpunished and commits future

overzealous prosecution, inadequate defense and

crimes (Acker, 2013). Out of the 132 actual perpe-

sometimes intentional misconduct contribute to

trators who were identified in the first 300 DNA

the conviction of the innocent. We believe that

exoneration cases, 65 had committed 137 addi-

most actors in the criminal justice system do

tional violent crimes — 74 rapes, 33 murders and

their jobs with integrity and have every interest in

30 other violent crimes (Innocence Project, 2013).

protecting the innocent. However, it is clear that
inadvertent errors committed by any (and all)

Moreover, it is well known that the length of time

individuals in the system eventually and directly

it takes to conduct an investigation and the prob-

contribute to wrongful convictions.

ability of finding the perpetrator are inversely
correlated: the longer the actual perpetrator goes

In recent years, research has led to several rec-

undetected, the less likely it becomes he or she

ommendations that can serve as checks on the

will ever be identified and apprehended. Thus,

system to prevent unnecessary miscues before

police will derive a public safety advantage from

an individual is falsely accused and to correct

adopting these recommendations, as they will

wrongs in the event they do occur. This paper

reduce the likelihood of being distracted from

focuses on the following questions: What poli-

pursuing the real perpetrator because of mis-

cies should police departments adopt to help

leading evidence and instead will help identify

prevent wrongful convictions, and how can

the actual perpetrator more quickly during the

police best assist in post-conviction investiga-

initial investigation.

tions? Specifically, we identify recommended
procedures for eyewitness identification, record-

We acknowledge that some of these recommen-

ing of interrogations, informant testimony, and

dations will be challenging to implement and will

evidence storage and preservation. We then look

need the entire community’s support to be fully

at the importance of learning from error through

effective. We also acknowledge that wrongful

Policing and Wrongful Convictions | 5

convictions occur as a result of systemwide errors,

do not (a concept psychologists call “confirma-

not simply mistakes by police officers. However,

tory bias”). Furthermore, scholars have shown

to stand by and do nothing at any level of the

that people are likely to hold fast to their own

criminal justice system is a dereliction of respon-

theories, even in the face of contradictory evi-

sibility and leadership.

dence (Simon, 2012).

We firmly believe that police departments are

It is common for people to experience this kind

the best advocates to catalyze this kind of change,

of “belief perseverance,” and it is often difficult for

and our reasoning is twofold. First, because they

them to realize their beliefs may be in error (Dror

are at the front end of the criminal justice process,

and Fraser-Mackenzie, 2009). This can happen

police officers are the initial point of contact in

at any stage of the criminal justice process —

an investigation, and their work affects the entire

from forensic scientists interpreting data in a

outcome of the case. If one piece of evidence is

manner consistent with law enforcement theo-

contaminated, it can often prejudice and con-

ries of guilt to prosecutors being so convinced of

taminate other evidence (this is known as the

a defendant’s guilt that they neglect to focus on

“snowball biasing effect,” which is discussed in

factors that contradict their theories — and it is

more detail in section II). Second, years of study

clear that every individual, regardless of his or

and research have produced highly specific, sci-

her level of responsibility or experience, is sus-

entifically based reforms to implement at the

ceptible to this phenomenon. The likelihood of a

police level. Best practices for other stages in the

wrongful conviction is further heightened by the

criminal justice system have not yet been empiri-

volume of arrests and the burden placed on the

cally tested and determined. Thus, it is logical for

criminal justice system as a whole. As it is cur-

these reforms and recommendations to be intro-

rently constructed, the criminal justice system

duced to the criminal justice system at the police

does not favor preventing or identifying wrongful

level.

convictions.

II. The Psychology Behind Cognitive
Biases and Their Relationship to
Wrongful Convictions

Humans are constantly flooded with sensory

Cognitive scientists have greatly helped us under-

everything quickly and efficiently (Findley and

stand wrongful convictions. It is now universal

Scott, 2006). To handle this overwhelming

knowledge that human perception is not objec-

amount of information, a host of shortcuts (heu-

tive. Our perceptions are heavily influenced by

ristics) have been developed that allow people

previous experiences and events, and humans

to make decisions rapidly without being over-

tend to remember the factors that fit with their

whelmed by unnecessary or redundant input.1

constructed mindset more clearly than those that

Although these shortcuts are generally helpful

information — we hear, read and see information from a variety of sources — and must process

and effective strategies for synthesizing and

6 | New Perspectives in Policing

processing complex information, they sometimes

humans do not (and cannot) process information

lead to faulty conclusions (Snook and Cullen,

on a blank slate. Individuals interpret informa-

2009). This system of categorization and selective

tion as a product of their beliefs, experiences,

attention makes humans subject to error.

expertise and understanding, a concept known
as top-down processing. It is a critical function

Tunnel vision, a heuristic-like model, is “the prod-

because humans have limited processing capaci-

uct of a variety of cognitive distortions that can

ties, and top-down processing helps manage the

impede accuracy in what we perceive and in how

huge amount of input received by the brain; how-

we interpret what we perceive” (Findley and Scott,

ever, it can also cause error because people often

2006, p. 307). Although a confluence of cognitive

do not realize they are combining new and old

biases is involved in tunnel vision, confirmatory

information (Dror, 2014). This lack of awareness,

and hindsight biases play the most prominent

along with a propensity to be overconfident about

roles.2 Confirmatory bias is the tendency to seek

one’s beliefs, can reinforce a premature conclu-

information and evidence that bolster existing

sion (Arkes et al., 1988; Findley and Scott, 2006).

expectations and hypotheses (Findley and Scott,
2006). It is usually not driven by conscious or

These cognitive biases are not limited to nov-

explicit errors, occurs almost automatically, and

ices and have been widely observed in experts

is most likely to appear when evidence is ambig-

in a variety of professions, including medicine,3

uous. In fact, research has shown that people

journalism,4 negotiating5 and psychotherapy.6

easily hold onto weak ideas and that providing

Dror explains that there is a “paradoxical nature

hypothetical explanations for an idea will bol-

of expertise, showing that with extraordinary

ster one’s belief in its likelihood (Simon, 2012).

abilities come vulnerabilities and pitfalls” (Dror,

Once a belief is in place, people tend to distort

2011, p. 184). Experts perform tasks and solve

new information to validate their expectations,

problems differently, which leads to high per-

increasing the value of confirming evidence and

formance in most situations. To achieve such

labeling disconfirming evidence as insignificant

high performance, experts need to have well-

or irrelevant.

organized knowledge; make sense of complex,
often ambiguous, signals and patterns; quickly

Hindsight bias refers to the tendency to think that

apply automaticity (reflexive skills that experts

an eventual outcome was much more likely to

can perform effortlessly); and perform challeng-

occur than one originally expected (Findley and

ing tasks that perplex novices (Dror, 2011). Experts

Scott, 2006). Memory is malleable and is greatly

also have highly trained information processing

affected by subsequent perceptions and expe-

mechanisms, with an acute ability to select the

riences. As a belief becomes more articulated

most pertinent information, group steps together

and comprehensive (and thus more persua-

into a single entity, and apply automaticity.

sive), it affects an individual’s recollection of
events (Stubbins and Stubbins, 2009). Moreover,

Policing and Wrongful Convictions | 7

Ironically, these linchpins of expertise can lead

prejudice and contaminate another line of evi-

to a lack of flexibility and to error. “Chunking”

dence. This is particularly relevant to police work

information makes the components less avail-

because it is usually the first point of contact in

able, if not inaccessible altogether, and fine-tuned

the criminal justice system and all later stages

selective attention also leads to the filtering of

feed off the information gathered in a police

potentially important information. For example,

investigation (Findley and Scott, 2006). Although

“chess experts are indeed better than novices in

these cognitive biases, to a certain degree, are

encoding and remembering board positions, but

inevitable in any case investigation, research

this is limited only to realistic board positions.

has shown that several best practices can help

Experts are no better, and are even worse than

prevent wrongful convictions and correct wrongs.

novices, in board positions in which constituting individual pieces are placed at random” (Dror,
2011, p. 181).

III. What Police Can Do To Help
Minimize the Likelihood of a
Wrongful Conviction

Thus, we can expect that experts in the crimi-

Twenty years of study have helped identify the

nal justice system are as susceptible as experts

leading factors that contribute to wrongful

in other fields to these cognitive biases. For

convictions — eyewitness misidentification, false

example, Dror and Hampikian (2011) found

confessions, untruthful informants, unvalidated

that a forensic scientist’s interpretation of DNA

or improper forensic science, government mis-

mixtures can be biased by knowledge of spe-

conduct and ineffective legal representation.

cific circumstances of a criminal case. Indeed,

Research has revealed changes in practice that

the seminal report on forensic sciences in the

can decrease the likelihood of any one of these

United States by the Committee on Identifying

errors recurring or at least increase the likelihood

the Needs of Forensic Sciences Community et

that the error will be discovered before convic-

al. (2009) explicitly acknowledges the suscepti-

tion (Committee on Identifying the Needs of

bility of bias in the forensic evidence discipline.

Forensic Sciences Community et al., 2009; Drizin

This fallibility plagues all the actors in the justice

and Leo, 2004; Kassin, Bogart, and Kerner, 2012;

system: defense attorneys, for instance, who only

Natapoff, 2006; Steblay et al., 2001; Wells et al.,

litigate certain types of cases are prone to form

1998; Wogalter, Marwitz, and Leonard, 1992).

generalizations based on previous experiences
that hamper their ability to envision alternate

We call on police leaders to (1) adopt best prac-

possibilities (Rand, 2003).

tices that can reduce wrongful convictions and
lead to the correct perpetrator, (2) embrace a

Tunnel vision is problematic in the criminal

culture of learning from error that works to

justice system especially because of what Dror

understand when errors or near misses occur, and

(2014) refers to as the “biasing snowball effect,”

(3) develop investigative practices that continuously

where knowing one piece of evidence can often

8 | New Perspectives in Policing

challenge the unwarranted assumption —

best practices have been developed to manage

shared by witnesses, investigators and attorneys

system variables (Innocence Project, 2014b).

— that we are always right.
First, numerous studies show that when lineup

A. Adopt Best Practices Known To Reduce the
Likelihood of Wrongful Convictions

administrators know who the suspect is, they

In the sections below, we review specific steps

In a study comparing the effects of blind and

in the justice process and present several best

non-blind administration, Canter, Hammond,

practices in the following areas: eyewitness

and Youngs (2013) found that the target photo

identification, interrogation, informant proce-

was selected more than twice as often with an

dures, and evidence preservation and storage.

“informed administrator” than with a “blind

These recommendations will reduce errors and

administrator.” Thus, implementing blind

improve the administration of justice.

administration — when the officer administering

can have a strong biasing effect on eyewitnesses.

the lineup is unaware of who the suspect is — can
1. Eyewitness Identification

significantly reduce the likelihood of misidentifi-

Eyewitness misidentification played a role in

cation (Haw and Fisher, 2004; Phillips et al., 1999;

almost 40 percent of the 1,281 overturned con-

Wells et al., 1998).

victions reported by the National Registry of
Exonerations (2014) and close to 75 percent of
the first 250 DNA exonerations (Garrett, 2011).
Research illustrates that it continues to be a major
factor, contributing to 73 percent of the current
312 DNA exonerations (Innocence Project, 2014b).
Since the landmark National Institute of Justice
study on eyewitness identification was published
almost 15 years ago (Technical Working Group for
Eyewitness Evidence, 1999), scholars have conducted extensive additional research showing
that eyewitnesses are prone to error. Two main
categories of factors interfere with eyewitness
accuracy — estimator variables (e.g., lighting,
presence of a weapon, degree of stress) and system variables (the ways that police retrieve and
record witness memory). Although police officers
have no control over estimator variables, several

Second, eyewitnesses tend to use relative judgments in making identifications, where they
compare lineup members to each other to decide
which one most resembles their memory of the
perpetrator. The major issue with relative judgments in eyewitness lineups is that one lineup
member will always look more like the perpetrator
than other members. This is of particular concern when the actual perpetrator is not included
in the lineup, making the innocent suspect who
most looks like the perpetrator more susceptible
to selection (Wells, Steblay, and Dysart, 2011).
Sequential rather than simultaneous presentation of lineups decreases the rate at which
innocent people are selected by limiting the comparison of photos that eyewitnesses are prone to
make in a simultaneous presentation (Cutler and
Penrod, 1988; Lindsay and Wells, 1985; Steblay et
al., 2001). Organizations such as the Innocence

Policing and Wrongful Convictions | 9

Project (2014d) and the International Association

often than expected by chance. If lineups were

of Chiefs of Police (IACP) (2010) also support the

fair, the mock witnesses would have identified the

use of blind-sequential lineup presentations.

suspect 11 percent of the time (evenly distributed
among the nine individuals in the lineup), yet

Bradfield, Wells, and Olson (2002) show that eye-

the suspect was identified 25 percent of the time

witnesses who were given confirmatory feedback

(Valentine and Heaton, 1999).

on their identification inflated their recollections
of the confidence they felt in their selections. This

To minimize biased lineups, fillers should be

study and others support the best practice that

selected based on a two-part procedure. First,

eyewitnesses provide a statement on their level

filler picks should match critical details of the

of confidence in the identification at the time

witness’s description of the perpetrator. This

the identification is made, when the possibility

will lessen the chance that an innocent suspect

of feedback and memory alterations is minimal.

looks most like the perpetrator. Second, filler

Furthermore, research indicates that having wit-

picks should not look too different from the sus-

nesses instantly answer questions about certainty

pect to ensure that the suspect does not stand

has a partially prophylactic effect against future

out (Koehnken, Malpass, and Wogalter, 1996).

feedback effects (Quinlivan et al., 2009), and in

Finally, all lineups should be properly recorded

the event that feedback does occur, the state-

and documented to allow for review of the lineup

ments can be used to temper jurors’ evaluations

in the event that it was improperly conducted

of the eyewitness identification (Bradfield and

(International Association of Chiefs of Police,

McQuiston, 2004).

2010; Innocence Project, 2014b).

Lineup composition is crucial for a fair lineup, as

According to public information requests in nine

a biased lineup can cause an innocent suspect

states conducted by the Innocence Project and

to stand out and thus increase the likelihood of

affiliated entities, many jurisdictions have imple-

a false identification (Wogalter, Marwitz, and

mented blind-sequential procedures voluntarily

Leonard, 1992). In a fair lineup, mock witnesses’

through the adoption of an official policy. These

selections should be evenly distributed over all

jurisdictions include but are not limited to Santa

of the lineup members, which is consistent with

Clara, California; Monterey, California; Denver,

random chance. If mock witnesses focus on or

Colorado; Palm Beach County, Florida; Baltimore,

choose one lineup member above others, this

Maryland; Hyattsville, Maryland; Norwood,

demonstrates a biased lineup. In one study that

Massachusetts; Hennepin County, Minnesota;

used archival records to examine the fairness of

Colstrip, Montana; North Charleston, South

live lineups conducted in actual cases, research-

Carolina; and Virginia Beach, Virginia. Six states

ers found that mock witnesses, who had never

have also adopted both blind and sequential

seen the suspect before and had only heard a

administration, and five other states have rec-

description, identified the suspect 2.2 times more

ommended the double-blind reform package and

10 | New Perspectives in Policing

incorporated it into police training (Innocence
Project, 2014d). Yet many departments are reluc-

Folder Shuffle Method

indicates there are fewer misidentifications using

Many police departments do not have enough
independent administrators available to perform
the number of blind lineups that are needed. This
problem can be remedied by following folder shuffle
procedures:

a sequential lineup presentation. They argue that

1.

tant to adopt these procedures. That reluctance
seems to stem largely from three concerns. First,
many investigators question the research that

the research has been conducted in a laboratory environment and may not apply in the field.
Second, they contend that blind administration
is an operational problem because it is difficult
to find someone who can act as a blind administrator. Third, investigators feel that turning over
a victim or witness to someone he or she is not
familiar with would hinder the lineup process.
The first of these concerns has been addressed
both in the laboratory and in a recent randomized trial field study in four sites that showed
sequential lineups with blind administration
produced significantly fewer filler picks than the

Place the suspect’s and five filler photos in
individual folders.

2. Shuffle the folders before giving them to the
witness and place four empty folders at the
bottom of the pile (you do not want the witness
to know when he/she is viewing the last photo,
because witnesses may feel an increased need
to make an identification if they believe they are
viewing the last photo).
3. The officer administrating the array should
position himself or herself so he or she cannot
see inside the folders as the witness is reviewing
them.
This method will achieve the same effect as having
an independent blind administrator (International
Association of Chiefs of Police, 2010; Office of the
Attorney General, State of Wisconsin, 2010).

blind simultaneous procedure (11 and 18 percent,
respectively). This reduction in filler picks was

with a training key, to guide police departments

not associated with a reduction in suspect picks,

in implementing these practices (International

with approximately one-quarter of witnesses

Association of Chiefs of Police, 2010).7 Many

choosing a suspect using both procedures (Wells,

departments have adopted policies to address

Steblay, and Dysart, in press). This peer-reviewed

the concerns about staffing and witness/victim

study, the only one of its kind, reinforced in the

comfort. Despite initial resistance, experience

field what has been known in the laboratory for

has proved that implementation is simpler than

30 years and has contributed to the increase in

expected, and the jurisdictions that adopt them

police agencies adopting the recommended poli-

prefer the new procedures.

cies. The experiences of departments that have
implemented these practices also illustrate that

2. Interrogation Procedures

the perceived obstacles are not serious areas

Another major contributing cause of wrong-

of concern (Police Executive Research Forum,

ful convictions is false confessions — innocent

2013). The IACP developed a model policy, along

defendants made outright confessions or pled
guilty in almost 16 percent of exoneration cases,

Policing and Wrongful Convictions | 11

and innocent individuals incriminated them-

suggests that people who are falsely accused tend

selves in some way in more than 25 percent of

to believe that justice will prevail and that their

DNA exonerations (Gross and Shaffer, 2012;

innocence will become transparent to investiga-

Innocence Project, 2014c). Innocent people con-

tors. As a result, they often cooperate fully with

fess for myriad reasons, among them duress,

police, waiving their rights to silence and lawyers

ignorance of the law, and intoxication. Juveniles

and speaking freely to defend themselves. In a

and people with mental disabilities are consid-

study of mock criminals and innocent individu-

ered vulnerable populations at an increased risk

als, the mock criminals varied the amount they

for falsely confessing. Drizin and Leo (2004) ana-

disclosed, but the innocent individuals were

lyzed 125 cases of proven false confessions and

uniformly forthcoming (Kassin, 2005). Another

found that 63 percent were younger than age 25

study compared the willingness of innocent

and 32 percent were younger than age 18. They

individuals to waive their rights and submit to

also found that 22 percent were mentally chal-

questioning to that of guilty individuals; the

lenged and 10 percent had a diagnosed mental

rates were 81 percent and 36 percent, respectively

illness. Although these populations may be the

(Kassin and Norwick, 2004).

most vulnerable, the exonerations include many
fully competent adults who confessed to crimes

Finally, because the decision to interrogate a

they did not commit.

suspect is based on a preliminary suspicion of
that individual, there can be an inherent bias

Several core principles of psychology play a role

during interrogations (known as “investiga-

in false confessions (Kassin, Bogart, and Kerner,

tor bias”). This can lead to a propensity to view

2012; Meissner and Russano, 2003). First, “peo-

suspects as guilty and can inflate officers’ con-

ple make choices that they think will maximize

fidence in their ability to detect deceit (Simon,

their well-being given the constraints they face,

2012). For example, research shows that people

making the best of the situation they are in,”

who were led to believe that a truthful individual

(Kassin et al., 2010, p. 15). Second, people tend to

was being untruthful overestimated the amount

view immediate outcomes more favorably than

of gaze aversion (an action typically associated

delayed outcomes and are impulsive in their

with lying) the individual had displayed during

decision-making (Kassin et al., 2010). Malleability

the interrogation (Levine, Asad, and Park, 2006).

of memory and the ability to manipulate an

Kassin and colleagues (2007) surveyed 631 police

individual’s perception of the past, which were

investigators, who estimated they were accurate

discussed in section II, are also heightened dur-

at truth and lie detection 77 percent of the time

ing stressful situations (Drizin and Leo, 2004;

although the average accuracy rate is actually

Kassin and Kiechel, 1996; Leo and Ofshe, 1998).

about 54 percent, which is essentially no better
than a random guess.

Ironically, innocence itself may put innocent
individuals at risk of falsely confessing. Research

12 | New Perspectives in Policing

Recording custodial interrogations, from the
reading of Miranda rights through the conclusion
of the interrogation, is one way to reduce the likelihood that a false confession will go undetected
through a criminal prosecution (Drizin and Leo,
2004; Kassin, Bogart, and Kerner, 2012). If there is
an accurate record of exactly who said what first,
a review of the interrogation will confirm whether
the suspect provided details that “only the true
perpetrator could know” or whether the interrogator inadvertently introduced those facts. It
creates a record of all the inconsistent statements
made that support the claim that the confession
was not reliable. A recording also has the benefit
of preventing false claims of torture or undue
coercion. Seventeen states, including Arkansas,
Connecticut, Indiana, Maryland, North Carolina,
and Oregon; the District of Columbia; and almost
1,000 law enforcement agencies, many in large
cities such as Atlanta, Boston, Dallas, Denver,
Philadelphia, San Francisco, and St. Petersburg,
now record the entire custodial interrogation in at
least some categories of felonies (Sullivan, 2014).
National organizations, including the IACP and
the American Bar Association Criminal Justice
Section, have endorsed the practice.
Police agencies face challenges in implementing video recording of interrogations. They must
find the money in tight budgets to purchase and
install the equipment needed to implement this
best practice and must modify interview rooms
to obtain clear images and sound. Policies and
procedures must be developed, and officers must
be trained. The greatest challenge is introducing
a change that some will resist; however, with

Interview With William Brooks
William G. Brooks III, Chief of the Norwood,
Massachusetts, Police Department, is a staunch
advocate for eyewitness identification best practices and recording of interrogations. He says about
both policies: “As resistant as I was to them at first,
it was really a breeze. It just works. It’s one of the
best things we’ve done.”
The Norwood Police Department began recording
interrogations in 2004. Although there was initial
concern about individuals being unwilling to talk,
the department has seen no additional resistance
from individuals now that they are being recorded.
Chief Brooks advises discreetly setting up a camera
with a hidden microphone so there is not a constant
reminder of the recording. He believes the benefits
of recording interrogations are significant, including
a reduction in the time detectives spend testifying
about the interview procedures and no longer needing to explain to the jury why police did not record
the interrogation.
The Norwood Police Department implemented the
recommended eyewitness identification practices
after Chief Brooks attended a training hosted by
the District Attorney’s office that promoted the
practices. The adoption of the practices was seamless and the advantages were immense, especially
at trial. Chief Brooks has noticed that because
the lineup procedures now have a scientific basis,
detectives are more professional on the witness
stand, and there has been little cross-examination
about the procedures.
According to Chief Brooks, the key to successfully
adopting these best practices is the training. “From
the police end, there’s so much to know and if you
don’t tell the police the reason behind it, they won’t
get it and they resist it. In the departments that have
the policy and have done the training, there is really
no resistance… . Everywhere I go, they are always
on board.”

Policing and Wrongful Convictions | 13

proper training and policies, experience has

or make the drug purchases that provide the

shown that police departments willingly comply

evidence to support an arrest and prosecution.

with and benefit from the adoption of interroga-

Police and prosecutors also use jailhouse infor-

tion recordings. Sullivan (2004) spoke with 238

mants to obtain information from suspects in

police agencies in 38 states and found that “vir-

custody to strengthen their case.

tually every officer with whom we spoke, having
given custodial recordings a try, was enthusias-

The use of jailhouse informants also plays an

tically in favor of the practice” (Sullivan, 2004,

important role in wrongful convictions. A study

p. 6). Officers noted several benefits to record-

by the Center on Wrongful Convictions (2004)

ing, including a major decrease in the number

found that incentivized testimony was a lead-

of defense motions to suppress statements and

ing factor in 45.9 percent of 111 capital wrongful

confessions and increased capacity to focus on

convictions since the 1970s. In more than 15 per-

the suspect rather than taking copious notes

cent of convictions overturned by DNA testing,

during the interview, and they encountered no

an informant testified against the defendant at

issues affecting their ability to obtain confessions

the original trial (Innocence Project, 2014e).

(Sullivan, 2004). The IACP also has beneficial

Statements made by informants can often be the

training keys to help police agencies institute

main piece of evidence against a defendant, and

electronic recording of custodial interrogations

in some cases it is the only evidence. Although

(International Association of Chiefs of Police,

we understand the value to police in using infor-

2004).

mants to obtain evidence, it is apparent that
measures should be taken to reduce the risk that

3. Informant Practices

incentivized statements may not be truthful and

Informants, or those who provide evidence in

may contribute to more wrongful convictions.

expectation of some benefit from police or prosecutors (e.g., monetary reward, release from
prison, or leniency in their own cases), have a
significant impact in many police investigations
and are particularly prominent in drug investigations. The overwhelming majority of drug cases
are made through the use of informants who are
involved in or on the fringes of the drug trade.
According to federal statistics, 60 percent of drug
defendants cooperate with prosecutors in some
way in exchange for a reduction in their sentence
or a dismissal of charges (Natapoff, 2006). In
these situations, informants identify the dealer

To ensure informants are managed properly, the
IACP recommends requiring that officers complete an Initial Suitability Report for the potential
informant. The report includes information about
the individual’s relationship with the suspect, the
potential informant’s possible motivations, his or
her criminal history and the individual’s ability
to gain information (International Association of
Chiefs of Police, 2008). Understanding and monitoring the motivations of informants is vital to
reducing the likelihood of a wrongful conviction.

14 | New Perspectives in Policing

In Washington, D.C., changes were recently made

informants are when they are considering using

so that the defense is now given information that

one for a case.

might impeach informant testimony at least two
weeks before trial; previously the information

4. Evidence Storage and Preservation

was provided only hours or days before trial (Hsu

In recent decades, it has become clear that

and Alexander, 2013). California passed legisla-

evidence gathered at crime scenes can have a

tion in 2011 requiring prosecutors to corroborate

substantial impact on the determination of inno-

information and incriminating statements pro-

cence. In a preliminary review of closed cases

vided by jailhouse informants by presenting

with claims of innocence over a 10-year period,

forensic evidence or uncompromised testimony

the Innocence Project (2013) found that 32 per-

(Williams, 2011).

cent were closed because evidence had been lost
or destroyed. For example, evidence handling

All case-relevant discussions with informants

played a key role in the 1986 murder conviction

should be electronically recorded, and copies

of Paul House. Blood on House’s jeans was found

of the recordings should be given to the defense.

to have matched the victim’s, but testing in 2008

Because most of these conversations occur in a

determined that the blood had actually come

custodial setting, this would not be a major bur-

from a post-autopsy sample. House’s jeans and a

den for police departments, especially in light

test tube of blood had been shipped in the same

of the increased use of recording interrogations

container to the FBI for analysis, and most of the

(Center on Wrongful Convictions, 2004). Finally,

blood was missing from the tube when it arrived,

departments should create informant files and

suggesting it had been spilled (Balko, 2011).

maintain detailed records of all interactions with
their informants in a central database, and they

With increasingly advanced evidence-testing

should have a process for objectively evaluating

methods, the selection of items that can be used

the investigations. Keeping track of informants

in the conviction or exoneration of an individual

who are later determined to have offered inac-

is growing exponentially. Thus, correct and care-

curate information can minimize the likelihood

ful evidence handling, tracking and storage are

that law enforcement will use them again. For

crucial for establishing innocence both before

example, the FBI maintains records on infor-

and after conviction; preservation is also key

mants’ value to investigations by aggregating

to post-conviction investigations.8 Individuals

their “statistical accomplishments” and assess-

involved in the intake, storage or disposition of

ing their credibility (Natapoff, 2009). Not only will

biological evidence should consider taking online,

this decrease the number of wrongful convictions

in-person or other forms of guided instruction on

due to unreliable informants, it will also benefit

evidence management.

police to have an educated, data-driven understanding of how useful and trustworthy potential

The National Institute of Standards and
Technolog y Technical Work ing Group on

Policing and Wrongful Convictions | 15

Biological Evidence Preservation (2013) recog-

of evidence are best stored (Technical Working

nizes the importance of biological evidence for

Group on Biological Evidence Preservation, 2013).

all parties and at all stages in the criminal jus-

Although there are several ways to improve the

tice system, and it advocates for the long-term

efficiency of tracking and retrieving evidence, the

retention of evidence in several crime categories.

Charlotte-Mecklenburg (North Carolina) Police

Specifically, it recommends the following:

Department has developed a well-organized
cataloging system that barcodes evidence dating

Biological evidence should be preserved

back to 1978 (Moffeit and Greene, 2007).

through, at a minimum, the period of
incarceration in the following crime

Finally, departments should have clear guide-

categories, as defined in NIBRS [National

lines on chain-of-custody documentation that

Incident-Based Repor t ing System

identify “all persons who have had custody of

Resource Guide], regardless of whether

evidence and the places where that evidence

or not a plea was obtained: homicides,

has been kept in chronological order from col-

sexual assault offenses, assaults, kid-

lection to destruction” (Technical Working Group

napping/abductions, and robberies

on Biological Evidence Preservation, 2013, p. 25).

(Technical Working Group on Biological

This is especially important because failure to

Evidence Preservation, 2013, p. 5).

maintain proper documentation of evidence handling may result in evidence being inadmissible.

Police departments in jurisdictions such as Dallas
County, Texas, which have had significantly more
exonerations than others, have been systemati-

B. Learn From Error Using an Organizational
Accident Model

cally saving evidence from rapes, murders and

Implementing an organizational accident model

other major crimes since the 1970s. It stores “tiny

allows police departments to review errors as

lab slides on at least a portion of biological evi-

systemwide weaknesses instead of single-cause

dence collected from all major cases over the

mistakes.

years. As a result, the department has flushed
out culprits in decades-old crimes” (Moffeit and

As noted earlier, research has identified a number

Greene, 2007, p. 4).

of factors that contribute to wrongful convictions
and the best practices that can be implemented

It is further recommended that police agencies

to address these contributing factors. Although

develop written protocols for standardizing

these policies are certain to help reduce wrong-

evidence packaging and tracking in property

ful convictions that might have involved

rooms. These policies should include packaging

eyewitnesses, false confessions, poor evidence

directions with digital photos and brief narrative

preservation or jailhouse informants, they will

descriptions that highlight the approved meth-

not address the systemwide failures that contrib-

ods and a discussion on how various categories

ute to wrongful convictions.

16 | New Perspectives in Policing

Currently, many police departments review

these individual mistakes to lead to tragedies

wrongful convictions through a single-cause,

(Doyle, 2012; Shane, 2013). This type of review is a

one-component review, where the focus of the

forward-looking tool that considers these acci-

search is to pinpoint the one mistake or “bad

dents as a culmination of errors at several stages

apple” that caused the wrongful conviction. The

and treats mistakes as learning opportunities.

underlying assumption in these assessments
is that once the lying informant, eyewitness

There are obstacles to creating and implementing

misidentification or unethical prosecutor is

an organizational accident model for the review

uncovered and remedied, the system will once

of wrongful convictions and near misses. One

again be efficient. However, this type of review

formidable obstacle is that the criminal justice

is flawed because it overlooks the multitude of

system is not a true system — it is an intercon-

less obvious errors that should be corrected to

nected group of individual entities that come

enhance the accuracy of the system as a whole

together to play specific roles in holding individu-

(Doyle, 2012; Shane, 2013). What is needed is an

als accountable for violations of the law. Winning,

approach to reviewing wrongful convictions (and

rather than a search for truth and justice, has

near misses) that not only focuses on the major

become a leading objective. Bringing police,

factors but also looks at the fundamental etiology

prosecutors, defenders and the courts together to

of wrongful convictions.

determine what went wrong is a challenge. One
must overcome the tendency to assign blame even

We recommend that police agencies develop a

though it is clear that the responsibility for wrong-

review system that avoids simply blaming indi-

ful convictions cuts across all parts of the system.

viduals and concentrates on understanding the

Civil liability is also a significant obstacle —

organizational factors that contribute to errors.

how can all of the stakeholders come together in

This type of organizational accident model does

a culture where accountability for errors often

not ask why the individual erred but asks what

comes in the form of institutional or individual

structural issues played a role in the individual’s

financial judgments?

actions that triggered the error. Accidents are
assumed to take place because of one or more

Review mechanisms have been established that

levels of organizational failure: organizational

may help answer that question and serve as a

influences, unsafe supervision, preconditions

guide to developing a process for wrongful con-

for unsafe acts and the unsafe acts themselves.

viction and near-miss assessment. Child fatality

The model accepts that mistakes are inevitable

review teams exist in all 50 states and the District

in a human-dominated system and that no sin-

of Columbia (Langstaff and Sleeper, 2001), and

gle error independently is sufficient to cause a

elder abuse and domestic violence fatality review

significant accident. It shifts the focus to how

teams have also been established in many com-

inherent system weaknesses converge with

munities across America (National Domestic

Policing and Wrongful Convictions | 17

Violence Fatality Review Initiative, 2013; Stigal,

We believe that this model does not need to

2005). These teams bring together police (lead-

be implemented immediately at all levels of

ership and labor), prosecutors, social services,

the criminal justice system and that police

mental health workers, coroners, public health

departments are in the best position to be the

officials and others to examine fatalities. They

forerunners for this kind of change. Through

search for the causes of these deaths and aim to

police leadership, all stakeholders (e.g., defenders,

implement policies or procedures that will help

prosecutors, forensic scientists) will eventually

prevent similar deaths in the future. Their success

become involved, and a new element of profes-

depends on creating an environment where there

sionalism within the criminal justice system will

can be honest dialogue about system flaws and

be created.

approaches for fixing them. The discussions are
confidential, and members are prohibited from

Although hurdles must be overcome to establish

disclosing them outside the team (Stigal, 2005).

a systematic process for wrongful conviction

Some states, such as California, have laws that

review, it is essential that it be done. As Shane

support this confidentiality and protect oral and

(2013, p. 3) states:

written communications from discovery (Stigal,
2005).

In reality, even the best decisions are
constrained by organizational policies,

The field of health care also illustrates opportuni-

personal preparedness and situational

ties for learning from error and has developed a

circumstances beyond the individual’s

comprehensive approach to reviewing mistakes.

control. So, punishing the individual has

In one examination of a wrong-patient surgery,

limited impact; there is symbolic value,

the review team found at least 17 errors, ranging

but that does little to correct the under-

from the patient’s face being draped so physi-

lying problem, shape a culture of safety

cians could not see it to poor communication

and develop organizational learning and

between doctors and nurses. Most critically, the

personal mastery.

review team concluded that no single error could
have caused the wrong-patient surgery. Instead,
the combination of all of these mistakes resulted
in this major error (Chassin and Becher, 2002).
Medical practitioners realized that viewing error
as a failure of character was unproductive and
left latent system weaknesses uninvestigated.
Instead, focusing on the error review in a nonblaming, systems-oriented way would lead to
much more useful information.

Thus, while we implore police leaders to adopt the
best practices described in section III.A of this
paper, we also call on these leaders to help move
the system toward one that continuously learns
from error and seeks truth, better outcomes and
continual system improvement.

C. Test Initial Assumptions
Using the knowledge acquired from wrongful
conviction and near-miss reviews, police can

18 | New Perspectives in Policing

also develop new ways to test initial assumptions

not require a lot of time and serve very important

about a case or a suspect. As we have explained,

purposes. Checklists are very useful in prosecu-

belief perseverance is difficult to overcome. It is

tors’ offices: they remind prosecutors of the steps

exacerbated by the immense institutional pres-

they should take to avoid error, they provide

sures police experience to close cases. Thus,

supervisors with a record of how an employee has

creating checklists from scientifically based

handled a case to date, and they increase the like-

research, previous experiences, and investiga-

lihood that prosecutors are basing their charging

tions that serve as opportunities to question

decisions on all available evidence. For example:

initial reactions can help direct police away
from an innocent individual and toward the real

The Manhattan District Attorney’s Office

perpetrator. Moreover, they provide a concrete

has assembled a number of checklists

record of how an officer handled a case and can

designed to assist its prosecutors in

be used as an example of good practice or to

investigating and prosecuting their cases.

uncover errors as the case progresses.

These checklists did not, for the most part,
create new practices. Instead, they repre-

Dating back to the 1930s, the use of checklists has

sented the formalization of existing office

proven helpful in avoiding pitfalls for business

policies and practices. However, even

and industry in general. As technology improved,

though the checklists did not contain

the amount of data and information available to

new policies, the Manhattan DA’s Office

practitioners in all areas grew, and managing and

wanted to emphasize the importance

digesting it became more difficult. Checklists

of distilling existing office policies into

were developed to help practitioners perform

checklists (Center on the Administration

a long list of steps reproducibly and to manage

of Criminal Law’s Conviction Integrity

complex tasks. It is human nature to overlook

Project, 2011, p. 18).

some steps in the multitude of things we do daily.
Gawande (2009) explains that in some of the most

Checklists can be tailored to promote compli-

complex professions, such as medicine and avia-

ance in a variety of areas. The Manhattan District

tion, checklists are essential, helping in memory

Attorney’s Conviction Integrity Unit has devel-

recall and defining, at a minimum, the necessary

oped a questionnaire to help prosecutors identify

steps needed to understand an overall process.

potential areas in cases where exculpatory material might exist that needs to be disclosed. The

Departments and offices within the criminal

questionnaire is not a formal checklist — a pros-

justice arena are already using checklists to rein-

ecutor does not need to literally check a box for

force best practices because they are low cost, do

each type of evidence that does or does not exist.

Policing and Wrongful Convictions | 19

Instead, the questionnaire provides flexibility

programs designed in the early 1960s used a

to prosecutors in assessing their cases while

daily checklist to evaluate performance and to

still serving the purpose of a standard checklist

ensure trainees practiced skills important to their

(Center on the Administration of Criminal Law’s

success (Kaminsky, 2000). Police have also used

Conviction Integrity Project, 2011).

checklists in domestic violence, homicide, assault
and property crime investigations (Gerberth,

The police have a long history of using check-

2013; Governor’s Commission on Domestic and

lists in a number of areas. Officer field training

Sexual Violence, 2012; Greenberg, 2010; Sadusky,

Case Example: Douglas Warney
On January 3, 1996, the police discovered William Beason’s body in his bedroom. Beason had been stabbed to death
in the neck and chest. They also found a bloodstained knife, towel and tissues in the bathroom hamper.
On the same day the media publicized the details of the murder, Douglas Warney called the police to discuss information he claimed to have about the homicide. Warney is cognitively impaired and has a recorded IQ of 68. He knew the
victim casually, having done housework for him several times in prior years. Warney was taken in for an interrogation,
where police obtained a signed confession stating that he had killed Beason alone.
Before the trial, biological evidence from the crime scene was analyzed. Testing on the towel and tissue found blood
that did not come from either Warney or the victim. Nevertheless, in February 1997, Warney was convicted of seconddegree murder and sentenced to 25 years to life in prison. In 2004, the Innocence Project accepted Warney’s case,
and DNA testing of evidence was issued. The results were entered into the New York State DNA databank of convicted
felons, and the profile from the evidence matched an inmate who was already serving a life sentence for murder.
Warney’s conviction was vacated in 2006, and he was released from prison.
Mistakes occurred at all levels of the criminal justice process, but we will limit our review to police errors. First,
police became focused on Warney as a primary suspect too quickly because of his suggestion that he had information about the murder. Warney had made similar calls to the police before that had not been credible, and he had
recently been admitted to a psychiatric ward. Investigators should have asked questions that would have shed light
on all possibilities as to why Warney called the police. These questions might have revolved around whether police
had exhausted all other possible explanations for Warney’s behavior and whether Warney had any other motives for
calling about the crime.
Second, Warney’s account of the crime changed several times, and his alleged confession was full of inconsistencies. Furthermore, it is important that the mental state and capabilities of an individual be considered when deciding
if and how to interrogate someone, and research indicates that an interrogation that lasts more than four hours
can increase the risk of an individual falsely confessing (Kassin et al., 2010). Warney was interrogated for 12 hours.
Detectives would have benefitted from considering whether there were any factors that put Warney at an increased
risk of falsely confessing.
Finally, when testing on the towel and tissue showed that the blood could not have come from either Warney or the
victim, little was done to look into alternative explanations. After Warney was arrested, police did not explore any other
possible suspects even though there were individuals who could have been questioned. If detectives had investigated
other potential leads, this may have helped offset the effects of tunnel vision and the police’s sole focus on Warney
after his alleged confession (Innocence Project, 2007).

20 | New Perspectives in Policing

2010). Sidebottom, Tilley, and Eck (2012) believe

also as a catalyst to test their initial assumptions

checklists can play an important role in problem-

about the direction of a case or a specific suspect

oriented policing:

during the initial investigation. The sidebar on
page 19 presents a case example that uses this

We see checklists as supplementing the

strategy and offers sample questions that could

numerous strategies already in place to

have been asked and that may have helped law

help manage the complexity involved in

enforcement identify the correct perpetrator

doing problem-oriented policing, not to

instead of the innocent individual.

be worked through slavishly but to act as
an aide-memoir to ensure that actions
that research evidence suggests is important are considered and not forgotten.
Police departments could also profit greatly from
the use of checklists in areas such as eyewitness
identification, interrogations, evidence collection
and laboratory controls. Although police agencies
are encouraged to improve these checklists, the
objective is not to come up with an exhaustive list
of questions. Doyle (2012) notes that as environments change and science advances, checklists
will need to be continuously maintained, evaluated, monitored and perhaps replaced, and an
overreliance on checklists can provide a false
sense of security that everything in an investigation has been covered.
Each case, no matter how similar it may seem to a
previous case, is unique in some way, and boiling
investigations down to a checklist-type system
may engender tunnel vision. Checklists are not a
panacea, but when properly used they minimize
the potential for overlooking important steps in
the investigative process.
Thus, it is critical that police officers use checklists not only as a framework for investigation but

IV. Facilitating and Assisting
Investigations of Post-Conviction
Claims of Innocence
Adopting these recommendations will greatly
reduce the likelihood of wrongful convictions
and decrease the number of perpetrators who
are free to commit additional crimes. Yet some
wrongful convictions are bound to continue —
the burden on every individual in the criminal justice process is great, and human error is
unavoidable. It is essential that police and prosecutors be able to question prior beliefs and be
amenable to the possibility of error when they
are approached by innocence organizations,
convicted defendants or members of their departments who harbor lingering or newfound doubts
about cases or convictions.
Research indicates that this type of openness
is increasing — in 2012, police or prosecutors
initiated or cooperated in 54 percent of the 63
exonerations, whereas in the previous 24 years
they cooperated in only 30 percent of the cases on
average (National Registry of Exonerations, 2014).
Dallas County, Texas, District Attorney Craig
Watkins established the first conviction integrity
unit in 2007 to review and reinvestigate claims

Policing and Wrongful Convictions | 21

of innocence in collaboration with innocence

information from the prosecutor or innocence

organizations and defense attorneys. Other pros-

advocates conducting the investigation. Policies

ecutorial offices are creating similar units. Texas

that spell out how officers are to investigate

leads the nation with 114 exonerations between

claims of innocence ensure that these cases are

1989 and 2012, with police or prosecutors playing

handled consistently and appropriately.

a major role in 53 of them (National Registry of
Exonerations, 2014).

Developing post-conviction units is a more ambitious undertaking, but they lead to enormous

It is clear that police and prosecutors are becom-

benefits. They not only help safeguard the pub-

ing more engaged in these investigations, but

lic, but they can also improve the accuracy of

exactly how they should become involved is not

the cases a department brings and can augment

clear. What is the appropriate role for police in

public confidence in the criminal justice system.

investigating claims of innocence? Should they

Similar to cold case units where police focus

establish something similar to the conviction

exclusively on unsolved crimes, post-conviction

integrity unit or cold case squads that focus on

units specialize in investigating cases where offi-

unsolved homicides and sexual assaults? Should

cers determine that the convicted individual may

they create policies to guide their actions in

be innocent. Although these judgments can be

responding to information requests or when

difficult to assess, police may bring cases to light

officers discover someone might have been

for a variety of reasons, including the emergence

wrongfully convicted?

of new evidence, a witness coming forward, or an
officer believing something went wrong during

We believe that at a minimum, police agencies

an investigation after he or she reviews a case.

should have written policies on how to react to

Once police have identified possible cases, it is

claims of innocence. Although practice var-

important to involve advocates and counsel for

ies from one jurisdiction to another, police

the potentially innocent individual and to work

are frequently the custodians of evidence or

with the relevant members of the community to

investigative files that may contain important

ensure that the case is handled properly.

information — in most states and many large
cities, the crime laboratories where biological

The following sidebar includes three case exam-

evidence is stored and analyzed are under police

ples in which individual police officers and entire

control. In the course of their investigations,

police departments were indispensable in the

police also come across information that may

post-conviction review process. These cases dem-

raise questions about the guilt of someone con-

onstrate the immense benefit of police support.

victed of a crime. Most often, however, the police
will become involved after receiving requests for

22 | New Perspectives in Policing

Case Examples: Post-Conviction Review
Exoneration of LaMonte Armstrong, with assistance from Detective Michael Matthews and the Greensboro Police
Department. LaMonte Armstrong was convicted of a 1988 murder in Greensboro, North Carolina, and was sentenced
to life in prison. His conviction was based entirely on the testimony of an informant who then recanted. After looking
at the case, Duke Law School’s Wrongful Conviction Clinic reached out to the Greensboro Police Department, which
was open to listening to the concerns and revisiting the case. Greensboro Detective Michael Matthews and two law
students reviewed the entire case file, which contained documentation that strongly suggested Armstrong’s innocence.
Duke’s Wrongful Conviction Clinic filed a motion for a new trial, and Armstrong was soon exonerated when DNA
evidence at the crime scene was retested and excluded him. The clinic, Assistant District Attorney Howard Neumann
and the Greensboro Police Department were all instrumental in Armstrong’s exoneration (Duke Law News, 2012).
Exoneration of Glenn Tinney, with assistance from the Mansfield, Ohio, Police Department. Glenn Tinney confessed several times to a 1992 murder in Mansfield, Ohio; however, the multiple confessions were riddled with
inconsistencies (he had also been diagnosed with several mental disorders, including schizophrenia, paranoia and
depression) and the Mansfield Police Department was not convinced of his guilt. Nevertheless, he was still charged and
convicted. In 2006, the Mansfield police brought the case to the Ohio Innocence Project. The Ohio Innocence Project
was able to get Tinney’s guilty plea withdrawn through court hearings in which Mansfield police officers testified about
their belief that Tinney was innocent. He was exonerated in 2013 after serving 21 years in prison (10TV.com, 2012).
Exoneration of Jonathan Moore, with assistance from the Aurora, Illinois, Police Department. Jonathan Moore
was convicted of a 2000 murder in Aurora, Illinois, and was sentenced to more than 50 years in prison. In April 2011,
a confidential informant met with two Aurora detectives and said that someone other than Moore had committed the
murder. Based on that information, police detectives pursued the new lead. Soon after, the Illinois Innocence Project
(2012) became involved in the case; through collaboration with the police and state prosecutors, Moore’s conviction
was vacated and he was exonerated in 2012 (Hanley, 2012).

Conclusion
Police will derive numerous benefits from adopting the recommendations that we have presented
in this paper. First, as we have noted, the costs of
wrongful conviction are substantial from both a
justice and a public safety viewpoint. Every criminal justice official would argue that any rate of
wrongful convictions is unacceptable, and that
all reasonable measures must be taken to ensure
that no innocent individuals are wrongly convicted. By the same token, public safety demands
accuracy. When an innocent person is falsely
convicted, the real perpetrator remains at large
and can (and often does) commit future crimes.
Reducing the likelihood of these events will

inevitably bolster the public’s perception of law
enforcement’s legitimacy.
Second, a department t hat has scientif ically founded procedures in place promotes a
systems-oriented approach to learning from
error. Instead of searching for the one “bad apple,”
it focuses on understanding the structural factors that contribute to errors. This strategy will
improve investigations and encourage police
departments to move beyond an unproductive
culture of blame and toward a culture of continual improvement. Health care reformers realized
a deep reservoir of useful information by focusing
on the mistakes instead of the successes (Sparrow,

Policing and Wrongful Convictions | 23

2008), and we believe this opportunity transfers

best catalysts for this change throughout the

to the criminal justice system.

system.

Finally, we believe that these recommendations

Endnotes

are exemplary opportunities for police to build
partnerships within the criminal justice system.
Although the criminal justice system is not a “system” in the traditional sense, it is clear that all
stakeholders are connected and that a decision or
action by one player will affect the others. Thus, it
is vital for this community, including police, prosecutors, defenders, forensic scientists, judges and
the numerous other actors, to work together and
strive for the best quality of work.
We have proposed that the following will greatly
aid in preventing future harms: adoption of best
practices and procedures, knowledge of existing
or past problems in the system, and guidelines
on how to be most effective in post-conviction
investigations. These are continuously evolving
improvements that require the support of the
entire criminal justice system along with continued research.
We fully acknowledge that police are only one
part of the criminal justice system; however, we
remain steadfast in the belief that they play a
critical role because they are at the front end of
the system and have a major impact on the rest
of the process. Furthermore, police departments
are the most visible segment of the criminal justice community and often bear the brunt of social
criticism for system errors. Police not only have
the most to gain from adopting best practices and
a comprehensive system of recognizing error and
working to minimize it, but they can also be the

1. Heuristics are problem-solving strategies that
often, but not always, result in correct solutions.
For example, when someone is about to cross a
street, he or she most likely does not use complex
mathematical or physics equations to calculate
the distance or the speed of an oncoming car.
Instead, when the person looks both ways and
sees a car approaching, he or she will quickly
form a sense of whether or not there is time to
cross the street while the car is still a distance
away. This use of visual cues and past experiences,
which allows humans to make rapid judgments,
is an example of a heuristic (Weinstein, 2002).
2. Simon (2012, pp. 36-39) describes five mechanisms of biased reasoning that are most
applicable to law enforcement: (1) selective
framing is the tendency to look for information
that should be present if the hypothesis is true,
(2) selective exposure is the tendency to selectively expose oneself to evidence that confirms
one’s hypothesis while ignoring discordant evidence, (3) selective scrutiny is the tendency to
apply strict standards of scrutiny to information
that is incompatible with one’s hypothesis and lax
standards to compatible information, (4) biased
evaluation is the tendency to distort the evaluation of evidence based on one’s hypothesis, and
(5) selective stopping is the tendency to cease
inquiries after finding a sufficient amount of evidence to bolster one’s hypothesis.

24 | New Perspectives in Policing

3. A study on the influence of expertise on X-ray

8. The National Institute of Standards and

image processing demonstrated that although

Technolog y Technical Work ing Group on

expert radiologists were better at detecting

Biological Evidence Preservation recently pub-

abnormal X-ray films, “recognition memory

lished a manual that includes best practices in

for normal X-ray films actually decreased with

the handling, tracking and retention of evidence.

radiological experience to a chance level” (Myles-

The recommendations in this paper are derived

Worsley, Johnston, and Simons, 1988, p. 556).

in large part from this publication. For more
information, refer to Technical Working Group

4. Donsbach (2004) studied the cognitive

on Biological Evidence Preservation (2013). The

processes at work during journalists’ news

International Association for Property and

decision-making and found they are based in

Evidence also offers a wide range of materials,

part on the need for social validation of percep-

including professional standards and manuals,

tions and the need to confirm existing beliefs.

to assist law enforcement agencies in the proper

5. An experiment conducted with expert (those
with more than 10 years of experience) and amateur negotiators showed that while the abilities
of experts to make mutually beneficial trade-offs
were better than those of amateurs, their abilities
to recognize when they had compatible interests
with the other party were relatively similar to
those of amateurs (Thompson, 1990).

handling of evidence. For example, Property
and Evidence by the Book, 2nd ed., extensively
addresses questions about property and evidence
preservation and storage related to law enforcement (Latta and Bowers, 2011).

References
10TV.com. 2012. “Police, Victim’s Family Both
Trying to Free Convicted Killer,” http://www.10tv.

6. In a study examining the use of confirma-

com/content/stories/2012/09/28/mansfield-

tory strategies in therapy, psychotherapists

innocense-project-tries-to-free-man.html.

interviewed students to assess introversion and
extroversion. The results showed that psychotherapists tended to sample behavioral evidence
that would confirm their initial hypotheses when
they chose questions from a list (Dallas and
Baron, 1985).
7. This includes a procedure for presenting a
photo array blindly through a folder shuffle
method even if an independent administrator is
not available.

Acker, James. 2013. “The Flipside Injustice of
Wrongful Convictions: When the Guilty Go Free.”
Albany Law Review 76: 1629-1712.
Arkes, Hal, David Faust, Thomas Guilmette, and
Kathleen Hart. 1988. “Eliminating the Hindsight
Bias.” Journal of Applied Psychology 73: 305-307.
Balko, Radley. 2011. “Wrongful Convictions: How
Many Innocent Americans Are Behind Bars?”
Reason (July 7).

Policing and Wrongful Convictions | 25

Bradfield, Amy, Gary Wells, and Elizabeth Olson.

Chassin, Mark, and Elise Becher. 2002. “The

2002. “The Damaging Effect of Confirming

Wrong Patient.” Annals of Internal Medicine

Feedback on the Relation Between Eyewitness

136(11): 826-833.

Certainty and Identification Accuracy.” Journal
of Applied Psychology 87: 112-120.

Committee on Identifying the Needs of Forensic
Sciences Community; Committee on Science,

Bradfield, Amy, and Dawn McQuiston. 2004.

Technology, and Law Policy and Global Affairs;

“When Does Evidence of Eyewitness Confidence

Committee on Applied and Theoretical Statistics;

Inflation Affect Judgments in a Criminal Trial?”

and Division on Engineering and Physical

Law and Human Behavior 28: 369-387.

Sciences, National Research Council of the
National Academies. 2009. Strengthening Forensic

Canter, David, Laura Hammond, and Donna

Science in the United States: A Path Forward.

Youngs. 2013. “Cognitive Bias in Line-Up

Washington, D.C.: National Academies Press.

Identifications: The Impact of Administrator
Knowledge.” Science and Justice 53: 83-88.

Cut ler, Brian, and Steven Penrod. 1988.
“Improv ing t he Reliabilit y of Eyew itness

Carson, E. Ann, and Daniela Golinelli. 2013.

Ident if icat ion: Lineup Const ruct ion and

Prisoners in the United States, 2012: Trends in

Presentation.” Journal of Applied Psychology 73:

Admissions and Releases, 1991-2012. Washington,

281-290.

D.C.: U.S. Department of Justice, Bureau of Justice
Statistics.

Dallas, Mercedes, and Robert Baron. 1985. “Do
Psychotherapists Use a Confirmatory Strategy

Center on the Administration of Criminal Law’s

During Interviewing?” Journal of Social and

Conviction Integrity Project. 2011. “Establishing

Clinical Psychology 3: 106-122.

Conviction Integrity Programs in Prosecutors’
Offices.” New York, New York: New York University

Donsbach, Wolfgang. 2004. “Psychology of

School of Law, Center on the Administration of

News Decisions: Factors Behind Journalists’

Criminal Law’s Conviction Integrity Project.

Professional Behavior.” Journalism 5: 131-157.

Center on Wrongful Convictions. 2004. “The

Doyle, James. 2012. “Learning About Learning

Snitch System: How Snitch Testimony Sent Randy

From Error.” Ideas in American Policing 14 (May).

Steidl and Other Innocent Americans to Death
Row.” Chicago, Illinois: Northwestern University

Drizin, Steven, and Richard Leo. 2004. “The

School of Law, Bluhm Legal Clinic, Center on

Problem of False Confessions in the Post-DNA

Wrongful Convictions.

World.” North Carolina Law Review 82: 891-1007.

26 | New Perspectives in Policing

Dror, Itiel. 2011. “The Paradox of Human

Gawande, Atul. 2009. The Checklist Manifesto:

Expertise: Why Experts Get It Wrong.” In The

How to Get Things Right. New York, New York:

Paradoxical Brain, ed. Narinder Kapur. New

Metropolitan Books.

York, New York.: Cambridge University Press, pp.
177-198.

Gerberth, Vernon. 2013. Practical Homicide
Investigation Checklist and Field Guide, 2nd ed.

Dror, Itiel. 2014. “Cognitive Forensics: Human

Boca Raton, Florida: CRC Press.

Cognition, Contextual Information, and Bias.” In
Encyclopedia of Criminology and Criminal Justice,

Governor’s Commission on Domestic and Sexual

ed. G.J.N. Bruinsma and D.L. Weisburd. New York,

Violence. 2012. “A Model Protocol for Response

New York: Springer Science+Business Media.

to Adult Sexual Assault Cases.” Concord, New
Hampshire: State of New Hampshire, Office of

Dror, Itiel, and Peter Fraser-Mackenzie. 2009.

the Attorney General.

“Cognitive Biases in Human Perception, Judgment,
and Decision Making: Bridging Theory and the

Greenberg, Mor r is. 2010. “Invest igat ing

Real World.” In Criminal Investigative Failures, ed.

Proper t y Crimes: A Check list for

D. Kim Rossmo. Boca Raton, Florida: CRC Press,

Success.” Investigations eNews (July 20),

pp. 53-67.

http://www.policeone.com/police-products/
investigation/articles/2142763-Investigating-

Dror, It iel, a nd Greg Ha mpi k ia n. 2011.

property-crimes-A-checklist-for-success/.

“Subjectivity and Bias in Forensic DNA Mixture
Interpretation.” Science & Justice 51(4): 204-208.

Gross, Samuel. 2008. “Convicting the Innocent.”
Annual Review of Law and Social Science 4:

Duke Law News. 2012. “Duke Law Faculty,

173-192.

Alumni, and Students Secure Release, New Trial,
for LaMonte Armstrong,” http://law.duke.edu/

Gross, Samuel, and Michael Shaffer. 2012.

news/duke-law-faculty-alumni-and-students-

“Exonerations in the United States, 1989-2012:

secure-release-new-trial-lamonte-armstrong/.

Report by the National Registry of Exonerations.”
Ann Arbor, Michigan: University of Michigan

Findley, Keith, and Michael Scott. 2006. “The

School of Law.

Multiple Dimensions of Tunnel Vision in Criminal
Cases.” Wisconsin Law Review 2006: 291-307.

Hanley, Matt. 2012. “Prosecutors Drop Murder
Case Against Aurora Man Imprisoned Ten Years.”

Garrett, Brandon. 2011. Convicting the Innocent:

Chicago Sun Times (March 6).

Where Criminal Prosecutions Go Wrong.
Cambridge, Massachusetts: Harvard University

Haw, Ryan, and Ronald Fisher. 2004. “Effects of

Press.

Administrator-Witness Contact on Eyewitness

Policing and Wrongful Convictions | 27

Identification Accuracy.” Journal of Applied

University, Benjamin N. Cardozo School of Law,

Psychology 89: 1106-1112.

Innocence Project, http://www.innocenceproject.
org/understand/Eyewitness-Misidentification.

Hsu, Spencer, and Keith Alexander. 2013. “D.C.

php.

Plans Changes in Police Lineups, Informants
to Stop Error.” The Washington Post (February

Innocence Project. 2014c. “False Confessions.”

13), http://w w w.washingtonpost.com/local/

New York, New York: Yeshiva Universit y,

crime/dc-plans-changes-in-police-lineups-

Benjamin N. Cardozo School of Law, Innocence

informants-to-stop-errors/2013/02/13/315673d4-

Project, http://w w w.innocenceproject.org/

7617-11e2-8f84-3e4b513b1a13_story.html.

understand/False-Confessions.php.

Illinois Innocence Project. 2012. “The Jonathan

Innocence Project. 2014d. “The Implementation

Moore Case.” Springfield, Illinois: University of

of Double Blind-Sequential Presentation of

Illinois-Springfield, Illinois Innocence Project,

Lineups.” New York, New York: Yeshiva University,

http://www.uis.edu/innocenceproject/cases/

Benjamin N. Cardozo School of Law, Innocence

exonorees/TheJonathanMooreCase.html.

Project.

Innocence Project. 2007. “Douglas Warney.” New

Innocence Project. 2014e. “Informants.” New

York, New York: Yeshiva University, Benjamin N.

York, New York: Yeshiva University, Benjamin

Cardozo School of Law, Innocence Project, http://

N. Cardozo School of Law, Innocence Project,

www.innocenceproject.org/Content/Douglas_

http://www.innocenceproject.org/understand/

Warney.php.

Snitches-Informants.php.

Innocence Project. 2013. “DNA Exoneration

International Association of Chiefs of Police. 2004.

Cases W here Evidence Was Believed Lost

“Interrogations and Confessions Model Policy.”

or Destroyed.” New York, New York: Yeshiva

Alexandria, Virginia: International Association

University, Benjamin N. Cardozo School of Law,

of Chiefs of Police.

Innocence Project, http://www.innocenceproject.
org/Content/DNA_Exoneration_Cases_Where_

International Association of Chiefs of Police.

Evidence_Was_Believed_Lost_or_Destroyed.php.

2008. “Confidential Informants Model Policy.”
Alexandria, Virginia: International Association

Innocence Project. 2014a. “DNA Exoneree Case

of Chiefs of Police.

Profiles.” New York, New York: Yeshiva University,
Benjamin N. Cardozo School of Law, Innocence

International Association of Chiefs of Police.

Project, http://www.innocenceproject.org/know/.

2010. “Eyewitness Identification Model Policy.”
Alexandria, Virginia: International Association

I n nocence P rojec t . 2014b. “Eyew it ness

of Chiefs of Police, http://www.theiacp.org/about/

Misidentification.” New York, New York: Yeshiva

whatsnew/tabid/459/default.aspx?id=1336&v=1.

28 | New Perspectives in Policing

Kaminsky, Glenn. 2000. The Field Training

Research.” In Psychological Issues in Eyewitness

Concept in Criminal Justice Agencies. Upper

Identification, ed. S. Sporer, R.S. Malpass, and

Saddle River, New Jersey: Prentice Hall.

G. Koehnken. Mahwah, New Jersey: Lawrence
Erlbaum Associates, Inc., pp. 205-231.

Kassin, Saul. 2005. “On the Psycholog y of
Confessions: Does Innocence Put Innocents at

Langstaff, John, and Tish Sleeper. 2001. The

Risk?” American Psychologist 60: 215-228.

National Center on Child Fatality Review. Fact
Sheet. Washington, D.C.: U.S. Department of

Kassin, Saul, Daniel Bogart, and Jacqueline

Justice, Office of Juvenile Justice and Delinquency

Kerner. 2012. “Confessions That Corrupt:

Prevention.

Evidence from the DNA Exoneration Case Files.”
Psychological Science 23: 41-45.

Latta, Joseph, and Gordon Bowers. 2011. Property
and Evidence by the Book, 2nd ed. Burbank,

Kassin, Saul, Steven Drizin, Thomas Grisso, Gisli

California: International Association for Property

Gudjonsson, Richard Leo, and Allison Redlich.

and Evidence.

2010. “Police-Induced Confessions: Risk Factors
and Recommendations.” Law and Human

Leo, Richard, and Richard Ofshe. 1998. “The

Behavior 34: 3-38.

Consequences of False Confessions: Deprivations
of Liberty and Miscarriages of Justice in the Age of

Kassin, Saul, and Katherine Kiechel. 1996.

Psychological Interrogation.” Journal of Criminal

“The Social Psychology of False Confessions:

Law & Criminology 88: 429-496.

Compliance, Internalization and Confabulation.”
Psychological Science 7: 125-128.

Levine, Timothy, Kelli Jean Asad, and Hee Sun
Park. 2006. “The Lying Chicken and the Gaze

Kassin, Saul, Richard Leo, Christian Meissner,

Avoidant Egg: Eye Contact, Deception and Causal

Kimberly Richman, Lori Colwell, Amy-May Leach,

Order.” Southern Communication Journal 71:

and Dana La Fon. 2007. “Police Interviewing and

401-411.

Interrogation: A Self-Report Survey of Police
Practices and Beliefs.” Law and Human Behavior

Lindsay, R.C., a nd Ga r y L . Wel ls. 1985.

31: 381-400.

“Improving Eyewitness Identification from
Lineups: Simultaneous versus Sequential Lineup

Kassin, Saul, and Rebecca Norwick. 2004. “Why

Presentations.” Journal of Applied Psychology 70:

People Waive Their Miranda Rights: The Power

556-564.

of Innocence.” Law and Human Behavior 28:
211-221.

Meissner, C.A., and Melissa Russano. 2003.
“The Psychology of Interrogations and False

Koehnken, Gunter, Roy Malpass, and Michael
Wogalter. 1996. “Forensic Applications of Line-Up

Confessions: Research and Recommendations.”

Policing and Wrongful Convictions | 29

Canadian Journal of Police & Security Services

Phillips, Mark, Bradley McAuliff, Margaret Bull

1: 53-64.

Kovera, and Brian Cutler. 1999. “Double-Blind
Photoarray Administration as a Safeguard

Moffeit, Miles, and Susan Greene. 2007. “Room for

Against Investigator Bias.” Journal of Applied

Error in Evidence Vaults.” The Denver Post (July 23).

Psychology 84: 940-951.

Myles-Worsley, Marina, William Johnston,

Police Executive Research Forum. 2013. “A

and Margaret Simons. 1988. “The Influence of

National Survey of Eyewitness Identification

Expertise on X-Ray Image Processing.” Journal

Procedures in Law Enforcement Agencies.” Final

of Experimental Psychology: Learning, Memory,

report for National Institute of Justice, grant

and Cognition 14: 553-557.

number 2010-IJ-CX-0032, NCJ 242617, http://

Natapoff, Alexandra. 2006. “Beyond Unreliable:
How Sn itc hes Cont r ibute to Wrong f u l
Convictions.” Golden Gate University Law Review
37: 106-130.
Natapoff, Alexandra. 2009. Snitching: Criminal
Informants and the Erosion of American Justice.
New York, New York: New York University Press.

www.policeforum.org/assets/docs/Free_Online_
Documents/Eyewitness_Identification/a%20
national%20survey%20of%20eyewitness%20
identification%20procedures%20in%20law%20
enforcement%20agencies%202013.pdf.
Quinlivan, Deah, Jeffrey Neuschatz, Angelina
Jimenez, Andrew Cling, Amy Bradfield Douglass,
and Charles Goodsell. 2009. “Do Prophylactics

National Domestic Violence Fatality Review

Prevent Inflation? Post-Identification Feedback

Initiative. 2013. “Frequently Asked Questions —

and the Effectiveness of Procedures to Protect

What Is a Domestic Violence Fatality Review?”

Against Confidence-Inflation in Earwitnesses.”

Flagstaff, Arizona: Northern Arizona University,

Law and Human Behavior 33: 111-121.

National Domestic Violence Fatality Review
Initiative, http://www.ndvfri.org/faqs.php.

Rand, Joseph. 2003. “Understanding Why Good
Lawyers Go Bad: Using Case Studies in Teaching

Nat ional Reg ist r y of Exonerat ions. 2014.

Cognitive Bias in Legal Decision-Making.”

“Exonerations in 2013.” Ann Arbor, Michigan:

Clinical Law Review 9: 731-782.

University of Michigan School of Law.

Risinger, Michael. 2008. “Innocents Convicted:

Office of the Attorney General, State of Wisconsin.

An Empirically Justified Factual Wrongful

2010. “Model Policy and Procedure for Eyewitness

Conviction Rate.” Journal of Criminal Law &

Identification.” Madison, Wisconsin: Wisconsin

Criminology 97: 761-806.

Department of Justice.

Sadusky, Jane. 2010. “Best Practice Checklist
for Improv i ng Com mu n it y Response to

30 | New Perspectives in Policing

Domestic Violence: Police Investigations.”

Rates in Sequential and Simultaneous Lineup

Duluth, Minnesota: Praxis International. Report

Presentations: A Meta-Analytic Comparison.”

submitted to the Office on Violence Against

Law and Human Behavior 25: 459-474.

Women, grant number 2004-W T-A X-K052,
http://w w w.pra x isinternat ional.org/f iles/

Stigal, Lori. 2005. Elder Abuse Fatality Review

praxis/files/Safety%20Audits/TTABestPractice

Teams: A Replication Manual. Washington, D.C.:

ChecklistPoliceInvestMar2012.pdf.

American Bar Association.

Shane, Jon. 2013. Learning from Error in Policing:

Stubbins, David, and Nelson Stubbins. 2009. “On

A Case Study in Organizational Accident Theory.

the Horns of a Narrative: Judgment, Heuristics,

New York, New York: Springer.

and Biases in Criminal Investigation.” In Criminal
Investigative Failures, ed. D. Kim Rossmo. Boca

Sidebottom, Aiden, Nick Tilley, and John Eck.

Raton, Florida: CRC Press, pp. 99-142.

2012. “What of Checklists for Problem-Oriented
Policing?” Community Policing Dispatch 5(5).

Sullivan, Thomas. 2004. “Police Experiences with

Washington, D.C.: U.S. Department of Justice,

Recording Custodial Interrogations.” Chicago,

Office of Community Oriented Policing Services,

Ilinois: Northwestern University School of

http://cops.usdoj.gov/html/dispatch/05-2012/

Law, Bluhm Legal Clinic, Center on Wrongful

POPChecklists.asp.

Convictions.

Simon, Dan. 2012. In Doubt: The Psychology

Sullivan, Thomas. 2014. “Departments that

of the Criminal Justice Process. Cambridge,

Currently Record a Majorit y of Custodial

Massachusetts: Harvard University Press.

Interrogations.” Chicago, Illinois: Jenner & Block
LLP.

Snook, Brent, and Richard Cullen. 2009. “Bounded
Rationality and Criminal Investigations: Has

Technical Work ing Group for Eyew itness

Tunnel Vision Been Wrongfully Convicted?”

Evidence. 1999. Eyewitness Evidence: A Guide

In Criminal Investigative Failures, ed. D. Kim

for Law Enforcement. Washington, D.C.: U.S.

Rossmo. Boca Raton, Florida: CRC Press, pp.

Department of Justice, National Institute of

71-98.

Justice.

Sparrow, Malcolm. 2008. The Character of Harms:

Technical Working Group on Biological Evidence

Operational Challenges in Control. New York, New

Preservation. 2013. The Biological Evidence

York: Cambridge University Press.

Preservation Handbook: Best Practices for
Evidence Handlers. Gaithersburg, Maryland: U.S.

Steblay, Nancy, Jennifer Dysart, Solomon Fulero,

Department of Commerce, National Institute of

and R.C.L. Lindsay. 2001. “Eyewitness Accuracy

Standards and Technology.

Policing and Wrongful Convictions | 31

Thompson, Leigh. 1990. “An Examination of

Williams, Carol. 2011. “California May Curb

Naive and Experienced Negotiators.” Journal of

Use of Unsupported Jailhouse Testimony.”

Personality and Social Psychology 59: 82-90.

Los Angeles Times (July 31), http://articles.
latimes.com/2011/jul/31/local/la-me-jailhouse-

Valentine, Tim, and Pamela Heaton. 1999. “An

informants-20110731.

Evaluation of the Fairness of Police Line-ups
and Video Identifications.” Applied Cognitive

Wogalter, Michael, Bradley Marwitz, and David

Psychology 13(S1): S59-S72.

Leonard. 1992. “Suggestiveness in Photospread
Line-ups: Similarity Induces Distinctiveness.”

Weinstein, Ian. 2002. “Don’t Believe Everything

Applied Cognitive Psychology 6: 443-453.

You Think: Cognitive Bias in Legal Decision
Making.” Clinical Law Review 9: 783-834.
Wells, Gary L., Mark Small, Steven Penrod,
Roy Malpass, Solomon Fulero, and C.A.E.
Brimacombe. 1998. “Eyewitness Identification
Procedures: Recommendations for Lineups and
Photospreads.” Law and Human Behavior 22:
603-647.
Wells, Gary L., Nancy K. Steblay, and Jennifer
E. Dysart. 2011. A Test of the Simultaneous vs.
Sequential Lineup Methods: An Initial Report of
the AJS National Eyewitness Identification Field
Studies. Des Moines, Iowa: American Judicature
Society.
Wells, Gary L., Nancy K. Steblay, and Jennifer E.
Dysart. In press. Double-Blind Photo-Lineups
Using Actual Eyewitnesses: An Experimental
Test of a Sequential Versus Simultaneous Lineup
Procedure. Law and Human Behavior.

Author Note
A nt hony W. Bat ts, D.P.A., is t he Pol ice
Commissioner in Baltimore, Maryland, and former Chief of Police, Long Beach and Oakland,
California.
Maddy deLone is Executive Director of the
Innocence Project, a nonprofit legal organization committed to exonerating the wrongfully
convicted through the use of DNA testing and
criminal justice system reform.
Darrel W. Stephens is the retired Chief of the
Charlotte-Mecklenburg Police Department. He is
the Executive Director of the Major Cities Chiefs
Association.
Special thanks to Elizabeth Janszky, who provided invaluable research and drafting assistance
with this bulletin. Without her hard work, this
bulletin would not have been possible.

U.S. Department of Justice
Office of Justice Programs
National Institute of Justice

presorted standard
postage & fees paid
DOJ/NIJ
Permit No. G­–91

*NCJ~246328*

Washington, DC 20531
Official Business
Penalty for Private Use $300

MAILING LABEL AREA
(5” x 2”) DO NOT PRINT THIS
AREA (INK NOR VARNISH)

Members of the Executive Session on Policing and Public Safety
Commissioner Anthony Batts, Baltimore
Police Department

Chief Edward Flynn, Milwaukee
Police Department

Professor David Bayley, Distinguished
Professor (Emeritus), School of Criminal
Justice, State University of New York at
Albany

Colonel Rick Fuentes, Superintendent,
New Jersey State Police

Professor Anthony Braga, Senior Research
Fellow, Program in Criminal Justice Policy
and Management, John F. Kennedy School
of Government, Harvard University; and
Don M. Gottfredson Professor of EvidenceBased Criminology, School of Criminal
Justice, Rutgers University
Chief Jane Castor, Tampa Police Department
Ms. Christine Cole (Facilitator), Executive
Director, Program in Criminal Justice Policy
and Management, John F. Kennedy School
of Government, Harvard University
Commissioner Edward Davis, Boston
Police Department (retired)
Chief Michael Davis, Director, Public Safety
Division, Northeastern University

District Attorney George Gascón, San
Francisco District Attorney’s Office
Mr. Gil Kerlikowske, Director, Office of
National Drug Control Policy
Professor John H. Laub, Distinguished
University Professor, Department of
Criminology and Criminal Justice, College of
Behavioral and Social Sciences, University
of Maryland, and former Director of the
National Institute of Justice
Chief Susan Manheimer, San Mateo
Police Department
Superintendent Garry McCarthy, Chicago
Police Department
Professor Tracey Meares, Walton Hale
Hamilton Professor of Law, Yale Law School
Dr. Bernard K. Melekian, Director, Office

Mr. Ronald Davis, Director, Office of
Community Oriented Policing Services,
United States Department of Justice

of Community Oriented Policing Services
(retired), United States Department of
Justice

Ms. Madeline deLone, Executive Director,
The Innocence Project

Ms. Sue Rahr, Director, Washington State
Criminal Justice Training Commission

Dr. Richard Dudley, Clinical and Forensic
Psychiatrist

Commissioner Charles Ramsey,
Philadelphia Police Department

Professor Greg Ridgeway, Associate
Professor of Criminology, University of
Pennsylvania, and former Acting Director,
National Institute of Justice
Professor David Sklansky, Yosef
Osheawich Professor of Law, University of
California, Berkeley, School of Law
Mr. Sean Smoot, Director and Chief Legal
Counsel, Police Benevolent and Protective
Association of Illinois
Professor Malcolm Sparrow, Professor
of Practice of Public Management, John F.
Kennedy School of Government, Harvard
University
Mr. Darrel Stephens, Executive Director,
Major Cities Chiefs Association
Mr. Christopher Stone, President, Open
Society Foundations
Mr. Richard Van Houten, President, Fort
Worth Police Officers Association
Lieutenant Paul M. Weber, Los Angeles
Police Department
Professor David Weisburd, Walter E. Meyer
Professor of Law and Criminal Justice,
Faculty of Law, The Hebrew University;
and Distinguished Professor, Department
of Criminology, Law and Society, George
Mason University
Dr. Chuck Wexler, Executive Director,
Police Executive Research Forum

Learn more about the Executive Session at:
www.NIJ.gov, keywords “Executive Session Policing”
www.hks.harvard.edu, keywords “Executive Session Policing”

NCJ 246328