Vol. 77, No. 3, March
Looks can be Deceiving:
Safeguards for Eyewitness Identification
Recent cases in Wisconsin and elsewhere point to two particularly
troubling problems with current eyewitness identification techniques ...
memory contamination through investigator suggestion and a witness's use
of the relative judgment process. Read how some jurisdictions are
minimizing misidentification and erroneous conviction of the
Winn S. Collins
egal researchers have determined repeatedly that
the single leading cause of wrongful conviction in the United States is
eyewitness misidentification.1 A study
conducted in Wisconsin reached a similar conclusion, determining that
eyewitness misidentification was involved in approximately one half of
Wisconsin exoneration cases.2 Erroneous
identification is especially troubling because social scientists already
have determined that the criminal justice system can significantly
reduce the rate of misidentification by implementing subtle changes to
This article reviews several misidentification cases in Wisconsin,
including the highly publicized release of Steven Avery in September
2003, explains the leading causes of misidentification, and recommends
changes to eyewitness evidence collection techniques. The article also
recognizes several commendable efforts recently undertaken by some
within the justice system to implement new collection procedures.
Misidentification in Wisconsin
The Wisconsin criminal justice system is not immune to the dangers of
eyewitness misidentification and conviction of the innocent. In the
early 1970s, a Milwaukee County jury convicted Francis Hemauer of
abduction, rape, and attempted murder after the victim identified him at
both a photograph and live lineup procedure.3 Hemauer served eight years in prison before
testing of seminal deposits excluded him as the source of the
semen.4 In 1996, Anthony Hicks was
exonerated when DNA testing revealed that he was not the source of the
hair found at the scene of a sexual assault that occurred in
1990.5 Hicks was convicted, in large part,
based upon the victim's identification at a lineup. In September 2003,
Steven Avery became yet another example illustrating the ramifications
of eyewitness misidentification.
The Steven Avery Exoneration. Steven Avery was
convicted of first-degree sexual assault, attempted murder, and false
imprisonment of a Manitowoc woman.6 On July
29, 1985, the woman was jogging on a Lake Michigan beach when a man
attacked and sexually assaulted her. The victim later testified that,
during the assault, she remembered thinking, "I have to stay calm and
get a good look at this guy."7 The victim
provided a detailed description of her attacker to investigators from
the Manitowoc County Sheriff's Department. The sheriff suspected that
Avery was the attacker because his features matched those provided in
the victim's description. Only a few hours after the attack, the victim
selected Avery from a photograph lineup during which the sheriff told
her "there was a chance that the suspect might be in there."8 Three days later, the victim "immediately"
identified Avery from a live lineup after the sheriff told the victim
that the man whose photograph she had selected had been arrested.9 At trial, the victim testified "[t]here is
absolutely no question in my mind" that he was the attacker.10 The jury convicted Avery. The judge sentenced
Avery to 32 years in prison.
On Sept. 11, 2003, Avery was released from prison, after DNA testing
matched a hair found at the crime scene to Gregory Allen, a man serving
a 60-year sentence for a similar sexual assault.11 Before the exoneration, Avery had argued that
the identification procedure used by the sheriff was impermissibly
suggestive. In an unpublished opinion, the Wisconsin Court of Appeals
denied Avery's claim, noting "the photo array constitutes one of the
fairest ones this court has seen."12 A
decade later, in a second appeal, Avery unsuccessfully tried to obtain a
new trial after testing revealed the presence of DNA from an unknown
third party.13 Avery finally proved his
innocence with the help of the Wisconsin Innocence Project and a
two-year-old law that increased a defendant's access to DNA
testing.14 Avery had served nearly 18 years
in prison before proving his innocence.
The Eugene Glenn Exoneration. Only one month after
Avery's release, the Milwaukee Journal Sentinel reported the
exoneration case of Eugene Glenn, another case involving
misidentification.15 Glenn was convicted
for the armed robbery of a Milwaukee woman on Oct. 11, 2001. A Milwaukee
police detective suspected Glenn because he matched the general
description provided by a man who received some of the property stolen
during the robbery. The victim identified Glenn at a lineup and, at
trial, she told the jury that she "recognized his face immediately." The
jury convicted Glenn and the judge sentenced him to 20 years in prison.
On his appeal, an investigator and public defender contacted the man who
originally had received the stolen property. The man implicated Eric
Gardner, who confessed to the crime during a police interview on Oct. 4,
Causes of Misidentification
Experts on eyewitness identification have known the dangers and
causes of misidentification well before the Avery and Glenn
exonerations. As long ago as 1932, Yale law professor Edwin Borchard
reported that identification procedures are susceptible to producing
misidentification and new procedures based on scientific study should be
developed.16 More recently, in the last two
and a half decades, researchers have identified two particularly
troubling problems with current identification techniques. The first is
memory contamination through investigator suggestion and the second is a
witness's use of the relative judgment process.
Memory Contamination. Memory contamination occurs
during an identification procedure when an investigator makes an
accidental or subtle remark. For example, before a lineup, a detective
may comment to a witness that the detective has a suspect in mind,
thereby encouraging the witness to identify someone during the
identification procedure. Memory contamination likely occurred in the
Avery case before the photo lineup because the sheriff
commented that the victim could assume the suspect was present in the
lineup. The sheriff further contaminated the victim's memory before the
live lineup by telling the victim that his department had arrested the
man she selected in the earlier photo lineup.
Relative Judgment Process. The second problem, the
relative judgment process, occurs when a witness views a simultaneous
lineup (that is, all members of the lineup are shown to the witness at
the same time). In a report by the U.S. Department of Justice -
published the same month as Avery's exoneration - identification experts
explained that "[r]esearch shows that eyewitnesses tend to select the
person who looks most like the perpetrator relative to the other lineup
members."17 When the actual perpetrator is
absent from the lineup, the witness often selects the innocent
individual who most closely resembles the witness's memory of the
culprit. The relative judgment phenomenon almost certainly contributed
to the misidentification of Avery because the actual perpetrator,
Gregory Allen, was not presented at either lineup procedure.
The preferable identification approach invokes the absolute judgment
process, in which an eyewitness compares each lineup participant only to
the witness's memory of the perpetrator and not to the other members
within the lineup. Absolute judgment involves a cognitive process in
which the witness almost automatically recognizes the actual culprit
from a lineup because "the suspect looks sufficiently like the person in
the witness's memory to meet some internal threshold of
recognition."18 An investigator may
encourage a witness to invoke the absolute judgment process by showing
the lineup participants one person at a time, thereby preventing a
witness from comparing lineup members to one another.
Solutions to Misidentification
By the late 1990s, academic and government officials recognized that
broad reforms were needed to minimize the misidentification and
erroneous conviction of the innocent. In 1998 the American Psychology
and Law Society published "good-practice guidelines for constructing and
conducting lineups and photospreads for eyewitnesses to crime."19 The following year, in October 1999, the U.S.
Department of Justice published Eyewitness Evidence: A Guide for Law
Enforcement (DOJ Eyewitness Guide).20
These publications recommended a host of reforms to lineup procedures,
including recommendations for blind identification procedures and
Blind Testing. The American Psychology and Law
Society report recommended the adoption of blind identification
procedures. That is, "[t]he person who conducts the lineup ... should
not be aware of which member of the lineup ... is the suspect."21 The report stated that the recommendation for
blind testing does not presume any deliberate impropriety on the part of
law enforcement investigators; instead, the report explained that
investigators often accidentally reveal clues to the eyewitness through
eye contact, visible facial expressions, and verbal exchanges. The use
of blind procedures prevents an investigator from accidentally providing
suspect information to a witness, thereby significantly decreasing the
likelihood of misidentification based upon memory contamination. The use
of blind procedures in the Avery case would have prevented the
sheriff from telling the victim that the man whose photograph she had
selected in the photo lineup had been arrested.
Sequential Lineups. The abovementioned publications
also supported the use of sequential lineup techniques (that is, lineups
conducted by displaying one photo or one person at a time to the
eyewitness). The American Psychology and Law Society report recognized
that "the sequential procedure ... reduces the propensities of
eyewitnesses to make relative judgments."22
In a supplement to the original DOJ Eyewitness Guide, the authors stated
that the sequential procedure prevents the eyewitness from making
relative judgments, thereby encouraging the use of the absolute judgment
process.23 It is impossible to predict
whether sequential procedures would have prevented the misidentification
of Avery, but experts on the subject agree that "it is clear that the
sequential procedure produces a lower rate of mistaken identifications
in perpetrator-absent lineups."24
States Adopt the Procedures. In October 2001, New
Jersey became the first state to officially implement a statewide
identification procedure using blind testing and sequential
lineups.25 New Jersey Deputy Attorney
General Lori Linskey stated that recent findings from a two-year review
of the program demonstrated that law enforcement agencies implemented
the procedures without much difficulty. More than 80 percent of law
enforcement departments implemented blind testing either with no
difficulties or with only minor difficulties that were easily
overcome.26 Likewise, more than 85 percent
of departments had either no difficulties or only minor difficulties
implementing sequential lineups.27 In
October 2003, following New Jersey's lead, North Carolina became the
second state to support blind testing and sequential lineups after the
chief justice of the North Carolina Supreme Court recommended the use of
Hesitation to Change
In 2001 members of Wisconsin's legal community called on the criminal
justice system to follow the example set by New Jersey. In September
2001, the president of the State Bar of Wisconsin stated "we can
eliminate a great majority of misidentifications and not miss true
positive identifications by insisting on the use of ... a serial photo
lineup ... [and] [u]sing an officer other than the assigned detective to
conduct the photo lineup."29 Keith Findley,
codirector of the Wisconsin Innocence Project, added, "It's important
that we seize [this] opportunity, while we have it."30
Since then, however, the Wisconsin criminal justice and legal
communities have made only modest gains toward implementing the
statewide use of blind testing and sequential lineups. One notable event
occurred in May 2003, when Portage County Circuit Judge Fred Fleishauer
organized a half-day information session for Wisconsin judges on the
dangers and causes of misidentification.31
A second advancement occurred when approximately 150 law enforcement
officers attended private training sessions conducted by members of the
working group that had published the DOJ Eyewitness Guide.32 The Madison Police Department also hosted an
independent training session for law enforcement in April 2003.33
The results of these training sessions ultimately led to the Appleton
Police Department becoming one of the first departments in Wisconsin to
adopt a policy of sequential lineups. Appleton Police Department
Sergeant Dan Woodkey stated that after the department implemented
sequential procedures, he quickly recognized the value to the technique
because witnesses now make "an instant response" upon viewing a
suspect's photograph - an observation confirming that the witness used
the absolute judgment process.34 The
Madison Police Department recently became the first department in
Wisconsin to require both blind testing and sequential lineups.35
Despite this progress, the Wisconsin criminal justice system does not
have any statewide directive advocating for improved identification
techniques. The Wisconsin Department of Justice (DOJ) has not formed a
working group to examine the procedures implemented by the New Jersey
Department of Justice. Similarly, the Wisconsin court system has not
followed North Carolina's lead by creating a broad-based commission to
examine the use of blind testing and sequential lineups. The slow
response is all the more troubling in light of the Eugene Glenn case,
because investigators conducted his identification procedure after New
Jersey already had implemented blind testing and sequential lineups. One
can only speculate whether the victim would have incorrectly identified
Glenn had the identification procedure occurred in New Jersey.
The recent exoneration of Steven Avery has served as a spark to
awaken Wisconsin's criminal justice system. Shortly after the
exoneration, the Wisconsin DOJ responded by forming a Public Integrity
Unit to determine "whether or not any improprieties were had during the
court (sic) of the investigation or trial."36 In the final report, Attorney General Peg
Lautenschlager addressed many important matters surrounding the
Avery case, but she did not address broader issues regarding
how to improve identification procedures.37
The Wisconsin Legislature also responded, with Rep. Mark D. Gundrum
forming a taskforce "to review practices and procedures in Wisconsin's
criminal justice system which may need improving to help avoid the
wrongful conviction of innocent persons."38
Findley, codirector of the Wisconsin Innocence Project and a taskforce
member, stated that the taskforce plans to dedicate an entire session to
address eyewitness identification procedures before it completes its
work in spring 2004.39 Only time will
determine whether the taskforce serves as a basis for statewide
Winn S. Collins, U.W. 2003, is an
assistant DA in Outagamie County. He previously served as a project
assistant at the Wisconsin Innocence Project under the direction of
codirectors Keith A. Findley and John A. Pray. He has worked on several
cases involving the use of eyewitness evidence and lineup
The Avery exoneration likely will make the next year a decisive one
in determining whether the Wisconsin criminal justice system adopts a
statewide movement toward blind testing and sequential lineups. The
Avery case certainly serves as a powerful demonstration of the
need for reform, but Avery clearly is not the first - or last - person
exonerated after a conviction based upon misidentification. A person
would be naïve to believe that the Avery case alone will
change a system that did not change after previous exonerations.
Perhaps the better hope for reform lies in the lessons learned from
law enforcement departments already using blind testing and sequential
lineups. The preliminary findings from New Jersey's two-year review
demonstrate that an entire state can implement blind testing and
sequential procedures without too much difficulty. Likewise, the
Appleton and Madison police departments prove that such procedures work
well within the Wisconsin criminal justice system. As these departments
continue to use the improved techniques, the criminal justice system as
a whole may recognize that something as subtle as how an investigator
displays photographs has a tremendous impact on whether an innocent man
spends 18 years of his life behind bars.
1Winn S. Collins, Improving
Eyewitness Collection Procedures in Wisconsin, Wis. L. Rev. 529,
529 n.1 (2003).
2Id. at 534 (citing
Jessica Harry et al., Guilty Until Proven Innocent: A Study of Wrongful
Convictions in the Wisconsin Criminal Justice System (May 16, 2001)
3Hemauer v. State, 64 Wis.
2d 62, 218 N.W.2d 342 (1974).
4E. Michael McCann, Opposing
Capital Punishment: A Prosecutor's Perspective, 79 Marq. L. Rev.
649, 657 (1996).
5See State v. Hicks, 202
Wis. 2d 150, 549 N.W.2d 435 (1996).
6See generally State v.
Avery, 213 Wis. 2d 228, 570 N.W.2d 573 (Ct. App. 1997); State
v. Avery, No. 86-1831-CR, 1987 Wisc. App. LEXIS 4019 (Ct. App. Aug.
7Avery, 213 Wis. 2d at
244, 570 N.W.2d at 580.
8Avery, 1987 Wisc. App.
LEXIS 4019, at *4.
9Avery, 213 Wis. 2d at
244, 570 N.W.2d at 580; Tom Kertscher, A Mistake, Milwaukee J.
Sentinel, Sept. 21, 2003, at 1A.
10Kertscher, supra note
11A Milwaukee Journal
Sentinel investigation revealed that, at the time of the sexual
assault, the Manitowoc Police Department suspected Allen of several
sexual assaults that had occurred in the first half of 1985. The
Manitowoc Sheriff's Department never pursued Allen as a suspect. For
more information on what officers knew about Allen, see Tom Kertscher,
Police Were Watching Man Now Linked to Avery Case, Milwaukee J.
Sentinel, Oct. 20, 2003, at 1A.
12Avery, 1987 Wisc. App.
LEXIS 4019, at *13.
13Avery, 213 Wis. 2d at
232-33, 570 N.W.2d at 575-76.
14Wis. Stat. § 974.07
(2001-02). For more information on this statute, see Keith A.
Laws Reflect the Power and Potential of DNA, 75 Wis. Law. 20
15Jessica McBride, Reversed
Conviction Reflects Problems with Testimony, Milwaukee J. Sentinel,
Oct. 13, 2003, at 1A.
16Edwin M. Borchard,
Convicting the Innocent: Sixty-Five Actual Errors of Criminal
Justice 367 (Garden City Publ'g Co. 1932).
17U.S. Dep't of Justice,
Eyewitness Evidence: A Trainer's Manual for Law Enforcement 38 (Sept.
2003), avail. at www.ncjrs.org/nij/eyewitness/188678.pdf
(712K) [hereinafter Eyewitness Evidence II].
18Donald P. Judges, Two
Cheers for the Department of Justice's Eyewitness Evidence: A Guide for
Law Enforcement, 53 Ark. L. Rev. 231, 256 (2000).
19Gary L. Wells et al.,
Eyewitness Identification Procedures: Recommendations for Lineups
and Photospreads, 23 Law & Hum. Behav. 603, 603 (1998).
20U.S. Dep't of Justice,
Eyewitness Evidence: A Guide for Law Enforcement (Oct. 1999), avail.
(216K) [hereinafter Eyewitness Evidence I].
21Wells et al., supra
note 19, at 627.
22Id. at 639-40.
23Eyewitness Evidence II,
supra note 17, at 41.
24Wells et al., supra
note 19, at 639.
25Letter from John J. Farmer Jr.,
New Jersey Attorney General, to County Prosecutors et al. 1-2 (Apr. 18,
2001), avail. at www.state.nj.us/lps/dcj/agguide/photoid.pdf
26Telephone Interview with Lori
Linskey, Deputy Attorney General, New Jersey Department of Justice (Jan.
28Letter from I. Beverly Lake
Jr., Chief Justice, Supreme Court of North Carolina, to Scott Perry et
al., Director, Criminal Justice Training & Standards, North Carolina
Department of Justice (Oct. 9, 2003).
29Gerry Mowris, President's
Message: DNA Lessons Learned, 74 Wis. Law. 5 (Sept. 2001).
30Dianne Molvig, DNA
Evidence: Freeing the Innocent, 74 Wis. Law. 14 (Apr.
31Telephone Interview with Fred
Fleishauer, Portage County circuit court judge (Nov. 7, 2003).
32Telephone Interview with Paul
Carroll, Technical Working Group for Eyewitness Evidence, National
Institute of Justice, U.S. Dep't of Justice (Oct. 10, 2003).
33Telephone Interview with Cheri
Maples, captain of personnel and training, Madison Police Dep't (Oct.
34Interview with Dan Woodkey,
sergeant in the Investigative Services Unit, Appleton Police Dep't,
Appleton, Wis. (Nov. 7, 2003).
35Telephone Interview with Cheri
Maples, supra note 33.
36J.R. Ross, State to Review
Freed Man's Case; Officials Will Look at Whether Old Rape Case was
Handled Properly, Wis. St. J., Sept. 20, 2003, at B1.
37Memorandum from Peg
Lautenschlager, Wisconsin Attorney General, to Mark Rohrer, Manitowoc
County District Attorney (Dec. 17, 2003), avail. at www.doj.state.wi.us/news/rep121803_DCI.asp.
38Press Release, Mark D. Gundrum,
Gundrum Announces Avery Task Force: Group to Examine Practices and
Procedures Contributing to Wrongful Convictions (Dec. 10, 2003).
39Telephone Interview with Keith
A. Findley, codirector of the Wisconsin Innocence Project (Jan. 6,
2004); see also Telephone Interview with Mark D. Gundrum, chair
of the Committee on Judiciary, Wisconsin Assembly (Oct. 17, 2003).