3.2 Overview of Risks of Misidentification

A. Nature of the Problem

Research and studies. Eyewitness identifications play a major role in the charging and

conviction of criminal defendants, providing the basis for criminal charges against

approximately 77,000 people each year. Gary L. Wells & Elizabeth A. Olson, Eyewitness

Identification: Information Gain from Incriminating and Exonerating Behaviors, 8 J.

EXPERIMENTAL PSYCHOL.: APPLIED 155 (2002).

Scientific research, academic literature, and overturned convictions, however, have raised

concerns about the reliability of eyewitness identifications in general and cross-racial

identifications in particular. See also infra “State guarantee of due process” in § 3.4A,

Due Process (discussing recent cases from the U.S. Court of Appeals for the Fourth

Circuit and state supreme courts reviewing research raising concerns about eyewitness

identification evidence). The U.S. Supreme Court has observed that “the annals of

criminal law are rife with instances of mistaken identification.” United States v. Wade,

388 U.S. 218, 228 (1967); see also Perry v. New Hampshire, 565 U.S. __, 132 S. Ct. 716,

738–39 (2012) (Sotomayor, J., dissenting) (“[A] staggering 76% of the first 250

convictions overturned due to DNA evidence since 1989 involved eyewitness

misidentification.”). Experts believe that “eyewitness error is the leading contributing

factor in wrongful convictions in the United States.” See ELIZABETH F. LOFTUS ET AL.,

EYEWITNESS TESTIMONY: CIVIL AND CRIMINAL § 1-2 (5th ed. 2013) [hereinafter

YEWITNESS TESTIMONY]; see also EDWARD CONNORS ET AL., NATIONAL INSTITUTE OF

JUSTICE, CONVICTED BY JURIES, EXONERATED BY SCIENCE: CASE STUDIES IN THE USE OF

DNA EVIDENCE TO ESTABLISH INNOCENCE AFTER TRIAL (1996); C. Ronald Huff et al.,

Guilty Until Proved Innocent, 32 CRIME & DELINQUENCY 518 (1986); Innocence

Project, Facts on Post-Conviction DNA Exonerations, INNOCENCEPROJECT.ORG (last

visited Sept. 25, 2014).

A review of post-conviction DNA exonerations found that at least 40% of cases in which

the defendant was exonerated as a result of DNA evidence involved cross-racial

eyewitness identifications. Innocence Project,

Facts on Post-Conviction DNA

Exonerations, INNOCENCEPROJECT.ORG (last visited Sept. 25, 2014). As discussed in

greater detail below, researchers have found that White eyewitnesses are more likely than

Black eyewitnesses to make erroneous cross-racial identifications and that most cross-

racial identification errors made by White eyewitnesses are false positives—that is, the

erroneous identification of a person as the perpetrator. Christian A. Meissner & John C.

Brigham, Thirty Years of Investigating the Own-Race Bias in Memory for Faces: A Meta-

Analytic Review, 7 PSYCHOL. PUB. POL’Y & L. 3 (2001); see infra § 3.3, Cross-Racial

Impairment.

Case study: The role of race in the Shawn Massey case. Below are the reflections of

Duke University School of Law John S. Bradway Professor of Law and Wrongful

Convictions Clinic Co-Director James E. Coleman on the Shawn Massey case:

Raising Issues of Race in North Carolina Criminal Cases

Ch. 3: Eyewitness Identifications (Sept. 2014)

Shawn Massey spent twelve years in prison for crimes he did not commit, and the hard truth is that

race played a role in this miscarriage of justice. In 1999, a jury wrongfully convicted him of

kidnapping, armed robbery, and breaking and entering based on allegations that in May 1998 he

forced a woman and her children at gunpoint into their Charlotte apartment and took $60 from

them. The adult victim in this case, a White woman, described the perpetrator of the crime as a 5-

foot-9, 180-pound African American male who wore his hair “pulled back from his face and four

braids on the back of his head.” At trial, the victim clarified that she meant to describe cornrows.

However, the White police officers and prosecutors handling the case did not understand the

cornrow hairstyle, and this misunderstanding caused them to focus on the wrong suspect, create a

photo lineup with inappropriate fillers, and prosecute an innocent man. If the police officers

investigating the case had been familiar with the cornrow hairstyle or had focused on Mr. Massey’s

appearance before concluding that he was the assailant, they would have discovered that he was

both shorter and slimmer than the perpetrator, did not wear his hair in cornrows, did not have

enough hair for that hairstyle, and should have been excluded as a suspect.

At the time of the offense, Mr. Massey was 26 years-old, working a construction job, and living with

his grandmother. He had been charged with petty offenses, but had never been to prison. I became

involved with his case through my work in Duke Law School’s Wrongful Conviction Clinic. We decided

to investigate his claim of innocence because the only evidence in the case was the victim’s

eyewitness identification of Mr. Massey, and her identification was always conditioned on the

assailant having the cornrow hairstyle. Because of the general unreliability of eyewitness

identification evidence, we believe that any prosecution based entirely on eyewitness identification

evidence merits an innocence investigation.

In this case, the misidentification problem arose when the victim was shown a photo lineup including

Mr. Massey and a number of fillers. Neither Mr. Massey nor the fillers wore their hair in cornrows. In

the photo shown to the victim, Mr. Massey’s hair was very short. Asked if she saw the assailant in

the six- photograph lineup, the victim told police that Mr. Massey looked most like the man, except

that Mr. Massey did not have braids and the assailant’s hair was longer. This kind of relative

judgment is typical in cases of misidentification. Because Mr. Massey’s hair was not in cornrows,

however, the victim stated that she couldn’t be sure of her identification unless she saw Mr. Massey

in person.

The first time the victim saw Mr. Massey in person was in court. Just before the trial started, she told

the prosecutor that she had doubts about her identification, both because the defendant’s hair was

not in cornrows and because he appeared smaller than her attacker. Mr. Massey’s trial attorney was

not informed of the victim’s eleventh-hour doubts; we uncovered this Brady violation during our

innocence investigation. The only person who consistently maintained that Mr. Massey was the

perpetrator was the investigating officer, a White detective who relied exclusively upon an alleged

statement by Mr. Massey’s friend that Mr. Massey wore his hair “pulled back and 4 or 5 braids on

the back of his head.” At trial, however, this friend denied making such a statement, and denied that

Mr. Massey wore braids or that his hair was long enough to braid. Mr. Massey’s friends and family

members uniformly testified that he had never worn his hair in cornrows or long enough to braid,

but the statements did not persuade the jury.

While investigating the case, we discovered seven photographs of Mr. Massey in the District

Attorney’s files, one of which was taken in March 1998, nine weeks before the crime. The seven

photographs were taken over a nine-year period. In all of the photos, including the one taken in

March 1998, Mr. Massey’s hair was very short. We showed these photographs to professional

barbers familiar with African American hairstyles, who all agreed that Mr. Massey could not have

Raising Issues of Race in North Carolina Criminal Cases

Ch. 3: Eyewitness Identifications (Sept. 2014)

grown his hair long enough to wear cornrows by the time of the offense. At the time of Mr. Massey’s

prosecution, police and prosecutors did not appreciate the significance of these photographs

because they were not familiar with the cornrow hairstyle; as a result, they did not disclose the

photographs to the defense. Until the victim testified at trial, the police and prosecutor assumed

that cornrows were worn only on the back of the head and neck, and therefore concluded that

photographs showing only the front of Mr. Massey’s head did not exclude him as a suspect.

At the conclusion of our investigation, we sent a letter to the Mecklenburg County District Attorney

laying out the evidentiary basis of Mr. Massey’s claim of innocence, along with a description of the

Brady violations we uncovered. We included the photograph of Mr. Massey with short hair taken in

March 1998, prior to the offense, along with the opinions of professional barbers familiar with

African American hairstyles. In response to this evidence, District Attorney Peter Gilchrist filed an ex

parte motion to vacate the conviction and dismiss the charges, which was granted by a superior

court judge. Mr. Massey was released in 2010, approximately two years before the expiration of his

sentence.

This was a case of cross-racial identification, and race played an important role in the wrongful

conviction of Shawn Massey, beginning with the police and prosecutor’s misinterpretation of the

witness’s description of the perpetrator. This misinterpretation illustrates the importance of

obtaining an accurate description of the assailant from the very outset and making sure one

understands what is being described. Prompt investigation of identification procedures also provides

counsel with an opportunity to clarify any confusion or misunderstanding on the part of the police or

prosecutor, or to litigate pretrial the admissibility of a questionable identification. If the issue is not

raised until trial, jurors likely will see the dispute as one involving only the credibility of witnesses,

and, as some studies have shown, the race of a witness may affect jurors’ perceptions of whether

the witness is telling the truth. In Mr. Massey’s case, the jurors chose to believe the White police

officers and not the Black witnesses who all testified Mr. Massey did not wear braids and did not

have hair long enough to braid.

When the introduction of a flawed identification can’t be avoided, defenders should emphasize not

only the problems with cross-racial identification, but also specific contextual factors that may make

the witness identification less reliable. For example, in this case, opinions from professional barbers

familiar with African American hairstyles could have been introduced as evidence supporting Mr.

Massey’s claim that his hair could not have been worn in cornrows at the time of the offense.

Without such information, jurors are often misled by the common but mistaken belief that a victim

never forgets the face of his or her attacker.

Mr. Massey’s experience also underscores the recommendation of scholar Elizabeth Loftus, that

whenever possible, officers of the same race as the suspect should be involved in identification

procedures. If an African American police officer had constructed the lineup in this case or been part

of the investigation, he or she probably would have been familiar with the cornrow hairstyle and Mr.

Massey might never have been wrongfully convicted. If the White police officers or prosecutor had

queried the victim about her description, Mr. Massey also might never have been convicted. These

are flaws that can be fixed or, at a minimum, raised early in the case by the defense attorney.

B. Factors Affecting Eyewitness Identifications

Assuming that witnesses for the most part are earnestly trying to construct an accurate

account of a past event, why do eyewitness errors occur and why do they lead to

Raising Issues of Race in North Carolina Criminal Cases

Ch. 3: Eyewitness Identifications (Sept. 2014)

wrongful convictions? This section discusses the causes of eyewitness errors, while

subsection C, below, discusses jury perceptions of eyewitness testimony.

Generally. Researchers have concluded that eyewitness mistakes generally result from

the “normal and natural processes that occur whenever human beings attempt to acquire,

retain, and retrieve information.” EYEWITNESS TESTIMONY at § 2-1. Some studies suggest

that because of these natural processes eyewitnesses make correct identifications only

50% of the time. BRIAN L. CUTLER & STEVEN D. PENROD, MISTAKEN IDENTIFICATION 218

(1995) [hereinafter MISTAKEN IDENTIFICATION].

Factors affecting the accuracy of eyewitness identification fall into three broad

categories: problems of acquisition, problems of retention, and problems of retrieval.

EYEWITNESS TESTIMONY at § 2-2. Familiarity with these factors may assist counsel in

assessing the possibility of a misidentification and preparing for discussions with expert

witnesses. See infra § 3.6D, Expert Testimony.

Acquisition stage. In the acquisition stage, when an eyewitness memory is formed, the

accuracy of the witness’s perception may be affected by factors such as lighting

conditions, the duration of the event at issue, violence, stress, fear, age, sex, race, and

expectations. EYEWITNESS TESTIMONY at § 2-2. Research has found that cross-racial

eyewitness identifications are particularly susceptible to error at this stage. See infra §

3.3, Cross-Racial Impairment.

Retention stage. The retention stage encompasses the time between when the witness’s

memory was formed and when the witness describes the memory. Memory is malleable.

The accuracy of a memory during the retention stage is influenced by the length of the

retention interval and post-event experiences. For example, a suggestive pretrial

identification procedure, especially one involving cross-racial identification, could distort

an eyewitness’s memory of the actual event. “An eyewitness who is told that it is very

important for her to view a photoarray or lineup immediately is more likely to infer that

the investigators have identified the perpetrator than is an eyewitness who is told that she

could drop by the station whenever it is convenient for her to do so,” and this inference

may exert subtle pressure on the witness to provide a positive identification. MISTAKEN

IDENTIFICATION at 113–14. Similarly, the memory of an eyewitness who is given positive

feedback after participating in an identification procedure may be influenced by such

feedback. Id. at 186–90. One study concluded that in the trials of offenders exonerated by

DNA evidence between 1989 and 2010 that involved eyewitness testimony, 57% of the

eyewitnesses were initially uncertain of their eyewitness identifications, an indicator of

unreliability at the retention stage. BRANDON L. GARRETT, CONVICTING THE INNOCENT:

WHERE CRIMINAL PROSECUTIONS GO WRONG 64 (2011).

Retrieval stage. In the context of a criminal trial, the retrieval stage refers to the moment

when the witness testifies about the identification. When retrieving events from memory,

eyewitnesses are susceptible to the manner in which information is solicited. Frequent

review of the event, through questioning, identification procedures, and preparation to

testify, may artificially increase witness confidence when retrieving the memory at trial.

Raising Issues of Race in North Carolina Criminal Cases

Ch. 3: Eyewitness Identifications (Sept. 2014)

MISTAKEN IDENTIFICATION at 186–87. Witness confidence exerts a powerful influence on

jurors, but researchers have found that “eyewitness confidence is not a great or consistent

indicator of eyewitness accuracy.” EYEWITNESS TESTIMONY at § 3-12; see also

MISTAKEN IDENTIFICATION at 95; Siegfried Ludwig Sporer et al., Choosing, Confidence,

and Accuracy: A Meta-Analysis of the Confidence/Accuracy Relation in Eyewitness

Identification Studies, 118 PSYCHOL. BULL. 315 (1995); Neil Vidmar et al., Rethinking

Reliance on Eyewitness Confidence, 94 JUDICATURE 16 (2010).

Illustration: The wrongful conviction of Ronald Cotton illustrates the problems that may

arise with the acquisition, retention, and retrieval stages of eyewitness identifications.

Ronald Cotton was convicted of the rape of Jennifer Thompson in Burlington, North

Carolina and sentenced to life plus 54 years based on Thompson’s eyewitness

identification. Ronald Cotton served 10.5 years in prison before he was exonerated by

DNA evidence. Jennifer Thompson, a White woman, made a cross-racial identification of

Ronald Cotton, a Black man. Ms. Thompson later observed that she “studied every single

detail on the rapist’s face . . . [and] was going to make sure he was put in prison.”

Jennifer Thompson, I Was Certain, But I Was Wrong, N.Y.

TIMES, June 18, 2000. In the

time between the crime and the trial of Mr. Cotton, Ms. Thompson grew more and more

confident in her mistaken identification. She reflected that, when she saw Ronald Cotton

in a photo array, she was “completely confident” that he was the assailant; and when she

picked him out of a lineup, she was “sure . . . [she] had picked the right guy.” Id.

However, her confidence did not correspond with accuracy. “From description, to

creating an Identikit, to reviewing a photo array, to identifying the wrong man in a lineup

and in court—each step unconsciously became a process of picking the individual most

resembling the prior step, not most resembling the perpetrator.” Joseph F. Savage Jr. &

James P. Devendorf, Conviction After Misidentification: Are Jury Instructions a

Solution?, THE CHAMPION, June 2011, at 30, n.7 (discussing factors contributing to

Jennifer Thompson’s misidentification of Ronald Cotton).

Estimator and system variables. Variables that affect eyewitness identifications can be

categorized as either “estimator” or “system” variables. See Gary L. Wells, Applied

Eyewitness Testimony Research: System Variables and Estimator Variables, 36 J.

PERSONALITY & SOC. PSYCHOL. 1546 (1978) (using these categories for the first time).

Familiarity with these variables may assist counsel in assessing the identification and

discussing it with an expert. Additionally, counsel may use these variables to structure

cross-examination in order to show how a given variable, such as the presence of a gun,

affected the reliability of eyewitness testimony.

Some variables that bear on the accuracy of eyewitness identifications are within the

control of the criminal justice system while others are not. Estimator variables occur

before the case enters the criminal justice system and include factors such as: the lighting

at the time of the event, the duration of the witness’s exposure to the perpetrator, and the

race of the witness and perpetrator. In contrast, system variables are or may be within the

control of the criminal justice system, such as the construction of a lineup, the

questioning of the eyewitness, and the method in which the lineup is presented. Examples

of system and estimator variables are as follows. (The list is reproduced verbatim from

Raising Issues of Race in North Carolina Criminal Cases

Ch. 3: Eyewitness Identifications (Sept. 2014)

Joseph F. Savage & James P. Devendorf, Conviction After Misidentification: Are Jury

Instructions a Solution?, THE CHAMPION, June 2011, at 30, 31, except that examples of

variables that have been added are set off in brackets.)

1. Wording of questions: The wording of questions posed to an eyewitness can affect the

witness’s testimony about an event. [For example, if an eyewitness is asked how

many minutes she was able to observe the assailant, she may be more likely to

estimate that the event took a matter of minutes, rather than seconds.]

2. Lineup instructions: The instructions given to the witness at a lineup can affect the

witness’s willingness to make an identification. [For example, studies have shown

that instructions implying that the suspect is in fact in the photo array “increas[e] the

likelihood that the eyewitness will make a positive—though not necessarily correct—

identification.” MISTAKEN IDENTIFICATION at 115–23.]

3. Confidence malleability: Factors unrelated to identification accuracy can influence a

witness’s confidence. [For example, a witness’s confidence may rise when she is told

that another witness has identified the same person. MISTAKEN IDENTIFICATION at

186–90.]

4. Mugshot-induced bias: Exposure to mugshots of a suspect increases the likelihood

that the witness will later choose that suspect in a lineup.

5. Post-event information: Testimony of eyewitnesses about an event often reflects not

only what they actually saw but information they obtained after the event.

6. Child witness suggestibility: Young children are more vulnerable than adults to

interviewer suggestion, peer pressures, and other social influences.

7. Attitudes and expectations: A witness’s perception and memory of an event may be

affected by his or her attitudes and expectations. [For example, if a person is attacked

in the dark in an area primarily frequented by Latinos, he or she may be more likely

to believe that his or her attacker was Latino.]

8. Hypnotic suggestibility: Hypnosis increases suggestibility to leading and misleading

questions.

9. Alcohol intoxication: Alcohol intoxication impairs an eyewitness’s later ability to

recall persons and events.

10. Cross-race bias: Eyewitnesses are more accurate when identifying members of their

own race than members of other races.

11. Weapon focus: The presence of a weapon impairs an eyewitness’s ability to

accurately identify the perpetrator’s face.

Raising Issues of Race in North Carolina Criminal Cases

Ch. 3: Eyewitness Identifications (Sept. 2014)

12. Accuracy-confidence correlation: A witness’s confidence is not a good predictor of

his or her identification accuracy.

13. Forgetting curve: The rate of memory loss for an event is greatest right after the

event and then levels off over time.

14. Exposure time: The less time an eyewitness has to observe an event, the less well he

or she will remember it.

15. Presentation format: Witnesses are more likely to misidentify someone by making a

relative judgment when presented with a simultaneous (as opposed to sequential)

lineup.

16. Unconscious transference: Eyewitnesses sometimes identify as a culprit someone

they have seen in another situation or context.

Defense attorneys can attempt to mitigate the influence of system variables by, for

example, ensuring that law enforcement officers present individuals or photographs to

witnesses sequentially rather than simultaneously. See infra § 3.5, Eyewitness

Identification Reform Act. Additionally, counsel can educate jurors about the impact of

system and estimator variables on the reliability of eyewitness testimony through expert

testimony, cross-examination of eyewitnesses, and jury instructions concerning factors

influencing eyewitness reliability. See infra § 3.6, Procedures for Challenging Eyewitness

Identification Evidence.

C. Jurors’ Perceptions of Eyewitness Identification

Concerns about eyewitness identification, discussed in the preceding section, are

compounded by the weight jurors may give such testimony. Studies have concluded that

jurors tend to overestimate the reliability of eyewitness testimony. See, e.g., Tanja Rapus

Benton et al., Eyewitness Memory is Still Not Common Sense: Comparing Jurors, Judges

and Law Enforcement to Eyewitness Experts, 20

APPLIED COGNITIVE PSYCHOL. 115

(2006); Richard S. Schmechel et al., Beyond the Ken? Testing Jurors’ Understanding of

Eyewitness Reliability Evidence, 46 J

URIMETRICS 177 (2006); Perry v. New Hampshire,

___ U.S. ___, ___, 132 S. Ct. 716, 739 (2012) (Sotomayor, J., dissenting) (observing that

“jurors routinely overestimate the accuracy of eyewitness identifications”). In 2004, the

Public Defender Service for the District of Columbia surveyed nearly 1,000 potential

jurors about eyewitness identification. The researchers concluded that the survey

members often underestimated the difficulties eyewitnesses experience in making cross-

racial identifications, the impact of stress on memory, and the ways in which police

procedures may undermine eyewitness accuracy. Timothy P. O’Toole et al., District of

Columbia Public Defender Survey, THE CHAMPION, Apr. 2005, at 28; see also, Richard S.

Schmechel et al., Beyond the Ken? Testing Jurors’ Understanding of Eyewitness

Reliability Evidence, 46 JURIMETRICS 177 (2006); EYEWITNESS TESTIMONY at § 6-6.

Raising Issues of Race in North Carolina Criminal Cases

Ch. 3: Eyewitness Identifications (Sept. 2014)

One reason that lay people misunderstand the reliability of eyewitness identification is

that much of what experts now know about memory and eyewitness testimony is counter-

intuitive. For example, even though experts now recognize that eyewitness confidence is

not reliably correlated with accuracy, “it would seem logical that a more certain witness

would be a more accurate one, [and therefore] it would be surprising if jurors understood

the relationship between confidence and accuracy as a matter of common sense.”

Timothy P. O’Toole et al., District of Columbia Public Defender Survey, THE CHAMPION,

Apr. 2005, at 28, 29. As one court observed, “while science has firmly established the

inherent unreliability of human perception and memory, this reality is outside the jury’s

common knowledge, and often contradicts jurors’ commonsense understandings. To a

jury, there is almost nothing more convincing than a live human being who takes the

stand, points a finger at the defendant, and says, ‘That's the one!’” United States v.

Brownlee, 454 F.3d 131, 142 (3d Cir. 2006) (quotations and citations omitted). See also

Phillips v. Allen, 668 F.3d 912, 916 (7th Cir. 2012) (stating that “nothing is obvious about

the psychology of eyewitness identification” and “most people’s intuitions on the subject

of identification are wrong”).

Practice note: Counsel may file a motion in limine requesting the court to preclude the

prosecutor from drawing a correlation between witness confidence and witness accuracy.

See Lisa Steele, Trying Identification Cases: An Outline For Raising Eyewitness ID

Issues, THE CHAMPION, Nov. 2004, at 8.

Raising Issues of Race in North Carolina Criminal Cases