CHAPTER 2 EYEWITNESS IDENTIFICATION
| Jurisdiction | North Carolina |
After reading this chapter, you should be able to:
• Distinguish, in general, which variables related to eyewitness identifications are under the control of the criminal justice system and which are not.
• Understand the implications for an accurate identification when the eyewitness and the culprit are of different races.
• Appreciate how the design and administration of suspect lineups influence identification accuracy.
• Appreciate the malleability of eyewitness confidence and the fallibility of human memory and the implications of both for accurate identifications.
• Understand the extent (and limitations) of the Sixth Amendment right to counsel at corporeal and noncorporeal identifications.
• List the reliability criteria set forth by the U.S. Supreme Court for determining the admissibility of identification testimony.
• Discuss the recommended changes to police identification procedures.
In one night in July 1984, a man broke into two apartments in Burlington, North Carolina, and committed sexual assault. One of the victims was 22-year-old college student Jennifer Thompson. The assailant had cut her phone line, searched through her items, and raped her at knife-point. Unable to escape, Thompson made a conscious effort to study the man's face and voice so that, if she survived, she would be able to help the police identify him. After the attack, Jennifer offered to make the man a drink and, when he agreed, fled the apartment.
Thompson worked with police to develop a composite sketch of her attacker, and on August 1, 1984, police arrested an African American man named Ronald Cotton. Cotton, who was the same age as Jennifer and had a prior record, was placed in a photo array, in which she identified him as the perpetrator. Shortly thereafter, she viewed a live lineup, and again chose Ronald Cotton. Thompson was told she had picked the same person both times, and she was sure that she had identified the man who had attacked her.
Cotton was put on trial for rape and burglary. The prosecution's evidence included a flashlight from his home that resembled the one used by the attacker and rubber from his sneaker that was consistent with rubber found at the crime scene, but the most important evidence was Thompson's identification of Cotton. In January 1985, Cotton was convicted of one count each of rape and burglary. The appeals court overturned the conviction because the second victim had identified a different person in the lineup, but that evidence was not allowed in the first trial. Before Cotton's second trial, Bobby Poole, a man who was in prison with Cotton for committing a series of brutal rapes, told another inmate that he actually committed the attacks, not Cotton. However, the trial judge did not allow this evidence to be admitted at trial, and in November 1987, by which time the second victim had also identified Cotton as the attacker, he was convicted for both assaults. Ronald Cotton was sentenced to life in prison plus 54 years. Cotton appealed his case, but his lawyer did not challenge the judge's failure to allow the other inmate's confession to be used, and lost the appeal.
In 1994, two new lawyers took over his defense and filed for relief on two grounds. First, they argued that Cotton's appellate lawyer was inadequate. Second, they filed a motion for DNA testing, which was granted in October 1994. The following spring, the Burlington Police Department turned over all of the evidence from the case, including samples of the attacker's semen for DNA testing. The sample from one of the attacks was deteriorated and could not generate a conclusive result. However, the DNA from the second attack was tested, and did not match Cotton. The defense then requested that the test results be run through the State Bureau of Investigation's DNA database, and they found a match for Poole, the second inmate who had earlier confessed to the attacks.
The DNA test results were reported in May 1995. The defense and District Attorney submitted a motion to dismiss all charges against Cotton. In June, he was officially cleared of all charges and released from prison, and in July, was pardoned by North Carolina governor James Hunt. Cotton had served more than 10 years in prison for crimes he did not commit and received $110,000 from the state.
Picking Cotton: Our Memoir of Injustice and Redemption (2009) — a memoir co-authored by Jennifer Thompson and Ronald Cotton, with Erin Torneo. Thompson was the rape victim who wrongly identified Cotton in Burlington, North Carolina. Cotton spent more than a decade in prison before being exonerated through DNA testing. The book helps readers understand the case from the perspective of those who were involved in it.
I Didn't Do It, Episode 3, Ronald Cotton: Shadow of Doubt—I Didn't Do It is a documentary series that began airing in 2012 and was produced by Canadian production company Lively Media for the Investigation Discovery cable channel. Episode 3 recounts the Cotton case through dramatic recreations and interviews with Cotton, Thompson, and others involved in the case.
What Jennifer Saw (1997) — a PBS Frontline documentary on the case, featuring interviews with Thompson and Cotton, as well as police involved in the case and leading researchers who discuss the psychology of identification errors.
Eyewitness errors, such as the one in the case of Jennifer Thompson and Ronald Cotton, may not be as rare as one might hope. According to the National Registry of Exonerations, approximately 30% of the first 2,038 exonerations through May 2017 involved a mistaken eyewitness identification. Furthermore, the Innocence Project reports that more than 70% of the first 350 DNA-based exoneration cases involved an eyewitness misidentification, many of which were similar to the Cotton case in which a sexual assault victim wrongly identified the attacker. These numbers suggest that such errors may not be atypical, but rather that eyewitnesses are less reliable than one might think.
Knowledge about eyewitness errors is not new. In fact, psychologist Hugo Munsterberg (1908) wrote about how unreliable witnesses may be and how they can have a negative effect on the outcomes of trials in the early 1900s, and systematic research since the 1970s has uncovered many of the factors that generate mistaken identifications. Some of these factors—for example, whether a crime occurred in the dark of night or under a brilliant sun, whether a witness has 20/20 vision and was standing five feet from the offender or instead is hopelessly nearsighted and was 50 feet away, or whether a suspect is quickly located and an identification attempt is made within minutes of the crime or several months pass before an attempted identification is possible—are beyond the control of the justice system. Such factors are called "estimator variables." On the other hand, some factors may be under the control of the justice system. Examples of so-called "system variables" (Wells, 1978) include the size of a lineup and the number of pictures used in a photo array, what the police say to a witness before and after an attempted identification, and whether a defense lawyer is allowed to be present during an identification procedure. Both estimator and system variables are vital to understanding how and why witnesses make mistakes and what can be done to improve witness accuracy.
Estimator Variables
Estimator variables are factors that are beyond the control of the criminal justice system. Many characteristics of a situation can influence the likelihood of a witness making an error, but cannot be controlled by those who work in the system. These characteristics include obvious visual issues such as the lighting, the distance from which a witness viewed an event, or the length of time a witness was exposed to the perpetrator. For instance, a witness who sees a crime at night from 100 yards away may be more likely to make an error than one who witnesses a crime up close in broad daylight. But other details of a criminal event may alter a witness's ability to accurately remember and recall information in important ways. For example, much research has explored the "weapon focus effect" which suggests that in crimes involving a weapon, witnesses may have difficulty remembering characteristics of the perpetrator because their focus is on the weapon rather than on remembering details of the offender's appearance. Indeed, while some research has not found evidence of a weapon-focus effect, several studies have found that the presence of a weapon may negatively affect the accuracy of identifications. Importantly, many psychologists have also found that the presence of a weapon negatively affects the accuracy of a witness's descriptions of the perpetrator (Carlson, Dias, Weatherford, & Carlson, 2017; Kocab & Sporer, 2016).
The presence of a weapon along with other situational characteristics, such as the type and severity of the crime, may increase the fear a witness feels and thus the stress they are under. A number of research studies have found that witnessing violent events can affect human memory (Deffenbacher, Bornstein, Penrod, & McGorty, 2004).
In addition to characteristics of the crime and the situation, individual characteristics of the witness may decrease the reliability of an identification. One such factor is the witness's age. A recent review of psychological studies on this topic found that, when the suspect was present in the lineup, young adults were more accurate in their identifications than both children and older adults. Importantly, they were also better at not making identifications in lineups when the suspect was not present; in other words, they were less likely to make a false identification when the culprit was not included in the lineup. The review also found similar...
To continue reading
Request your trial