Ronald Cotton and Bobby Poole: DNA Exoneration and ID Reform

The Ronald Cotton and Bobby Poole wrongful conviction case is the story of a North Carolina man who spent more than ten years in prison for two 1984 rapes he did not commit, convicted on the confident but mistaken identification of one of the victims, and finally exonerated in 1995 when DNA testing matched the crime scene evidence to another man, Bobby Poole. It became one of the most studied wrongful convictions in American history and drove sweeping changes in how police handle eyewitness identifications.¹

The 1984 Burlington Assaults

Around 3 a.m. on July 29, 1984, an intruder broke into the apartment of 22-year-old Jennifer Thompson-Cannino in Burlington, North Carolina, and sexually assaulted her. She escaped through her back door and got help. Later that same night, a second woman was assaulted in the same neighborhood. Police collected physical evidence from both apartments and took descriptions from the survivors.¹

During the attack, Thompson-Cannino made a conscious effort to study her attacker’s face so she could identify him afterward. That effort left her confident, but confidence is a poor proxy for accuracy, especially across race. Thompson-Cannino is white; Cotton is Black. Research on cross-racial identification has consistently shown that people are less accurate at recognizing faces of another race, and that a witness’s in-the-moment confidence is a particularly weak predictor of accuracy in those situations.²

How the Identification Went Wrong

Investigators zeroed in on Cotton and showed Thompson-Cannino a photo array. She studied it for several minutes before pointing to his photograph and saying, “I think this is the guy.” The detective asked, “You’re sure?” She replied, “Positive.” She then asked, “Did I do OK?” and was told, “You did great.”³

A physical lineup followed. Thompson-Cannino told police Cotton “looks the most like him.” When she said she was sure, the detective told her, “It’s the same person you picked from the photos.” She later described feeling a “huge amount of relief” at hearing that.³ Those exchanges are now a textbook illustration of confirmatory feedback: small cues from investigators that harden a tentative pick into unshakable certainty. By trial, her initial “I think” had become absolute conviction. The second victim viewed the same photo lineup and the same physical lineup and could not identify her attacker either time, saying she only remembered he had a flashlight.¹

Two Trials and a Rejected Alternative Suspect

Cotton went to trial in Alamance County in January 1985. Thompson-Cannino identified him with complete confidence. His family members testified he had been home asleep. There was no physical evidence tying him to either assault. The jury convicted him of one count of rape and one count of burglary, and the court sentenced him to life plus fifty years.⁴

The Supreme Court of North Carolina reversed the conviction and ordered a new trial. At the retrial in November 1987, prosecutors charged Cotton with both rapes. This time the second victim also identified him in court, saying she had been too afraid to do so earlier.¹

By then Cotton’s defense had a name: Bobby Poole. Poole had been arrested for two other sexual assaults, and his blood type matched a sample found in the second victim’s apartment. Poole was serving time at the same prison as Cotton, the two men bore a striking resemblance, and Poole had reportedly told other inmates that he was the one who committed the Burlington attacks. With the jury excused, Poole was brought into the courtroom. Both victims looked at him and said he was not their attacker. Thompson-Cannino later wrote that when she saw Poole, she was certain she had never seen him before.⁵ Because neither victim identified him, the judge refused to let the defense present Poole as an alternative suspect. The jury convicted Cotton of both rapes and two counts of burglary.¹

While Cotton served his sentence, the consequences of misidentifying the attacker kept accumulating. In the nine months after Cotton was locked up, Poole committed at least twenty more crimes, including robberies, burglaries, and another rape, because police had stopped looking for the real offender.

The DNA Testing That Freed Cotton and Matched Poole

In the spring of 1995, Cotton’s defense team requested DNA testing on the biological samples preserved in the original 1984 rape kits. The Burlington Police Department turned over the physical evidence. The lab used polymerase chain reaction testing, which can generate a reliable DNA profile from a sample as small as a pinhead and had not existed when Cotton was first tried in 1985.⁶

The results were unambiguous. The profile from the crime scene evidence did not match Cotton. Compared to Bobby Poole’s DNA, it was a perfect match. On June 30, 1995, the district attorney and the defense jointly moved to dismiss all charges. The court vacated the convictions, and Cotton walked out after more than ten and a half years in prison.¹ Poole died in prison in 1998, three years after the DNA match.

Pardon and State Compensation

Dismissal cleared the charges but did not formally declare Cotton innocent. For that, he needed a pardon. On July 12, 1995, North Carolina Governor James B. Hunt Jr. granted him a pardon of innocence.¹ Under North Carolina law, a pardon of innocence is what opens the door to state compensation for wrongful incarceration. The state’s compensation statute now provides $50,000 for each year of wrongful imprisonment, up to a $750,000 cap.⁷ Cotton received $110,000 under the law in effect at the time of his release.

What the Case Changed About Eyewitness Identification

Cotton’s case became a catalyst for rethinking police identification procedures. Eyewitness misidentification has been a factor in 69 percent of convictions later overturned by DNA evidence, making it the single leading cause of wrongful convictions in the United States.⁸

The confirmatory feedback Thompson-Cannino received is now recognized as one of the most dangerous contaminants in the process. When a detective tells a witness she “did great,” or confirms she picked the same person twice, her confidence inflates permanently. By trial the jury never sees the original hesitation. Current best practices call for taking a confidence statement from the witness at the moment of identification and avoiding any feedback that could alter it later.⁹

North Carolina became one of the first states to enact legislation governing how identification procedures must be conducted. Other jurisdictions have adopted rules requiring that the officer administering a lineup or photo array not know which person is the suspect, reducing the risk that body language or offhand comments influence the witness.⁹ Implementation remains uneven. Recent survey data indicates more than half of responding law enforcement agencies still are not aware of the strong relationship between a witness’s initial confidence and identification accuracy, the very gap that put Ronald Cotton in prison and let Bobby Poole keep offending.

1
Innocence Project. Ronald Cotton
2
National Center for Biotechnology Information. A Cross-Race Effect in Metamemory: Predictions of Face Recognition
3
National Academies Press. Identifying the Culprit: Assessing Eyewitness Identification
4
The Marshall Project. Eyewitness Testimony Is Unreliable… Or Is It?
5
The New York Times. I Was Certain, but I Was Wrong
6
National Institute of Justice. What Every Law Enforcement Officer Should Know About DNA Evidence – Polymerase Chain Reaction
7
North Carolina General Assembly. North Carolina Code Chapter 148 Article 8 – Compensation to Persons Erroneously Convicted
8
Innocence Project. How Eyewitness Misidentification Can Send Innocent People to Prison
9
National Policing Institute. Eyewitness Identification in Law Enforcement: Bridging the Gap Between Science, Policy, and Practice