She Was Certain. DNA Proved the Identification Wrong: The Ronald Cotton Case

Jennifer Thompson-Cannino had every reason to believe her memory.

In the early morning hours of July 29, 1984, a man broke into the apartment of the 22-year-old college student in Burlington, North Carolina, and sexually assaulted her. During the attack, Thompson-Cannino made a deliberate effort to study the man’s face and physical features. She later explained that she wanted to remember him well enough to identify him if she survived. Her determination to help police find the person responsible would eventually become one of the best-known examples of how an eyewitness can be sincere, attentive, and completely confident, yet still identify the wrong person.

Ronald Cotton was 22 years old when he became the focus of the investigation. He would spend more than ten years in prison before DNA testing established that he was not Thompson-Cannino’s attacker and identified another man as the perpetrator.

Unlike the Jermaine Hudson case previously discussed, there was no fabricated crime and no witness who knowingly lied. Thompson-Cannino had been the victim of a violent sexual assault. She genuinely believed Cotton was the man who attacked her. Understanding how that belief developed is one of the reasons the case remains so important to the study of eyewitness identification.

The Investigation

After the assault, Thompson-Cannino worked with police to create a composite sketch and described her attacker as a tall, light-skinned Black man. Police eventually showed her photographs of possible suspects, including Cotton.

When she viewed a six-photograph lineup, she did not immediately select him. According to accounts of the identification procedure, Thompson-Cannino initially narrowed her choice to two photographs. After looking again, she selected Cotton.

The response she received after making that identification is significant. Investigators gave her positive feedback indicating that she had done well. Research conducted in the years since Cotton’s case has shown why seemingly innocent feedback after an identification can be problematic. Once a witness receives confirmation that police believe the person selected is the suspect, the witness’s confidence in that selection may increase even though the underlying memory has not become more accurate.

The following day, Thompson-Cannino viewed a live lineup. Cotton was the only person from the earlier photographic lineup who also appeared in the live lineup. She again had some difficulty deciding between two people before identifying Cotton. Police then confirmed that she had selected the same person she had chosen from the photographs.

By the time the case reached trial, her confidence had grown considerably.

Another sexual assault had occurred in the same neighborhood on the same night. That victim viewed the same photographic lineup but did not identify Cotton. Police nevertheless arrested him on July 31, 1984. During a search of his home, investigators found a flashlight resembling one described in connection with the crimes and tennis shoes containing foam similar to a piece of foam recovered from Thompson-Cannino’s apartment.

The similarities gave prosecutors evidence to argue alongside the eyewitness identification, although neither item was uniquely connected to Cotton.

The First Trial

Cotton went to trial in January 1985 for the assault on Thompson-Cannino. She identified him in court as the man who had attacked her.

There were reasons for the defense to question the identification. Several similar incidents had occurred within hours of one another in the area, and another victim had viewed the same photo lineup and identified a different man. Cotton’s attorneys attempted to introduce that identification, but the trial court ruled that it was not relevant and excluded it.

Cotton’s relatives testified that he had been home asleep when Thompson-Cannino was assaulted. Prosecutors relied on her identification, along with the flashlight and foam evidence recovered during the investigation.

On January 17, 1985, the jury convicted Cotton of rape and burglary. He was sentenced to life in prison plus 50 years.

Two years later, the North Carolina Supreme Court reversed the convictions and ordered a new trial. The court concluded that the defense should have been permitted to introduce evidence showing that another victim had identified someone other than Cotton.

The new trial gave Cotton another opportunity to establish his innocence. It also produced an extraordinary development.

The Man Who Actually Committed the Crime Appeared in Court

By the time of Cotton’s second trial in November 1987, another man had attracted the defense’s attention. His name was Bobbie Poole.

Poole had been arrested for other sexual assaults. His blood type was also consistent with a drop of blood recovered from the apartment of the second woman who had been assaulted on the same night as Thompson-Cannino.

There was another reason to investigate Poole. While incarcerated, he reportedly told other prisoners that Cotton was serving time for crimes Poole had committed.

The defense attempted to present Poole as an alternative suspect. Outside the jury’s presence, Poole was brought into the courtroom so the victims could see him.

Thompson-Cannino did not recognize him as her attacker.

Neither did the second victim.

By this point, the second woman had also come to identify Cotton as the man who assaulted her, although she had failed to identify him shortly after the crime. She explained that fear had prevented her from identifying him earlier.

The court ultimately refused to allow the defense to present Poole to the jury as an alternative suspect.

Cotton was convicted again, this time of both rapes and two burglaries. His sentence was life in prison plus 54 years.

The result is difficult to view with the benefit of what was discovered later. The actual perpetrator had been physically brought into the courtroom, yet the eyewitnesses rejected him and remained convinced that Cotton was responsible.

Their certainty did not reflect dishonesty. It reflected what they had come to remember.

DNA Finally Answered the Question

Cotton continued to maintain his innocence. In 1995, his attorneys obtained access to biological evidence that had been preserved from both sexual assault investigations and requested DNA testing.

The results excluded Ronald Cotton.

Investigators then compared the unknown DNA profile with records maintained by the North Carolina State Bureau of Investigation. The DNA matched Bobbie Poole.

Poole was already serving a life sentence for other sexual assaults. He had also reportedly admitted to fellow prisoners that he had committed the crimes for which Cotton had been convicted.

The scientific evidence finally provided an answer that eyewitness certainty had not.

Cotton was not the attacker.

The district attorney and Cotton’s attorneys jointly sought dismissal of the charges. On June 30, 1995, Cotton was released after serving more than ten years in prison. Poole subsequently pleaded guilty to both rapes and received an additional 70-year sentence. North Carolina Governor James B. Hunt Jr. formally pardoned Cotton in July 1995.

Cotton eventually received compensation from the state for his wrongful conviction. He also took a job working in the warehouse at LabCorp, the same company whose DNA testing had helped establish his innocence.

Jennifer Thompson-Cannino Had to Confront What Had Happened

DNA testing resolved the legal question, but it created another difficult reality for Thompson-Cannino.

She had not casually identified Cotton. She had survived a terrifying crime and intentionally concentrated on her attacker because she wanted the person responsible to be caught. She had selected Cotton from photographs, identified him again in a live lineup, pointed him out in court, and remained convinced of his guilt through two trials.

She had also looked at Bobbie Poole, the actual perpetrator, and said he was not the man who attacked her.

Learning that Cotton was innocent required Thompson-Cannino to reconcile her certainty with scientific evidence proving that her memory was wrong.

Several years after Cotton’s exoneration, the two met in person. Thompson-Cannino apologized for the role her identification had played in his conviction. Cotton accepted her apology.

Their relationship continued, and they eventually became close friends. Rather than avoiding the history that connected them, they began speaking publicly about it. Together, they traveled around the country educating law enforcement officers, attorneys, lawmakers, students, and others about eyewitness identification and wrongful convictions. They later co-authored Picking Cotton: Our Memoir of Injustice and Redemption, which recounts the case from both of their perspectives.

Their advocacy also became part of a broader effort to improve identification procedures. North Carolina enacted the Eyewitness Identification Reform Act in 2007, requiring procedures intended to reduce the risk of mistaken identifications.

How Certainty Can Grow After an Identification

Cotton’s case has since been examined extensively by researchers studying human memory and eyewitness identification. The National Academies of Sciences included the case in its major report on eyewitness evidence.

One of the most important features is the difference between Thompson-Cannino’s initial uncertainty and the confidence she later expressed.

Her first identification was not instantaneous. She considered more than one photograph before choosing Cotton. During the subsequent live lineup, she again considered more than one person before selecting him. Each identification was followed by feedback that reinforced the selection she had made.

By the time Thompson-Cannino testified in court, the uncertainty present during the original identification was no longer apparent.

Modern research on eyewitness identification has led to recommendations that law enforcement document a witness’s confidence in the witness’s own words at the time of the initial identification. Researchers also recommend procedures designed to prevent an administrator from unintentionally influencing the witness, standardized instructions explaining that the perpetrator may not be present, and video recording of identification procedures.

Those safeguards exist because memory can change. Later confidence does not necessarily tell us how certain the witness was when the identification was first made.

Cotton’s case provides a particularly clear example. Thompson-Cannino did not become more accurate as she became more certain. Her confidence grew around an incorrect identification.

A Defense Investigation Has to Go Beyond the Identification

When reviewing an eyewitness case, the existence of an identification should be the beginning of the inquiry rather than the end of it.

A criminal-defense investigator should examine the witness’s earliest description before looking only at the identification eventually made. The procedures used during photo arrays and live lineups should be reviewed, along with the instructions provided to the witness, the people included as fillers, any feedback given afterward, and the witness’s original level of confidence.

Earlier failures to identify someone can be just as important as later identifications. Changes in a witness’s account deserve attention, particularly when a case has been pending for years and the witness has repeatedly seen photographs of the defendant, attended hearings, spoken with investigators, or discussed the case with other people.

Independent evidence should also be considered alongside the identification. In Cotton’s case, another victim initially identified someone else. Bobbie Poole emerged as an alternative suspect before the second trial, and his blood type was consistent with evidence recovered from one of the crime scenes. Poole reportedly admitted responsibility while incarcerated. Even after those developments, the eyewitnesses remained convinced Cotton was the attacker. Re-interviewing witnesses can reveal changes in memory, confidence, or detail that may not be obvious from the original report.

DNA eventually settled the question because biological evidence happened to exist and had been preserved.

Many criminal cases do not offer that luxury.

There may be no DNA capable of proving conclusively that an eyewitness is right or wrong. In those cases, careful investigation becomes even more important. Scene conditions, lighting, distance, the duration of an encounter, earlier descriptions, lineup procedures, witness statements, surveillance video, phone records, timelines, physical evidence, and alternative suspects may provide the only available means of testing an identification.

Ronald Cotton’s case does not teach us that crime victims should not be believed or that eyewitness testimony has no value. Thompson-Cannino was telling the truth as she understood it. Her identification was simply wrong.

That is precisely why corroboration is necessary.

A dishonest witness can send an investigation in the wrong direction, as Jermaine Hudson’s case demonstrated. Ronald Cotton’s case presents a more complicated problem because everyone involved may be acting in good faith. A victim can sincerely want justice, investigators can sincerely believe they have found the right suspect, and a jury can sincerely believe the testimony presented in court.

The wrong person can still be convicted.

Ronald Cotton lost more than ten years of his freedom before DNA provided the independent evidence necessary to correct the mistake. Thompson-Cannino spent years dealing with the knowledge that the man she confidently identified was innocent. The two eventually turned an experience that harmed both of them into an effort to improve the system that had failed them.

Their story remains valuable because it forces an uncomfortable but necessary distinction between honesty and accuracy.

A witness can be completely honest about what he or she remembers. The investigation still has to determine whether the memory is correct.

How Tinsley Investigations Can Help in a Criminal Defense Case

A criminal defense investigation should do more than repeat what is already contained in the discovery. Our investigators can locate and re-interview witnesses, compare current statements with earlier accounts, examine the circumstances surrounding an identification, revisit scenes, develop timelines, and look for independent evidence that supports or contradicts the State’s theory. When a case depends heavily on eyewitness testimony, those details can be especially important because the defense attorney needs to know not only who the witness identified, but how that identification developed and whether the surrounding evidence supports it.

Tinsley Investigations also provides attorney support services for criminal defense teams, including witness interviews, evidence review, investigative research, scene documentation, and other fact-development work that can assist counsel in preparing the case for negotiation or trial.This article is part of Tinsley Investigations’ series examining eyewitness testimony, wrongful convictions, and the importance of independent criminal-defense investigation.

Disclaimer: This article is provided for general educational purposes only and is not intended as legal advice. Every criminal case is different. Anyone facing criminal charges should consult a qualified criminal-defense attorney regarding the facts and circumstances of the case.

Tinsley Investigations

When the stakes are high, every move matters.

Professional investigative services for attorneys, businesses, and private individuals throughout Georgia, including criminal defense investigations, surveillance, child custody investigations, and litigation support.

Email

[email protected]

(912) 599-2119

© 2026. All rights reserved.

Phone

GA License# PDSC001777

Professional Organizations