Earlier this week, a man named Raymond Towler was released from prison after DNA evidence exonerated him. Towler had been convicted in 1981 of raping an 11-year-old girl in a Cleveland park and given a life sentence; he spent nearly 30 years in prison before a group called The Innocent Project took up his case and requested that DNA evidence be re-evaluated.

Towler’s freedom made national news. The judge in the case, Cuyahoga County Common Pleas Court Judge Eileen Gallagher, did not merely free Towler as a formality but left the bench to congratulate Towler at his defense table and gave him an Irish blessing. Towler was also the guest of honor at a Cleveland Cavaliers NBA playoff game.

It’s a tragic story with a bittersweet ending, but there’s an important element that has been largely ignored in the news surrounding Towler’s exoneration: the victim and witnesses identified him as the rapist from a photo. That is, the 11-year-old girl and other eyewitnesses pointed to a photograph of Raymond Towler and told police, “Yes, that is the man who did this.” But Raymond Towler was not the man who did this.

So what happened? Were they lying? Though false rape accusations do occur (as the Tawana Brawley and the Duke University lacrosse team cases show, to name just two high-profile scandals), it seems doubtful that’s what happened in this case. Instead, most likely the eyewitnesses were not lying, nor stupid, nor malicious. They were simply wrong.

Countless cases prove this point. Recently in the District of Columbia, a police sergeant involved in a chase positively identified a 14-year-old boy as the driver of a vehicle involved in a mass shooting. The boy was arrested and charged with four counts of first-degree murder. As the Washington Post noted, “An arrest based on this identification probably struck most people as reasonable, even laudable. After all, the sergeant is a trained, experienced police officer, and he was certain enough of his identification to commit it to a charging document.” Yet despite his positive identification, the trained police sergeant was wrong; other evidence proved that the boy was innocent.

Then there’s the case of the D.C. snipers who killed ten people and badly injured three others in October 2002. Police were baffled by the killings, though an apparent break in the case came when several eyewitnesses described the shooter: A white man driving a late-model white van or box truck. Based on these multiple eyewitness descriptions, police stopped white vans along the Capital Beltway hoping to stop the killer. Yet when the snipers were caught, it was clear that the sincere, believable eyewitnesses with no reason to lie or exaggerate were completely wrong.

Instead of a single white man driving a white box truck, the murders were committed by two Black men driving a dark blue 1990 Chevrolet Caprice. In that case, the eyewitness testimony likely cost human lives: Police had in fact noted the Chevrolet at several of the crime scenes but did not stop or check out the car because the police and public were focused on the non-existent white van reported by eyewitnesses.

Eyewitness misidentification is the leading cause of wrongful conviction in America. Of the more than 200 people exonerated by way of DNA evidence in the US, over 75% were wrongfully convicted because of eyewitness mistakes. Indeed, according to the Innocence Project, “While eyewitness testimony can be persuasive evidence before a judge or jury, 30 years of strong social science research has proven that eyewitness identification is often unreliable. Research shows that the human mind is not like a tape recorder; we neither record events exactly as we see them, nor recall them like a tape that has been rewound. Instead, witness memory is like any other evidence at a crime scene; it must be preserved carefully and retrieved methodically, or it can be contaminated.”

Often in criminal cases there’s a strong and understandable desire to believe the victim. No one wants to question or challenge a person who has obviously undergone a horrible experience—but it must be done. That eyewitness reports are often very unreliable is not news to psychologists or experienced police detectives, but the general public is often unduly impressed with an eyewitness who says, “I know what I saw, and I saw him do it.” Maybe, maybe not.

And it’s not just in crimes: many people who believe in cryptozoological curiosities such as Bigfoot and the Loch Ness monster also put great faith in eyewitness reports—especially sightings by police officers and others in authority. Yet the evidence is clear and uncontested: people are not good eyewitnesses, and often sincerely claim to see things they did not.

Think of how horrible the rape victim who accused Towler must feel, knowing her mistake took 30 years of an innocent man’s life. Those who offer eyewitness evidence against others should heed Oliver Cromwell’s famous plea: “I beseech you, in the bowels of Christ, think it possible you may be mistaken.”