Mayfield is one of a small but growing number of Americans who've been jailed on the basis a fingerprint match that turned out to be false. And as Correspondent Lesley Stahl first reported last year, it also happened to Rick Jackson.
When the police in Upper Darby, Pa., arrested Rick Jackson, they told him they had solid evidence that he had committed a gruesome crime - his friend's murder.
They said they had his bloody fingerprints, and showed Jackson Polaroids of what they said were his fingerprints.
"I actually felt relieved because I knew there was no way they could be, and that it was just going to be a matter of like, 'Ah, somebody made a mistake,'" says Jackson.
His father, Richard Jackson Sr., was with him when he was taken into custody. He says the local police told him their own fingerprint experts had made the match: "Several of them, with their professional experience, had read these prints, and they, without a doubt, knew they were my son's."
He hired a local attorney, Mike Malloy, to defend his son.
Malloy decided to call in his own expert, retired FBI examiner George Wynn, who drove up from his home in Virginia to see the evidence firsthand.
"When we leave the police department that day, he turns to me, in his southern drawl, and says to me, 'Michael, those prints don't match,'" recalls Malloy.
For a second opinion, Wynn sent the evidence to another former FBI examiner, Vernon McCloud. Between them, the two experts had 75 years of fingerprint experience.
"McCloud did to Wynn what Wynn had done to me and called up and said, 'Are you kidding me?'" says Malloy. "The expression that they would continually say to me was, 'It's not even a close call.'"
On TV, fingerprints are matched automatically, by computer. But that never happens in real-life forensic work.
Steve Meagher heads up the crimes scene fingerprint unit at the FBI. He says a computer, like the one at the Bureau, sorts through a database of 44 million sets of prints, but that merely narrows the search.
And then the humans, the fingerprint examiners, make the actual match the way they always have - by eye. In fact, it's arduous work involving human judgment. An examiner can spend hours, even days, analyzing a fragment of a fingerprint.
Usually, police lift only fragments at a crime scene, and they can be contaminated or distorted. The first thing an examiner compares is what's called ridge flow.
An examiner studies the width of the ridges and the direction in which they flow. "We could select any ridge that we wanted and follow the path of the ridge. At this point, you'd see a bifurcation [point of similarity]," says Meagher, who looked at several of those points and double-checked to make sure there wasn't the slightest discrepancy between the two.
Robert Epstein, a public defender in Philadelphia, is challenging the very notion that fingerprint identification is absolute, something he began to question when he was assigned to an armed robbery case, where the critical evidence was two partial fingerprints.
The uncertainty in his own mind led him to do some research, and he says he discovered that not only has there never been a study of the reliability of crime-scene fingerprint matching, there are no agreed-upon standards for what constitutes a match.
"There's complete disagreement amongst fingerprint examiners themselves as to what they need to see in order to declare a match," says Epstein.
In Italy, for example, examiners say they have to see 16 or 17 points of similarity. In Brazil, it's 30; in Sweden, it's seven points; and in Australia, it's 12. And most examiners in the United States, including those at the FBI, don't even use a point system.
Robert Epstein did something audacious. In a pretrial hearing, he argued that since the reliability of fingerprint matching had never been tested or proven, it should be barred as evidence from the courtroom.
"There isn't a single experiment that's ever been done, literally," says Dr. Ralph Haber, a forensic scientist.
Haber thinks Epstein is on to something, and he says he's ready to do the research to test reliability himself, if law enforcement will cooperate.
Over the last 100 years, there have been only a handful of cases where convictions have been reversed because of faulty fingerprint identification, but that doesn't mean examiners don't make mistakes. In fact, there's plenty of evidence they make lots of mistakes. Consider the certification tests given by the International Association of Identification, or IAI, which only professionals who are already working in the field can take.
"Half of the examiners who take the test fail it. There are 15 items. They can make no more than three mistakes, and they're out," says Haber.
But that doesn't mean they can't do analysis anymore. "Virtually no crime laboratory requires that you be certified to work in the lab, even in the FBI lab," says Haber.
Malloy, Jackson's defense lawyer, says he runs up against the same problem in court. Judges don't require that fingerprint experts who testify be certified.
"The underlying problem is not the evidence itself, but is who's allowed to be qualified as an expert," said Malloy. "The police experts were really just your local police officer, I mean, who, on a given day, might do anything from getting the cat out of the tree to examining the fingerprints."
The two local officers, along with a third certified examiner from out of state, testified that the bloody prints were - with 100 percent certainty - Rick Jackson's. But the defense experts, both IAI certified, swore that the prints could not be Jackson's.
If fingerprint identification is an exact science, as the FBI claims, a situation like this should never have occurred. But the judge allowed it, and after a two-week trial, the jury returned a verdict in a matter of hours. Rick Jackson was found guilty of first-degree murder.
Jackson began serving a life sentence, with no chance of parole. In the meantime, Epstein lost his bid to have fingerprint evidence barred from his trial, and his client was also convicted. But his effort has spawned a wave of similar challenges.
"It's hardly surprising that trial judges around the country didn't, all of a sudden, start excluding fingerprint evidence from the courtroom. This is the kind of issue that has to be raised and raised again, and people have to hear about it, and it has to sink in," says Epstein.
It's not the judge's fear of appeals, says Meagher, the FBI's fingerprint expert. He says they're rejecting the challenges because they accept this point: that fingerprint analysis is a reliable science, and it shouldn't be confused with the question of human fallibility.
"We're winning 41 times out of 41 challenges. I think that says something. We have 100 years of experience. Lets make sure that that's clearly out there. And if it wasn't reliable, this certainly would have been discovered many, many years ago," says Meagher.
It may be reliable in the abstract, but fingerprints come down to human judgment, subject to human error, as the case of Rick Jackson demonstrates.
His experts, outraged by the verdict, appealed to the IAI for a review of the prints and got a conclusion that they were not Rick Jackson's.
That eventually led the district attorney to ask the FBI for an opinion, something Jackson's attorney had asked for during the trial. When the FBI analyzed the prints, it determined the prosecution experts had been wrong.
With the prosecution experts discredited, the district attorney finally released Rick Jackson after he had spent more than two years behind bars.
As for the prosecution experts, the out-of-state examiner was decertified and lost his job - but not the two local officers.
"The men who put my son away for over two years are still allowed, and have never been removed from, the ability to read prints," says Jackson Sr.