In a furious effort to save their client from execution, lawyers for convicted cop killer Troy Anthony Davis are hinging much of their case on the fallibility of eyewitness testimony.

Racial bias, poor lighting, stress, alcohol, the passage of time, poorly conducted police lineups and other factors often play a role in misidentifications of criminal suspects.

More than 75 percent of the 205 people exonerated by post-conviction DNA evidence in the United States — including all six in Georgia — were imprisoned because of mistaken eyewitness identification.

Davis' legal team has hired an eyewitness expert at the University of Alabama in Huntsville to examine testimony about the events that led to Davis' conviction.

In his report, associate professor Jeffrey S. Neuschatz found numerous concerns with the identification of Davis as the man who fatally shot Officer Mark Allen MacPhail in a Burger King parking lot on a summer night in Savannah.

Seven of nine witnesses who helped implicate Davis for the murder have since recanted their testimony, so such analysis might prove valuable. Davis' lawyers seek to raise questions of how authorities may have steered some to point the finger at their client.

"The eyewitness testimony formed the central theme of the government's entire case," Danielle Garten, one of Davis' lawyers, said last week. "There was no weapon found, there was no physical evidence. The case relied on eyewitness testimony."

Eyewitness IDs analyzed

Whether concerns about that testimony will be enough to cast doubt on Davis' guilt remains to be seen. The parole board issued a 90-day stay of Davis' execution, which was scheduled for Tuesday. By Oct. 14, it must decide whether to commute his death sentence to life in prison (with or without parole) or allow the execution. The parole board has not made any public statements about why they halted the execution.

Neuschatz's report has been filed with both the parole board and in Davis' court appeals for a new trial. Neuschatz analyzed the eyewitness identifications using contemporary standards to determine if there were flaws in the procedures used to implicate Davis.

Neuschatz concluded that one witness, Dorothy Lee Ferrell, told police she had seen Davis' picture on the news as a suspect in MacPhail's slaying. "Prior exposure to the suspect's picture increases the likelihood that the suspect will be picked out of the lineup," Neuschatz wrote.

Neuschatz also made other observations, including: When a weapon is involved in a crime, witnesses tend to focus on it, rather than the suspect; the passage of time, in many cases 10 days, between the crime and the identification of Davis. Another witness testified that he had been drinking on the night of the shooting.

Neuschatz, in an interview, declined to comment on the specifics of the Davis case. But he emphasized that his work is to analyze data, not come to legal conclusions.

"I don't feel it's my role to make a decision whether this person is accurate," Neuschatz said. "I just argue what the state of the science is."

Attempts to reach Chatham County District Attorney Spencer Lawton by telephone and e-mail have been unsuccessful. Lawton and top Savannah police officials declined to talk with reporters when they came to Atlanta last week for Davis' clemency hearing.

Concerns over faulty eyewitness identification have reached the state Capitol. Earlier this year, state Rep. Stephanie Stuckey Benfield (D-Atlanta) — on the heels of the exoneration of a man imprisoned 21 years for a wrongful rape conviction — tried to pass legislation that would bring uniformity and higher standards to the way law enforcement officials conduct live and photo lineups.

House Speaker Glenn Richardson said he liked the idea, but it was opposed by prosecutors and did not pass.

But Benfield managed to push for a House study committee on eyewitness identification procedures, and Richardson recently appointed her to chair it.

Fall hearings planned

This fall, the committee is scheduled to hold a series of hearings on eyewitness identification. Barry Scheck, a DNA expert who gained fame as part of O.J. Simpson's legal defense team, has been invited to testify.

Prosecutors, defense lawyers, law enforcement, eyewitness ID experts and exonerees are also expected to participate.

The Georgia Innocence Project, which has played a role in three of the state's exonerations, is promoting lineup standards.

"In all six of those [Georgia] cases, the victims, and sometimes witnesses as well, incorrectly identified the attackers," said Lisa George, spokeswoman for the project. "It's not that these victims or witnesses were lying; it's just that they got it wrong. Human memory is extremely fallible."

Rick Malone, executive director of the Prosecuting Attorneys' Council of Georgia, said prosecutors don't object to better standards for lineups, but they don't want them codified into state law. Among his objections, Malone said such standards are promoted by defense lawyers who want "to put an additional set of rules out there between witnesses saying what they saw and their clients."

Malone, a former prosecutor, said he believes lawyers opposing Davis' clemency will have an easy time dismantling Neuschatz's report in a cross-examination.

"Here's a fella who, 18 years after the fact, thinks he can tell us what the law ought to be. Well, I have a lot of questions for people like that. What's he relying on? Did he do this research? Did he read the research that someone else did? What were the qualifications of the person who did the research? . . . You could go on forever. They're out there just simply speculating."