We've all seen that electrifying moment on TV or at the movies. In a packed, expectant courtroom, a witness with a quavering voice points a finger at the defendant and declares: "That's him there - that's who I saw."

Eyewitness identification makes for powerful drama. Unfortunately, in the real world of the criminal justice system, erroneous eyewitness identification is the single greatest cause of wrongful convictions.

Recently, the New Jersey Supreme Court's decision in State v. Henderson took the most determined steps in U.S. history to protect against wrongful convictions secured through eyewitness misidentification.

It is well recognized that eyewitness identification is subject to error, but since the US Supreme Court first set standards for eyewitness identification procedures in 1977, a vast body of research has consistently demonstrated just how fallible eyewitnesses can be, although they're often sincere, well-meaning and confident in their identification.

Although the certain frequency of witness misidentification is unknown, the research suggests 33 percent of witness identifications are mistaken. The Innocence Project determined that three-quarters of all convictions overturned by DNA proof were based on misidentification.

New Jersey's decision in Henderson draws from a comprehensive review of the scientific knowledge, including more than 200 studies and testimony from seven experts. The unanimous decision is a powerful statement that N.J. courts must incorporate the empirical data and scientific knowledge in their decision-making analyses regarding the reliability of the identification.

Previously, eyewitness identification testimony could be suppressed only in cases of the police being "overly suggestive" in the handling of a witness, and even then, only if the trial judge found the improper procedure "substantially likely" to lead to error.

In practice, as Chief Justice Stuart Rabner observed in his opinion, judges are loath to adopt this extreme remedy, leaving the defense no option for countering eyewitness identification other than cross-examination of the witness - oftentimes the victim - to demonstrate the unreliable nature of the identification. Of course, attacking a victim who proclaims 100 percent confidence in his identification is hardly a productive way to persuade a jury.

Jurors believe eyewitness testimony because witnesses usually believe it themselves, often with unshakable certainty. Rarely does a witness knowingly make a false identification. Rather, as the science demonstrates consistently, and the Supreme Court now acknowledges, perception is subjective, and memory is malleable and fragile.

"Reality," John Lennon wrote, "leaves a lot to the imagination." People view reality through the lens of their beliefs, attitudes and values. So many variables might impair the accuracy of a witness's observation: the stress of the situation; the duration, complexity and confusion of the unfolding events; the presence of a weapon; the lighting and the observer's distance from the action; the participants' racial makeup; the inexorable decay of all recollections; and the witness's age and acuity.

And then there are the many ways, both subtle and unsubtle, that police can suggestively influence whom a witness picks when viewing an array of photos. The New Jersey Supreme Court rejects the "bright line" called for by some advocates of suppressing eyewitness evidence in any instance of suggestive conduct by the police. Rather, trial courts will hold pre-trial hearings to examine both the investigative procedures and the variables affecting the reliability of the witness's observation.

Second, judges will now be required during their jury charge to educate juries regarding the reliability of eyewitness evidence and the various factors that affect it, and expert testimony may be a necessary part of the defense case to further educate the jurors.

The court said it expects identification testimony will still be presented in most cases. The threshold for suppression remains high. But jurors will have a much deeper understanding of the testimony they hear in order to weigh the reliability of that evidence in light of all the factors that can undermine it.

With the U.S. Supreme Court scheduled to take up these very issues in a New Hampshire case this fall, New Jersey sets a new standard for the treatment of eyewitness identification that will be difficult to ignore. Henceforth, one of the worst nightmare scenarios of criminal justice - the wrongful conviction of the innocent - should become less likely and less frequent.

Remi Spencer is a criminal defense lawyer and founding partner of Spencer & Associates in Newark. Roger Cohen is a public affairs consultant based in Bloomfield.