Schauer: Judge and Jury or Lie Detector?
While not accurate enough to use at trial, the results of lie detector tests may still be more reliable than some methods judges and jurors use to evaluate the credibility of a witness, Professor Frederick Schauer said during a lecture Tuesday.
Schauer, the David and Mary Harrison Distinguished Professor of Law, reframed the controversy, arguing that while the existing lie detecting technology is not accurate enough to use as evidence in a trial, it may be more accurate in determining veracity than methods currently used by judges and juries.
The most accurate lie detector test in use today is done by using functional magnetic resonance imaging to scan a person’s brain to discover what parts he or she uses for cognitive tasks. The parts of the brain being used will recruit oxygenated blood cells to help perform the task and will appear to “light up” on the fMRI, Schauer said.
“If — and it is a huge if — different parts of the brain are used for lying than for truth telling, then the possibility exists of using a brain scan to determine whether the person whose brain is being scanned is lying or telling the truth,” he said.
But leading neuroscientists have ruled much of the data in support of this model inconclusive. Schauer said a Stanford law professor and neuroscientist reviewed all such studies and urged a “legal moratorium” on the use of lie detecting technology until its accuracy can be drastically improved.
Even so, Schauer brought up a little-discussed point that he said is still important to the debate: Are the methods judges and juries currently use to detect veracity any more accurate than a lie detector test?
Schauer said jurors tend to supplement what witnesses say with judgments of how they act. For example, whether a witness looks up or down, fidgets, or speaks slowly or quickly will make jurors more likely to think the witness is lying, even if he or she is not, he said. Additionally, Schauer said both judges and jurors tend to believe a person is more likely to lie if he or she has been convicted of a crime.
“We know from serious research that these alleged indicators of veracity are at best highly unreliable and at worst systematically misleading,” Schauer said.
He cited a study that found people untrained in lie detecting are rarely more than 60 percent reliable in identifying lies, only 10 percent higher than relying on pure chance.
Schauer cautioned audience members to keep the spheres of law and science separate when appropriate. Scientists can determine an acceptable error rate for their experiments, but that error rate may not transfer directly into the field of law, he said.
A conservative estimate of the fMRI’s accuracy in detecting a lie is 60 percent, Schauer said. If those results were the sole evidence used against someone, Schauer said he would need a higher accuracy level to prove guilt beyond a reasonable doubt.
If that same fMRI were being used to prove someone’s innocence, however, the same 60 percent accuracy level is enough to cause reasonable doubt of guilt. In science, there is no such gray area, he said.
“The evaluative standard to be used by the law, even when it is science that is being evaluated, must be based on law’s goals, law’s purposes and law’s structure,” Schauer said. “And is so often the case, what is good outside of law may not be good enough inside it. Less obvious, but often more important, is the corollary: that what is not good enough outside of law may be good enough for parts of law.”
• Reported by lindsey wagner