Judge Dismisses Brain Scan Lie-Detection Technique in Court
The case that sparked the decision revolved around defendant Lorne Semrau, a CEO of two nursing homes who was accused of forcing his employees to fraudulently fill out their Medicaid and Medicare forms. Semrau maintained he was simply following governmental instructions, but the feds say his facilities siphoned off an extra $3 million through fraud. Semrau's lawyer wanted to use an fMRI brain scan to prove that the defendant was telling the truth, but, according to Judge Tu Pham, such evidence isn't nearly robust enough to hold up in court.
Pham justified his decision by citing the Daubert standard, which holds that all scientific evidence must have low error rates, and must have been thoroughly tested, peer-reviewed, and generally accepted by the scientific community. As Wired reports, Pham argued that the fMRI technique didn't meet the standard, since it had been tested in laboratory situations where the consequences of lying were non-existent. Pham also pointed to questionable practices on the part of Cephos, the company administering the scan. After Semrau failed one of two tests, Cephos issued a third one, on the grounds that the defendant was fatigued. Furthermore, the test cannot determine whether or not an individual is answering honestly to specific questions (e.g., "Did you lie to the government?"). Rather, it can only give an overall assessment of their truthfulness.
Pham did, however, mention that, even if test results can't currently be applied in the real world, brain scans could be used as evidence in the future, in the event that the approach undergoes further testing, and better refines its methods. Still, Pham's decision may have a ripple effect across other U.S. courts. Vanderbilt law professor Owen Jones remarked, "It has no automatic binding force on any other court, but because it's been so carefully done, it will very likely carry a lot of persuasive value." [From: Wired]