The American Association for the Advancement of Science's Science Insider has this article: "Brain Exam May Have Swayed Jury in Sentencing Convicted Murderer" by Greg Miller.
Earlier this month, a jury in Miami rejected the death penalty and chose life in prison for Grady Nelson, who in 2005 stabbed his wife 61 times, killing her, and stabbed and raped her 11-year-old, mentally handicapped daughter. A report in The Miami Herald last weekend suggests that measurements of Nelson's brain activity may have influenced some members of the jury, who viewed the results as evidence of a brain injury that would partially explain his behavior. But some scientists are critical of the way this technology was used in the case. During the sentencing phase of the trial, the court heard testimony from Robert Thatcher, a neuroscientist and president of by Applied Neuroscience Inc. of St. Petersburg, Florida. Thatcher's company examined Nelson using a method called quantitative electroencephalography (QEEG). As in standard EEG, technicians place electrodes on the skull to record electrical activity in the brain. In QEEG, a computer program analyzes these recordings to locate regions of abnormal activity...
The QEEG data Thatcher presented were riddled with artifacts, and his analysis was undermined by serious statistical flaws, says Charles Epstein, a neurologist at Emory University in Atlanta, who testified for the prosecution. Epstein adds that the sharp waves Thatcher reported looked more like blips caused by the contraction of muscles in the head. "I treat people with head trauma all the time," he says. "I never see this in people with head trauma."