On October 30, 2012, in Naperville, Illinois, 40-year-old Elzbieta Plackowska stabbed her 7-year-old son and a 5-year-old girl she was babysitting. She told evaluators she had been hearing voices. The psychiatrist retained by her defense concluded she was in the grip of a manic episode and could not appreciate that what she had done was wrong.
The psychiatrist called by the prosecution wasn’t buying it. Her story kept shifting. She said the children had seen the same visions she did, but real hallucinations don’t work that way. And the voices, she said, had spoken to her in English, though her first language was Polish. After a bench trial in 2017, the judge sided with the prosecution and convicted her.
Two forensic psychiatrists examined the same woman and reached opposite conclusions about whether she was lying. Cases like this turn on one of the hardest judgments in forensic work: whether a defendant who claims to be mentally ill is telling the truth. The fear of being fooled by a faker is real. The opposite mistake is just as costly: branding a sick person a liar. The error cuts both ways.
It Looks Like an Easy Out. It Rarely Is.
Ask the average juror, and you will hear that the insanity defense is a common loophole, easy to fake, and used all the time to dodge punishment. Decades of research say otherwise: Insanity pleas are rare, seldom succeed, and usually end in long psychiatric confinement, not release.
But the fear that a dangerous person will be set free to run amok in the community isn’t unfounded. Faking insanity happens. A 2025 analysis of roughly 1,300 people in a forensic psychiatry practice found that about a quarter failed at least three tests designed to catch invented or exaggerated symptoms.
Psychosis Shows Up at 3 a.m.
A validity measure tests whether someone is reporting honestly and putting in real effort. Some flag the person who reports too many dramatic, rare, or impossible mental health symptoms. Others track effort on tasks that look hard but are easy enough that even brain-injured patients pass. A deliberately poor score stands out.
The tests catch a lot. So does the conversation. With someone who is genuinely sick, you don’t just hear about the symptoms; you see them. They stop mid-sentence to listen to a voice no one else can hear. Their fear matches the paranoid delusion. A faker can describe all of that, but can’t produce it.
In nearly 30 years of competency and criminal responsibility evaluations, I’ve found that the genuinely ill rarely perform their illness. They’re too busy living inside it. It’s the performers who hand you the symptoms in the order a movie taught them.
Some defendants will agree to almost any symptom you ask them about. Some don’t even know what a hallucination is, mistaking something ordinary, like talking to themselves, for one. Others overdo it, piling on incompatible symptoms that no real illness produces: the wired, sleepless high of mania running full blast next to the dead weight of deep depression, hour after hour. That isn’t how illness works.
And the interview is only part of it. Listening to jail call recordings can be enlightening; despite clear indications that these calls are recorded, you’d be amazed at what some defendants reveal. Some of the most valuable sources are the officers and nurses who staff the jail, who are around the defendant for hours with no skin in the game. They see the defendant at three in the morning and in the lunch line. A performance is exhausting to sustain around the clock, and it often slips when the audience no longer matters.
When a Hunch Becomes Expert Testimony
The defense has its own legitimate worry. Law professor Deborah Denno recently tracked malingering claims across every American insanity case involving neuroscience evidence over the past 120 years. The accusation grew from a rarity in the early 1900s to more than a fifth of insanity cases by the 2000s. When it was raised, it shaped the outcome in roughly 70 percent of cases.
Yet much of the time it rested on thin support. Often, the finding came down to an examiner’s impression that the defendant seemed manipulative. Not a failed test, not a validity measure, but a feeling about the person across the table, written up in clinical language and carried into a courtroom as expertise.
None of this means faking isn’t real. It is, and modern tests can catch it. The trouble is that the courtroom malingering accusation, all too often, doesn’t include them. It rests on a clinician’s gut read, the same blunt hunch the field ran on a century ago, before those tests existed. The science got sharp. The accusation often didn’t. And it can carry a case without anyone running the assessment that might have settled it.
Spotting a Liar Is a Coin Toss
It’s tempting to think you’d know a faker if you saw one. The research says otherwise. Across hundreds of studies and more than 20,000 observers, the average person spots a lie about 54 percent of the time, barely better than a coin toss. We psychologists only do a smidgen better. And here’s an interesting fact: How confident you feel about your judgment has almost nothing to do with whether you’re right.
It’s why I don’t lean on my own impression of a defendant, and why I’m wary of any examiner who does. The face across the table is the least reliable instrument in the room.
When that certainty hardens into a courtroom opinion, someone pays. In a New York case, a young woman named Keila Pulinario shot a man she said had raped her and intended to do it again. The state’s psychiatrist called her a faker. At trial, she admitted she had lied to the evaluators about her sexual history, and the judge used those lies to bar her own expert from explaining her trauma to the jury. A federal court reversed, holding that her untruthfulness was a symptom of what she had survived, not a reason to silence her defense. Withholding and shame are not holes in a rape victim’s account. They are part of it.
Two Ways to Be Fooled
Mistaking a sick person for a faker denies them treatment and adds years to a sentence. Mistaking a faker for a sick person lets a fabrication corrupt the verdict. Both are real dangers, and both come from the same shortcut: trusting a feeling instead of doing the work, watching how the symptoms show up, running the tests built to catch a fake, and checking the story against the people who see the defendant when no one is performing for a judge. The difference between a genuine breakdown and a good act is there to be found. What hides it is an examiner who has already made up their mind.