The Investigative Failures Behind False Confessions

A false confession is the last domino to fall. Trace almost any case here backward and you'll find one or more of the following breakdowns setting it up.

​

Confirmation Bias and Tunnel Vision

Once an investigator decides who did it, the investigation quietly changes jobs — from finding out what happened to proving what they already believe. Evidence that fits the theory is embraced; evidence that doesn't is explained away or never pursued. Alternative suspects go unexamined. In the interview room this bias becomes self‑fulfilling: questions are aimed at confirmation, denials are read as resistance, and the case builds itself around a conclusion instead of the facts. Tunnel vision doesn't feel like bias from the inside. It feels like certainty.

​

Accusatory Interviewing

Legacy guilt‑presumptive, accusation‑driven interrogation is built to break down a suspect who is assumed to be lying — through persuasion, confrontation, minimization, implied promises, alternative questions, and sometimes hours of pressure. On a guilty suspect it may produce a true admission. On an innocent one — especially a juvenile, a person with an intellectual disability, or anyone frightened and exhausted — it can manufacture a confession that never should have existed. Decades of research point the other way: rapport‑based, information‑gathering approaches (such as science‑based interviewing) produce more information yield, more cooperation, and even more confessions with far fewer false confessions. The goal of an interview is to gather case-relevant information, not to win.

​

"Lie Detection" Junk Science

Much of the certainty and overconfidence that fuels a wrongful interrogation rests on methods that don't hold up. The behavior‑based "lie detection" — reading nervousness, eye contact, or body language — performs at barely better than a coin flip in study after study. When investigators believe they can see a lie, they stop testing the evidence and start pressing the person. Reliable deception research points to strategic, evidence‑based questioning, like the Strategic Use of Evidence — not demeanor cues — as the path to the truth. Junk science doesn't just fail to catch liars; it convinces good investigators that an innocent person is guilty.

​

These aren't rare instances. These are poor communication techniques littered in interview and interrogation curricula accross the United States. They are ordinary, human, and trainable — which is exactly why they can be prevented.