Case File 279 · Investigation Psychology

Why Are Confessions So Persuasive to Juries?

By Murder Book Editorial TeamPublished by Gungrounds, the studio making Murder BookPublished 3 October 2026Updated 3 October 2026

Confessions are among the most powerful forms of evidence in a criminal trial, and research consistently shows they have a stronger effect on jury verdicts than eyewitness identification, character testimony or other common evidence types. In a series of mock jury studies, Kassin and Neumann (1997) found that confession evidence produced the highest conviction rates across multiple crime scenarios, even when jurors were told the confession had been coerced. The core problem is not that jurors are careless but that confessions trigger a deeply held assumption: people do not confess to crimes they did not commit. Research shows this assumption is wrong more often than most people expect.

Key takeaways

Mock jury studies show confession evidence produces higher conviction rates than eyewitness identification or character testimony.
Even when jurors learn a confession was coerced or retracted, they often fail to fully discount it from their deliberations.
People exhibit a "truth bias" when evaluating confessions, tending to assume the confessor is telling the truth.
In DNA exoneration cases, false confessions contributed to roughly 25-29% of wrongful convictions documented by the Innocence Project.
Research suggests that a confession can contaminate other evidence in a case, influencing how forensic analysts, eyewitnesses and informants interpret their own information.

Why do confessions carry so much weight?

The persuasive power of confessions rests on a common-sense belief that is difficult to shake: the assumption that no rational person would confess to a crime they did not commit. This belief is so strong that even when people are educated about the existence of false confessions, they still tend to give confession evidence substantial weight.

Kassin and Neumann tested this directly in 1997. In their first experiment, participants read summaries of criminal trials involving murder, rape, assault and theft. Each summary contained one of four evidence types: a confession, an eyewitness identification, character testimony, or no additional evidence. The confession condition produced the highest conviction rates across all crime types. In follow-up experiments, even a confession described as coerced led to higher conviction rates than other evidence forms.

The researchers described this as evidence that confessions are "fundamentally different" from other types of evidence in their psychological impact, contradicting a 1991 US Supreme Court ruling in Arizona v. Fulminante that treated confessions as similar to other evidence for the purposes of harmless error analysis.

What happens when a confession is retracted?

In many false confession cases, the suspect recants before trial, claiming the confession was coerced or false. Research suggests juries struggle to set aside a retracted confession.

A study by Kassin and Sukel (1997) found that mock jurors who were presented with a confession obtained under high pressure judged it to be coerced, yet still voted to convict at higher rates than those who received no confession. The jurors could identify the coercion but could not prevent the confession from influencing their verdict. This pattern, sometimes called the "I know it but I can't help it" effect, has been replicated across multiple studies.

The difficulty of disregarding confession evidence is compounded by the way trials unfold. By the time a jury learns that a confession may have been coerced, they have already heard it. Judicial instructions to disregard evidence are notoriously limited in their effectiveness, and confession evidence appears to be especially resistant to such instructions.

Can a false confession contaminate other evidence?

Yes, and this may be one of the most concerning findings in the research. Kassin, Bogart and Kerner published a 2012 study titled "Confessions that corrupt" which examined how knowledge of a confession affects other participants in the justice system.

Their research found that when forensic examiners, eyewitnesses or alibi witnesses learned that a suspect had confessed, their own judgements shifted. Eyewitnesses became more likely to identify the suspect. Forensic analysts interpreted ambiguous evidence as more incriminating. In DNA exoneration cases studied by the Innocence Project, false confessions were followed by additional errors, such as invalid forensic science and unreliable informant testimony, roughly two-thirds of the time, and the confession typically came first in the sequence.

This contamination effect means that a false confession does not remain an isolated error. It can cascade through the investigation and trial, producing what appears to be a strong, multi-evidence case against an innocent person. For a closer look at how crime-scene details can enter a confession, see the article on confession contamination.

How common are false confessions in wrongful convictions?

The Innocence Project has documented that false confessions contributed to approximately 25-29% of the DNA exonerations in the United States through 2020. This makes false confessions one of the leading factors in proven wrongful convictions, alongside eyewitness misidentification.

Importantly, these figures represent only cases where DNA evidence was available and could be tested. The true prevalence of false confessions across the broader justice system is unknown, because most cases do not involve biological evidence that can be independently tested. Researchers such as Kassin have argued that the documented cases likely represent the visible portion of a much larger problem.

Why do people find it hard to believe in false confessions?

The idea that an innocent person would confess contradicts a powerful psychological intuition. People tend to believe that they themselves would never confess to something they did not do, and they extend that assumption to others. This creates a gap between what research has demonstrated and what jurors, judges and investigators feel is plausible.

Research on the fundamental attribution error suggests that people tend to attribute behaviour to character rather than to circumstances. When they see a confession, they assume it reflects the truth about the confessor rather than the truth about the circumstances, such as lengthy questioning, coercive tactics or the vulnerability of the suspect.

This bias is resistant to education. Even people who have been told about false confession research tend to believe that the specific confession in front of them is genuine. The pattern holds for both lay jurors and legal professionals.

What safeguards have been proposed?

Researchers have recommended several measures to help juries evaluate confession evidence more accurately. Expert testimony on false confessions can help jurors understand the circumstances under which innocent people confess, though its effectiveness varies. Mandatory recording of interrogations allows jurors to see the full context rather than only the final statement.

Some researchers have also recommended that courts require corroboration before admitting confession evidence, particularly in cases involving vulnerable suspects. A confession that contains details the suspect could only have known through involvement in the crime is more probative than one whose details could have been fed by investigators.

How does Murder Book explore the weight of a confession?

Murder Book: The Art of Interrogation places players in the position of evaluating what suspects say. Development screenshots show an evidence board linking statements such as "I NEVER met her!" and "I had no idea" to physical evidence, including a fabricated description of a gun. The game asks players to weigh whether these statements are consistent with the evidence or whether something does not add up.

The research above is directly relevant. A confession or admission in Murder Book is not the end of the case; it is something to be tested against other evidence. The game includes a branching case map with multiple endings, including a "not guilty" result, which means reaching the wrong conclusion is possible. This design reflects the research finding that confessions are persuasive but not infallible. Murder Book is in development for PC (Steam) and mobile, with release currently planned for the end of Q2 2027.