False Confessions and Memory: How Interrogation Shapes Memory and Wrongful Convictions

Short summary

The podcast examines how false confessions arise, distinguishing voluntary confessions from coerced ones and exploring how stress, suggestibility, and interrogation tactics can warp memory. Through historical and contemporary case studies, experts discuss the enduring danger of persuasive confessions and the memory science behind them, along with safeguards such as the PEACE interviewing model and interview recording.

• Types of false confessions: voluntary, coerced-compliant, and coerced-internalized
• Memory distortion driven by suggestibility and retrieval stress
• High profile cases and DNA exonerations highlight systemic risks
• Protective practices: peace model, audio/video recording, and police training shifts

Long form summary

The podcast All in the Mind on ABC Radio National tackles a challenging topic in forensic psychology: why false confessions occur, how memory can be distorted under interrogation, and what safeguards can reduce wrongful convictions. The conversation centers on the historical development of research into confession evidence, beginning with Saul Kasson, a psychology graduate student who explored how police interrogation tactics influence confession rates. Kasson drew on a leading interrogation manual and the broader literature on social influence and obedience, laying the groundwork for what would become a new field of investigating confession evidence. The discussion underscores that confessions can be extraordinarily persuasive to jurors, even when the confession is false, because a signed admission of guilt collapses into the narrative of the case and can eclipse other lines of evidence. The host places this in context by noting the significance of false confessions in both the United States and Australia, with real-world cases that reveal the system’s vulnerabilities.

Two main types of false confessions The episode clearly distinguishes voluntary false confessions, where individuals admit to crimes they did not commit, often seeking attention or protecting someone else, from coerced false confessions, which arise under legitimate pressure or deception during police interrogation. Kasson emphasizes that voluntary confessions are less common and are not trusted by police unless the confessor can provide details that could not have been learned from newspapers. In contrast, coerced confessions are ethically and legally troubling, especially when investigators use threats, carbons of evidence, or prolonged interrogations to induce a confession. Coercion is further categorized into two forms: compliant confessions, in which the suspect signs a confession under duress and may accept that signing is in their best interest, and internalized confessions, where innocent individuals come to believe that they committed the crime, sometimes even inventing memories of the event. The internalization phenomenon proves that innocence itself does not immunize a person from confessing. Kasson shares a powerful anecdote about a case in which a suspect believed the police had blood evidence and, believing he would be exonerated later, signed a confession that he later realized would guarantee a conviction even if the actual investigation pointed elsewhere. The counterintuitive insight is that innocence can sometimes facilitate false confessions, because innocent suspects expect the truth to emerge through investigation and, when pressed, feel less inclined to insist on their innocence, wrongly assuming that the authorities will uncover the truth in time.

Memory, suggestibility, and the science behind false confessions The podcast surveys foundational memory research that demonstrates how easily memories can be shaped by external influences. A central example is the Loftus car crash studies, which showed participants watched the same videos but were asked different questions about the speed of the vehicles based on the verb used ( crashed, collided, bumped). Those differences led participants to report varying speeds and even to recall non-existent broken glass. The Loftus studies illustrate how language and framing can generate false memories, a phenomenon that has direct implications for police interviews. The discussion also introduces Gudjonsson’s suggestibility scale, a tool used to measure how susceptible an individual is to being influenced during interrogations. It is noted that suggestibility tends to decline with age, but remains a stable trait with marked individual differences. The scientists explain that even people with lower general suggestibility can be swayed under stress and with specific questioning strategies. Retrieval stress—how difficult it is to recall information under conditions similar to an interrogation—also reduces accuracy and increases reliance on memory gaps filled with interviewer-provided details.

Stress, vulnerability, and the mechanics of memory distortion The experts outline that stress, isolation, and fear create cognitive load that impairs memory encoding and retrieval. When suspects are confronted with allegations and evidence, their confidence in their memory can be manipulated, leading to vaguer or hypothetical responses that interrogators may misinterpret as admissions. For instance, a suspect may resort to hypothetical or third-person statements during a pressure-filled interview, which authorities might mistakenly treat as the precursor to a confession. The discussion highlights that the vulnerability to false confessions is not limited to a single demographic; younger suspects and those with psychiatric histories or mental health vulnerabilities are often more susceptible to suggestive questioning and coercive strategies. A vivid example is given of a Nebraska case in which six people confessed to murder; years later, exonerations and compensations were granted, underscoring how lasting the effects of coercive interviewing can be on innocent individuals.

The Innocence Project, DNA evidence, and the real-world impact The podcast cites data from the Innocence Project in the United States, which analyzed 375 prisoners whose convictions were overturned on DNA grounds between 1989 and 2020. The analysis revealed that almost one third of those exonerations involved false confessions, with a notable proportion of those misled confessions involving individuals under 18. The Australian data is less robust, but experts suggest a similar pattern of wrongful convictions involving false confessions. The Innocence Project’s work has served as a catalyst for changes in policing practices and public policy around interrogation. Kasson points to a generational shift in police culture, arguing that newer cohorts trained under Innocence Project-inspired frameworks are more likely to use non-coercive techniques and to record interrogations, reducing the likelihood of false confessions.

Safeguards and policy changes In response to the risk of false confessions, policing practices have shifted toward the PEACE model of interviewing, developed in the United Kingdom to provide a non-confrontational, science-based approach. The PEACE model emphasizes preparation and planning, engagement and explanation, account or challenge, and closure and evaluation. The podcast notes that many Australian police forces now use the PEACE model, which focuses on rapport-building and structured, non-coercive interviewing. An additional safeguard is the routine recording of audio and video during formal interviews, a practice adopted by many jurisdictions since the late 1990s. The experts discuss evidence suggesting that such safeguards have preserved investigative effectiveness while reducing the incidence of false confessions. The UK experience is highlighted as an example where reforms coincided with no evident drop in case closures, suggesting that the reforms did not compromise investigative efficacy. The discussion ends on a cautiously optimistic note, arguing that the combination of improved interviewing techniques, better training, and generational shifts in policing has the potential to reduce false confessions. Australia’s cases, including Andrew Mallard’s wrongful conviction, demonstrate the ongoing relevance of these reforms, and the podcast closes by emphasizing that the field continues to evolve as researchers and practitioners collaborate to safeguard memory and memory-based evidence in the justice system.

Conclusion The podcast makes a strong case that false confessions, rooted in memory distortions and coercive interviewing practices, have led to wrongful convictions across countries. By combining robust memory science, standardized interviewing models such as PEACE, and explicit recording practices, the field aims to safeguard both memory and memory-based evidence in criminal investigations. The conversation also underlines the importance of public awareness and continued reform to ensure that confessions are truly reliable indicators of guilt rather than reflections of interviewing conditions or memory distortions. The overarching message is clear: understanding memory and confession dynamics is essential for fair trials, and ongoing reforms hold promise for a justice system that better distinguishes memory from memory manipulation.