False Confessions: Causes, Consequences, and Implications for Reform

Despite the commonsense belief that people do not confess to crimes they did not commit, 20 to 25% of all DNA exonerations involve innocent prisoners who confessed. After distinguishing between voluntary, compliant, and internalized false confessions, this article suggests that a sequence of three processes is responsible for false confessions and their adverse consequences. First, police sometimes target innocent people for interrogation because of erroneous judgments of truth and deception. Second, innocent people sometimes confess as a function of certain interrogation tactics, dispositional suspect vulnerabilities, and the phenomenology of innocence. Third, jurors fail to discount even those confessions they see as coerced. At present, researchers are seeking ways to improve the accuracy of confession evidence and its evaluation in the courtroom. KEYWORDS—police interrogation; confessions; evidence In criminal law, confession evidence is highly persuasive—yet fallible. Despite the pervasive myth that people do not confess to crimes they did not commit, the pages of American history, beginning with the Salem witch trials of 1692, bear witness to all the men and women who were wrongfully convicted and imprisoned, often because of false confessions. Although the prevalence rate is unknown, recent analyses reveal that 20 to 25% of prisoners exonerated by DNA had confessed to police, that the percentage is far higher in capital murder cases (White, 2003), and that these discovered instances represent the tip of an iceberg (Drizin & Leo, 2004). After reviewing a number of cases throughout history, and drawing on theories of social influence, Wrightsman and I proposed a taxonomy that distinguished three types of false confessions: voluntary, compliant, and internalized. Still used today, this classification scheme has provided an important framework and has since been used, critiqued, extended, and refined in subtle ways (Kassin & Wrightsman, 1985). Voluntary false confessions are those in which people claim responsibility for crimes they did not commit without prompting from police. Often this occurs in high-profile cases. When Black Dahlia actress Elizabeth Short was murdered in 1947, more than 50 people confessed. In 2006, John Mark Karr confessed to the unsolved murder of young JonBenet Ramsey. There are several reasons why innocent people volunteer confessions, such as a pathological need for attention or self-punishment, feelings of guilt or delusions, the perception of tangible gain, or the desire to protect someone else. In contrast, people are sometimes induced to confess through the processes of police interrogation. In compliant false confessions, the suspect acquiesces in order to escape from a stressful situation, avoid punishment, or gain a promised or implied reward. Like the social influence observed in Milgram’s classic obedience studies, this confession is an act of public compliance by a suspect who perceives that the short-term benefits of confession outweigh the long-term costs. This phenomenon was dramatically illustrated in the 1989 Central Park jogger case, in which five New York City teenagers confessed after lengthy interrogations, each claiming he expected to go home afterward. All the boys were convicted and sent to prison, only to be exonerated in 2002 when the real rapist gave a confession that was confirmed by DNA evidence. Lastly, internalized false confessions are those in which innocent but vulnerable suspects, exposed to highly suggestive interrogation tactics, not only confess but come to believe they committed the crime in question. The case of 14-year-old Michael Crowe, whose sister was stabbed to death, illustrates this phenomenon. After lengthy interrogations, during which Crowe was misled into thinking there was substantial physical evidence of his guilt, he concluded that he was a killer: ‘‘I’m not sure how I did it. All I know is I did it’’ (Drizin & Colgan, 2004, p. 141). Eventually, he was convinced that he had a split personality—that ‘‘bad Michael’’ acted out of jealous rage while ‘‘good Michael’’ blocked the incident from consciousness. The charges against Crowe were later dropped when a drifter from the neighborhood was found with Crowe’s sister’s blood on his clothing. Address correspondence to Saul Kassin, Department of Psychology, the John Jay College of Criminal Justice, 445 West 59th Street, New York, NY 10019, USA; e-mail: [email protected]. CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE Volume 17—Number 4 249 Copyright r 2008 Association for Psychological Science Inspired by tales of innocents wrongfully convicted, recent research has focused on three sets of questions: (a) Why are innocent people often misidentified for interrogation, (b) what factors put innocent suspects at risk to confess, and (c) how accurate are police, juries, and others at judging confession evidence? WHY INNOCENT PEOPLE ARE INTERROGATED Typically, the confrontational process of interrogation is preceded by an information-gathering interview conducted by police to determine if a suspect is guilty or innocent. In Criminal Interrogations and Confessions (Inbau, Reid, Buckley, & Jayne, 2001), the most influential manual on interrogation, police are thus advised on the use of verbal cues, nonverbal cues, and behavioral attitudes to detect deception—at, they claim, exceedingly high levels of accuracy. For a person who is falsely accused, this first impression may determine whether he or she is interrogated or sent home. Yet in laboratories all over the world, research has shown that people are only about 54% accurate in judging truth and deception; that training produces little, if any, improvement compared to naive control groups; and that police, customs inspectors, and other so-called experts perform only slightly better, if at all (Bond & DePaulo, 2006). In a study that examined performance in a criminal context, some lay participants but not others were randomly assigned to training in a popular law enforcement method of lie detection (Kassin & Fong, 1999). These students then watched videotaped interviews of suspects, some guilty and others innocent, denying their involvement in various mock crimes. As in past studies, observers could not differentiate between guilty and innocent suspects. Importantly, those who underwent training were less accurate, more confident, and more biased toward seeing deception than were those who had not received training. In a follow-up study with these same tapes, experienced detectives were tested—and they exhibited the same tendencies, making prejudgments of guilt, with confidence, that were frequently in error (Meissner & Kassin, 2002). At present, psychological scientists are seeking ways to improve human lie-detection performance. Some studies have shown that interviewers can boost their accuracy by withholding crime details while questioning suspects, trapping those who are guilty, but not those who are innocent, in inconsistencies when these facts are disclosed (Hartwig, Granhag, Strömwall, & Vrij, 2005). Other studies have suggested that because lying is more effortful than telling the truth, interviewers who tax a suspect’s cognitive load (e.g., by distraction or by having interviewees tell their stories in reverse order) can make more accurate true/false judgments by attending to effort cues such as hesitations (Vrij, Fisher, Mann, & Leal, 2006). WHY INNOCENT PEOPLE CONFESS Observational studies and surveys have shown that the modern American police interrogation—in which interrogators are legally prohibited from drawing confessions through violence, physical discomfort, threats, or promises—is a psychologically oriented process (Kassin et al., 2007; Leo, 1996; see Table 1). In their training manual, Inbau et al. (2001) recommend a multistep approach that is essentially reducible to an interplay of three processes: isolation, which increases anxiety and the suspect’s desire to escape; confrontation, in which the interrogator accuses the suspect of the crime, sometimes citing real or fictitious evidence to bolster the claim; and minimization, in which a sympathetic interrogator morally justifies the crime, leading the suspect to expect leniency upon confession. Situational Risk Factors Anecdotal evidence from DNA exonerations suggests that certain interrogation tactics exert too much influence. One potentially problematic tactic is the presentation of false evidence. American police are permitted to bolster their accusations by telling suspects that there is incontrovertible evidence of their guilt (e.g., a hair sample, eyewitness identification, or failed lie-detector test)—even if no such evidence exists. Can such trickery trap innocent people into confession? Over the years, basic research has shown that misinformation can alter people’s perceptions, beliefs, memories, and behaviors. With regard to confession, this hypothesis was tested in a laboratory experiment (Kassin & Kiechel, 1996). College students typed on a keyboard in what they were led to believe was a reaction-time study. At one point, subjects were accused of causing the computer to crash by pressing a key they had been instructed to avoid. They were asked to sign a confession. All subjects were truly innocent and all initially denied the charge. In some sessions but not others, a confederate said she witnessed the subject hit the forbidden key. This false evidence nearly doubled the number of students who signed a written confession, from 48% to 94%. As measured moments later, this manipulation also increased the number of subjects who actually believed TABLE 1 Ten Most Frequent Interrogation Practices, as Self-Reported by 631 North American Detectives (Kassin et al., 2007) Tactic Estimated Frequency Isolating the suspect from family and friends 4.49 Conducting interrogation in a small private room 4.23 Identifying contradictions in the suspect’s story 4.23 Establishing rapport/gaining the suspect’s trust 4.08 Confronting the suspect with evidence of guilt 3.90 Appealing to the suspect’s self-interest 3.46 Offering sympathy, moral justifications, & excuses 3.38 Interrupting the suspect’s denials and objections 3.22 Pretending to have independent evidence of guilt 3.11 Minimizing the moral seriousness of the offense 3.02 Note. Ratings were made on a 1-point (never) to 5-point (always) scale. 250 Volume 17—Number 4 False Confessions