Interrogation rooms remain some of the most secretive locations in the United States legal system. Police undertake interrogation to discover the truth about a crime. Police, along with society at large, want guilty people to confess and innocent people to resist. The stakes are particularly high because a confession is even more powerful than eyewitness testimony in a criminal trial (Kassin & Gudjonsson, 2004). A confession increases the likelihood of guilty verdicts even when the confession is coerced through threats or promises and even when judges admonish (i.e., instruct) jurors to ignore the confession (Kassin & Gudjonsson, 2004). Given the severity of a mistaken verdict, psychologists have investigated the process of interrogation and the possibility of false confessions.
Many factors increase the likelihood of true and false confessions. The physical structure of the room, a small, uncomfortable, soundproof space with an evident one-way mirror (see Kassin, 1997), interacts with the social influence exerted on the suspect by the interrogator to produce a situation in which confessions, both true and false, are likely. The interrogator initiates the process with a strong statement of the suspect’s guilt and then proceeds through a process that includes interrupting all denials, preventing the suspect from tuning out, and then showing sympathy and empathy for the suspect while encouraging the suspect to confess. These methods produce confessions, and researchers have argued that this high degree of social influence is necessary to ensure that guilty suspects confess, but the possibility of false confessions raises concerns (Kassin, 1997).
The number of annual false confessions remains unknown. Estimates vary widely, and some scholars argue that these numbers cannot be known (Kassin & Gudjonsson, 2004). In interviews of convicted defendants, both guilty and innocent defendants often maintain that they are innocent, and it remains extremely difficult to evaluate guilt in any absolute sense. When biological evidence exists and points to innocence, a confession may be overturned. The Innocence Project (2007) reports that in just over a quarter of their cases, a defendant incriminated him- or herself, confessed, or pleaded guilty when he or she was innocent.
Although it is hard to imagine why an innocent person might confess, researchers have delineated three possibilities. First, suspects may make voluntary false confessions. For example, in 1932 over 200 people confessed to kidnapping the Lindbergh baby. Second, innocent suspects may confess even though they know they are innocent; these coerced-compliant false confessions (Kassin & Gudjonsson, 2004) occur when participants believe that the benefits of confession outweigh the costs. For example, a coerced-compliant false confession could occur if the suspect believes that a conviction is inevitable and that a confession will lead to reduced punishment. Third, innocent suspects may confess and may truly believe that they are guilty. These coerced-internalized false confessions are most controversial and conceptually difficult (see Kassin, 1997). Suspects genuinely believe that they are guilty of crimes they did not commit, whether accusations are severe (e.g., the case of Paul Ingram) or minor, such as accidentally pressing an incorrect key during a computer task (Kassin & Gudjonsson, 2004).
Questions of true or false confessions would present fewer difficulties if the impact of confessions on juries were not so great. These issues are compounded when biases and beliefs of interrogators enter the mix. Interrogators overwhelmingly tend to believe that the suspect is guilty (Kassin, 1997). Although the effects of these biases on the interrogation process are difficult to measure in archival or ongoing legal cases, when student interrogators interviewed guilty or not guilty students, the longest and most intense interrogations occurred when the researchers led the student interrogator to believe that the innocent suspect was actually guilty (Kassin & Gudjonsson, 2004). Additionally, police interrogators endorse the “myth” (Kassin & Gudjonsson, 2004, p. 57) that they could recognize a false confession. When Kassin’s critics challenged his conclusions by arguing that trained observers would be more accurate than untrained observers, Kassin trained observers who were then less accurate but more confident than untrained observers (Kassin & Fong, 1999). These findings generated particular concern because confident expert witnesses, even those who are, as in this case, less accurate, have more influence on jurors’ decisions (Kassin & Fong, 1999).
These problems reach deeply into the legal system, but some solutions are available. Limiting the time that a suspect can spend in custody or in interrogation and working to reduce potential sleep or food deprivation could reduce risks. Extra precautions are required when the suspect is a member of a vulnerable population such as a child or an individual with a mental illness or developmental disabilities. Many interrogation researchers argue that requiring police to videotape confessions can reduce the likelihood of false confessions, protect suspects, and protect police from allegations of coercion (Kassin & Gudjonsson, 2004). These reforms are in place in some states and many local police departments, but many interrogations occur without the protection a video recording can provide.
Return to overview of Psychology and Law.