Why do people falsely confess?

Custodial interrogation: What are the background factors associated with claims of false confession to police?

In this article, we review the work of Gudjonsson, et al., on the important topic of false confessions to police and the background factors which were identified as associated. In particular, the researchers explored the relationship between false confessions during detention and reporting of harmful life events (e.g., abuse, self-harm, and previous psychological/psychiatric treatment).

A total of 1,896 students in further education in Iceland (ages 15 to 24) worked on the survey and said they had experienced police interrogation. The research was completed with a detailed questionnaire about their background, drug use, and psychological/psychiatric treatment. The result was, of the 1,896 participants, 138 (7.3%) reported they had given a false confession to the police. According to the authors, “Out of 27 variables entered into a logistic regression analysis, eight significant predictor variables were identified in the model. These were largely associated with multiple victimization (e.g., bullying, death of a significant other, being a victim of violence) and substance abuse (i.e., having attended substance abuse treatment, use of LSD).”[i] The researchers concluded that the findings suggested that multiple exposure to unpleasant or traumatic life events is associated with reported false confessions during the examination.

The key words we will deal with in this paper are: stressful events, risk, victimization and substance abuse.

Even a cursory review of the scientific literature supports the presence of a large number of false confessions. Several excellent studies on the topic include Gudjonsson, 2003; Gudjonsson, 2006; and Kassin & Gudjonsson, 2004. In their texts reviewing a large number of proven cases of false confession in the United States, Drizin and Leo (2004) argued that they represented “just the tip of a much larger iceberg” (p. 919). Also consider the comments from Kassin and Gudjonsson (2004): “Since no one knows the frequency of false confessions or devised an adequate method for calculating accurate incident rates, there is a multi-year debate about numbers." (p.48). Much of the research on false confession rates is in Iceland. Some of the examples of false confessions we have are from Gudjonsson and Sigurdsson (1994) and Sigurdsson and Gudjonsson (1996) where they asked Icelandic prison inmates if they had ever falsely confessed to the police. In both studies, 12% claimed to have made a false confession to the police at some time in their lives. Sigurdsson and Gudjonsson (2001) carried out a discriminant function analysis among 62 false confessors and 447 other prison inmates using 17 psychological and 16 criminological and substance misuse variables. The number of previous imprisonments and a low score on the Gough Socialization Scale were the two best predictors. It was concluded that antisocial personality traits and the severity of criminal behavior were good predictors of false confessions made during detention. There were also two community studies in which students from “Icelandic faculties and universities who reported being questioned by the police (3.7% and 1.2%), that is, claimed to have given a false confession” (Gudjonsson, Sigurdsson, Bragason, Einarsson, and Valdimarsdottir, 2004a; Gudjonsson, Sigurdsson and Einarsson, 2004b).

In a survey in which 10,472 students participated in further education in Iceland, a false confession rate of 7.3% of those surveyed by the police was established with a false confession rate of 1.6% for the total sample (Gudjonsson, Sigurdsson, Asgeirsdottir and Sigfusdottir, 2006). The researchers explored the association of creating false confessions with variables related to mental state that included anxiety, depression, anger, self-esteem, attitude toward school, and parental support, as well as involvement in delinquent behavior and delinquent behavior of friends. “False confessors felt more anxiety, depression, anger problems, lower self-esteem, less parental support and more delinquency during the previous year, and more delinquency among friends. It represents that they were more delinquent and emotionally disturbed than the other participants” per Gudjonsson et al. (2006).

In the study under review in this article, Gudjonsson et al. worked to expand previous research by researching the connection between claimed (unsubstantiated) false confessions and past experiences of a series of unpleasant or traumatic life events. Researchers have associated the general theory developed by Agnew (1992) with the general deformation relevant to crime and delinquency. One of the main layers according to this theory is the history of aversive life events that creates negative emotional reactions such as frustration, anger, and impairment in the ability to deal effectively with stressful life events, making people vulnerable to delinquent actions. In the currently reviewed article, the authors hypothesized that reporting false confessions during interrogation is associated with negative life events and tensions, including abuse (violent and sexual), family death, parental divorce, and serious substance abuse. These life events comprise a major continued emotional strain for the individual concerned, which places him or her in a position of frustration and avoidance when faced with the demands of a police interrogation.


Participants: Participants were students (1,896) in further education in Iceland who had been selected from a previous study on 10,472 students (Gudjonsson et al., 2006) on the basis that they had reported having at some time in their lives been interrogated as a suspect at a police station (in Iceland criminal responsibility is assumed from the age of 15). The students were from all 38 secondary education colleges in Iceland. There were 1,359 (72%) males and 529 (28%) females in the study (eight further participants did not specify their gender on the questionnaire). The average age was 18.09 years (range 16 – 24, SD¼ 2.00).

Instruments: All data used in the study come from a population survey of Icelandic high school students in 2004. The Icelandic Center for Social Research and Analysis conducted the survey in collaboration with the State Agency for Child Protection, the Ministry of Education and the Icelandic Institute of Public Health. The questionnaire contained 169 subjects, ranging from questions about family, education, and social background to sexual experiences and abuse, various forms of victimization, previous use of substances, delinquency, and delinquency by friends. All participants were asked about their experience of police interrogation and if they had ever given a false confession to the police. In the present study, 27 categorical items relating to adverse background events and substance misuse were included in order to investigate how they discriminated between the participants who had reported a false confession during interrogation and those who did not. The 27 items were selected on the basis of probable theoretical relevance to false confessions (Gudjonsson, 2003). In relation to substance misuse, the types of drug use included in the study reflected the main illicit drugs used in Iceland (Hibell et al., 2000).

The 27 items were as follows:

1. parents separated or divorced, rated yes or no

2. parent or a sibling died, rated yes or no

3. serious arguments with parent, rated yes or no

4. run away from home for more than one day, rated yes or no

5. been rejected by some of male or female friends, rated yes or no

6. death of a friend, rated yes or no

7. recent school performance, rated satisfactory or not satisfactory

8. been expelled from school, rated yes or no

9. being bullied in school, rated yes or no

10. experience of violence at home, rated yes or no

11. experience of violence at school, rated yes or no

12. experience of violence on the street, rated yes or no

13. history of sexual abuse, rated yes or no

14. drinking of any kind of alcohol, rated yes or no

15. ever used hashish, rated yes or no

16. ever used amphetamine, rated yes or no

17. ever used cocaine, rated yes or no

18. ever used LSD, rated yes or no

19. ever used ecstasy, rated yes or no

20. ever taken sleeping pills or sedatives not prescribed by a medical doctor, rated yes or no

21. attended substance misuse treatment, rated yes or no

22. history of self-harm, rated yes or no

23. history of serious suicide thoughts, rated yes or no

24. history of attempted suicide, rated yes or no

25. has asked a psychiatrist for help regarding personal matters during the last 12 months, rated yes or no

26. has asked a psychologist for help regarding personal matters during the last 12 months, rated yes or no

27. has taken or does take any psychotropic medicine, rated yes or no.


Students were approached by teachers in class and asked to participate in a survey on youth. They were told that their answers were anonymous and confidential. The survey took an hour to complete, and then the students enclosed their completed questionnaires in blank envelopes. All participants throughout Iceland completed the study on the same day.


Significant differences between the two groups were identified on 25 of the variables (no significance was found for items 3 and 14). Treatment of substance abuse was reported in only 5% of cases of non-recognition, but in 37% of false cases of confession. Similarly, rates for LSD use were 4% and 29%, respectively. Other significant differences have been found in the history of sexual abuse (9% and 31%, respectively), experiencing violence at home (11% and 30%, respectively), expelled from school (12% and 37%, respectively), experiencing the death of a parent, brother, or sister (8% and 29%, respectively), and bullying at school (3% and 16%, respectively). Regarding alcohol, differences were observed between groups (96% and 98%, respectively), cannabis use (51% and 62%, respectively), serious quarrels with parents (57% and 65%, respectively), and parental divorce (30% and 40%, respectively).

To investigate which variables were most influential, logistic regression analysis was performed on the data. Pearse, Gudjonsson, Clare, and Rutter (1998) used the method in a similar case with respect to variables that predicted false confession versus denial. The method aimed to formulate a statistical model that identified the predictive value of the independent variables in relation to the odds an individual falls into a particular group (dependent variable), which in the present study is having a history of making a false confession during custodial interrogation. The outcome variable was whether or not the participant ever reported a false confession to the police during the interrogation in prison (i.e., a non-false confessor versus a false confessor). A forward stepwise method (‘forward likelihood ratio’) was used to enter the data (Field, 2003). The authors had 27 variables entered, and of those 27 entered, 8 had a significant contribution to the model. They found statistical significance for each of the eight variables entered in the final step of the model. The results suggested that the individual had a history of false confessions that increased more than three times if the suspect reported they had used LSD (odds ratio = 2.94), been treated for substance abuse (odds ratio = 2.67 ), experienced street violence (odds ratio = 2.27), experienced the death of a parent or sibling (odds ratio = 2.14), expelled from school (odds ratio = 1.98), physically assaulted at home (odds ratio = 1.90), and had negative school achievement (odds ratio = 1.65). The authors had an additional separate analysis where they added three continuous variables from Gudjonsson et al. (2006) studies (depression, delinquency, and peer delinquency). The results indicated only delinquency of peers entered the model in the final step and it had the lowest odds ratio.


The present findings confirm the findings of Sigurdsson and Gudjonsson (2001), Gudjonsson et al. (2004a), Gudjonsson et al. (2004b), and Gudjonsson et al. (2006) that individual differences are important in discriminating between claimed false confessors and other suspects interrogated by the police. The authors found significant differences for 25 of the 27 life events. Most of the serious adverse life events, such as the death of a parent or sibling, sexual abuse, experiencing violence at home, being bullied at school, and being expelled from school, were quite rare among the non-false confessors (range 3 - 12%), but they were common among the claimed false confessors (range 16 - 37%). Many had also experienced serious suicidal thoughts or attempts and had been in contact with mental health professionals. Factors can be seen as psychological vulnerabilities that can increase the risk of false confession. Harassment, although less impactful than some factors, was a strong predictor in logistic regression analysis. Only 5% of the non-false confessors had attended substance misuse treatment in contrast to 37% of the claimed false confessors. Also, LSD was very rare (4%) among the non-false confessors but common (29%) among the claimed false confessors. A surprisingly high number of participants reported experiencing violence at school and in the street. Of those who reported to be false confessors, they reported being assaulted in the street in 62% of cases (the corresponding figure for school violence was 37%). The findings are very important because they give strong support to the hypothesis that many claimed false confessors have experienced a number of serious adverse life events and this may make them vulnerable to making a false confession during interrogation.

Of course, we are dealing with correlational data from a cross-sectional study, and it is not possible to separate cause and effect. Recent independent findings from a study of 47 suspects at police stations in Iceland corroborate the findings in the present study (Sigurdsson, Gudjonsson, Einarsson, & Gudjonsson, 2006). When interviewed immediately after police interrogation by a psychologist, four (9%) of the suspects claimed to have made a false confession to the police during the interrogation due to their immediate need or psychological problems. All four reported significant symptoms of clinical depression. This study has limitations. First, the findings are dependent on self-report. It was not possible to corroborate the participants’ accounts through official records and some participants may not have been accurate or honest about their involvement with the police. Second, in the present study the reasons for giving a false confession to the police were not investigated. Third, the type and severity of the crime the students reported they had confessed to falsely are unknown. Fourth, gender differences were not specifically investigated, because in the previous study (Gudjonsson et al., 2006) no differences in the rate of false confession were found once the rate of offending was taken into consideration. Future research should introduce a clinical interview with participants who report false confessions to the police and establish a primary clinical diagnosis (Sigurdsson & Gudjonsson, 2001).

In conclusion, the present study is unique in that it is the first to investigate the direct association between serious adverse life events and claims of having made a false confession to the police. Many of the claimed false confessors in this study reported a number of very adverse life events, which were far less frequent among the non-false confessors. These findings can be interpreted within Agnew’s (1992) general strain theory.


Agnew, R. (1992). Foundation for a general strain theory of crime and delinquency. Criminology, 30, 47–87.

Drizin, S. A., & Leo, R. A. (2004). The problem of false confessions in a post-DNA world. North Carolina Law Review, 82, 891–1007.

Field, A. (2003). Discovering statistics using SPSS for Windows. London: Sage.

Gudjonsson, G. H. (2003). The psychology of interrogations and confessions: A handbook. Chichester, UK: John Wiley & Sons.

Gudjonsson, G. H. (2006). Disputed confessions and miscarriages of justice in Britain: Expert psychological and psychiatric evidence in the Court of Appeal. The Manitoba Law Journal, 31, 489–521.

Gudjonsson, G. H., & Sigurdsson, J. F. (1994). How frequently do false confessions occur? An empirical study among prison inmates. Psychology, Crime and Law, 1, 21–26.

Gudjonsson, G. H., Sigurdsson, J. F., Asgeirsdottir, B. B., & Sigfusdottir, I. D. (2006). Custodial interrogation, false confession and individual differences: A national study among Icelandic youth. Personality and Individual Differences, 41, 49–59.

Gudjonsson, G. H., Sigurdsson, J. F., Bragason, O. O., Einarsson, E., & Valdimarsdottir, E. B. (2004a). Confessions and denials and the relationship with personality. Legal and Criminological Psychology, 9, 121–133.

Gudjonsson, G. H., Sigurdsson, J. F., & Einarsson, E. (2004b). The role of personality in relation to confessions and denials. Psychology, Crime and Law, 10, 125–135.

Hibell, B., Anderson, B., Ahlstro¨m, S., Balakireva, O., Bjarnason, T., Kokkevi, A., & Morgan, M. (2000). The 1999 ESPAD report: Alcohol and other drug use among students in 30 European countries. Stockholm: Centralfo¨rbundet fo¨r alkohol-och narkotikaupplysning (CAN).

Kassin, S. M., & Gudjonsson, G. H. (2004). The psychology of confessions: A review of the literature and issues. Psychological Science in the Public Interest, 5, 33–67.

Pearse, J., Gudjonsson, G. H., Clare, I. C. H., & Rutter, S. (1998). Police interviewing and psychological vulnerabilities: Predicting the likelihood of a confession. Journal of Community and Applied Social Psychology, 8, 1–21.

Sigurdsson, J. F., & Gudjonsson, G. H. (1996). The psychological characteristics of ‘false confessors’: A study among Icelandic prison inmates and juvenile offenders. Personality and Individual Differences, 20, 321–329.

Sigurdsson, J. F., & Gudjonsson, G. H. (2001). False confessions: The relative importance of psychological, criminological and substance abuse variables. Psychology, Crime and Law, 7, 275–289.

Sigurdsson, J. F., Gudjonsson, G. H., Einarsson, E., & Gudjonsson, G. (2006). Differences in personality and mental state between suspects and witnesses immediately after being interviewed by the police. Psychology, Crime and Law, 12, 619–628.

[i] Gudjonsson, G. H., Sigurdsson, J. F., Asgeirsdottir, B. B., & Sigfusdottir, I. D. (2007). Custodial interrogation: What are the background factors associated with claims of false confession to police? The Journal of Forensic Psychiatry & Psychology, 18(2), 266-275.

238 views10 comments

Recent Posts

See All