Confessions as Guaranteed Truths

In this study, Hasel and Kassin (2008) utilized 260 undergraduate students in a two phase experiment. Initially the students were told they would be participating in a persuasion technique study; however, once brought to the lab they witnessed an individual steal a laptop from the experimenter’s desk. Upon returning to the lab, the experimenter explained to the participants that he or she knew a study on criminal investigations was taking place. The experimenter reported that he or she would proceed by gathering eyewitness accounts, questioning a set of suspects and ultimately deciding whom to charge for the crime. The experimenter enlisted the assistance of all the participants in solving the crime. The participants were asked to view a six person photographic line-up in which the actual thief was missing and determine which, if any, of the individuals represented in the line-up committed the crime. Lastly, participants were asked to rate the confidence they had in their line-up decision.

In the second phase of the study participants were asked to return to the lab two days after the theft. Depending on whether the participants had chosen a suspect from the six person photographic line-up or not, the experimenter proceeded in different ways. If the participant had chosen an individual from the line-up the experimenter, by random assignment, indicated one of the following: that the chosen individual had indeed confessed, all of the suspects had denied participating in the crime, the chosen individual had denied involvement, or another suspect had confessed to the crime. If the experimenter presented a confession he or she also produced a handwritten and signed admission of the crime. If the participants had not chosen a subject from the six person photographic line-up, the experimenter proceeded, again by random assignment, to indicate one of the following: that all the suspects had denied participating in the crime, an unspecified suspect from the line-up had confessed, or a specified suspect from the line-up had confessed. After each of these various scenarios the experimenter continued by asking all the participants to again rate their confidence level in their original decision. The experimenter then gave participants the option to reconsider their first choice, regardless of if they had originally chosen a suspect from the line-up or not. Finally, participants were asked to rate their confidence in their second choice.

After considering attrition and exclusion of subjects because of subject suspiciousness, familiarity with the thief or other reasons that would compromise the study, results of 206 participants were used. A significant number of participants who were told their choice from the original line-up had confessed reported an increase in confidence in their choice. Interestingly, a majority of participants who were told that a different suspect confessed changed their mind and reported the confessor as the criminal. Although the authors mention possible flaws with their research due to the use of college students as participants, a mock crime and the lack of high-stake consequences rendering the identification decisions moot, the study is informative and future research in this area must be borne out of this body of work. This study should be replicated using different populations and simulating more various types of crimes. The issue of false confessions and tainted eyewitness accounts is of incredible importance not only for the integrity of our legal system, but for the thousands of individuals who are accused of crimes on a daily basis.

To cite this review, please use this reference:
Levi-Minzi, M. (2009). Confessions as Guaranteed Truths. Psychology Alert (3). http://psychologyalert.com/2009/01/confessions-as-guaranteed-truths.html