Professor Richard Leo has posted the above-titled article on SSRN. Download here. The abstract states:
The problem of police interrogation contamination (disclosing or leaking of non-public facts) is pervasive in documented false confessions leading to wrongful conviction. The presence of unique and detailed crime facts in a false confession creates the illusion that the defendant volunteered inside information about the crime that “only the true perpetrator could have known,” thus seemingly corroborating a false confession as verifiably true. This article argues that confession contamination occurs because (1) the guilt-presumptive psychology of American police interrogation is designed to trigger and perpetuate confirmation biases that (2) lead investigators – seemingly inadvertently – to provide detailed case information to suspects as part of their pre- and post-admission accusatory interrogation strategies, but (3) has no internal corrective mechanism to catch or reverse investigators’ misclassification errors or their confirmatory interrogation techniques. American investigators presume not only the guilt of the suspects they interrogate, but also their guilty knowledge of the crime facts. Interrogators almost invariably disclose detailed case information to presumed guilty suspects through the use a variety of information-conveying techniques – accusations, attacks on denials, evidence ploys, feigned omniscience, inducements and scenarios. In the pre-admission stage, investigators convey and disclose detailed case information as part of their interrogation strategies to move the suspect from denial to admission, whereas in the post-admission stage, interrogators do so as part of their strategies to elicit a complete and persuasive narrative of his guilt. Interrogation contamination corrupts the truth-seeking process and increases the risk that a false confession will lead to a wrongful conviction. In order to eliminate it, investigators must dispense with the presumption of guilt that currently underlies interrogations, seek to better understand the multiple sources of their misclassification errors, and create internal corrective mechanisms that help them identify the confirmation biases and tendency toward tunnel vision that lie at the heart of American-style police interrogation.