The Feasibility of an Analytic Assessment to Identify False Confessions

Written By: Reid
Nov 01, 2012
The medical, psychiatric and therapy communities have a long history of using diagnostic "checklists" to evaluate the presence or absence of physiological or mental disease or degree of impairment. These analytic assessments are useful to help develop appropriate treatment protocols and often essential in determining whether or not specific diagnoses will be covered under certain insurance programs.

A process of analytic assessment has been applied to confessions in an effort to determine whether or not a confession is false. In the case of People v. Kowalski (July 2012), Richard Leo's analysis of the defendant's confession revealed that the confession was false. The court rejected Leo's opinion finding, in part, that Leo's analysis used "unreliable and unscientific methodology" in validating his procedures and that his assessment contained biases toward finding any confession false (Leo's assessment starts with the assumption that a confession is false.)

However, Leo does raise an intriguing question: Could a means of quantitatively assessing a confession be developed to scientifically predict whether or not a confession is true or false? This web tip will explore the feasibility of this concept. We were unable to obtain a detailed description of the process or formula that Leo utilized to identify a false confession so we will apply our knowledge and case law in developing a False Confession Checklist.

There is general agreement that the trustworthiness of a confession is influenced through three variables. The first is the level of corroboration contained within a confession. A simple statement, "I murdered my wife" is not a confession. Legally the confession must contain some level of corroboration. 1 Within this category are four items for analysis:

Independent - The suspect's confession revealed information unknown to the investigator which was independently verified as true. This represents the best form of corroboration.

Dependent - The suspect's confession contained information about the crime that was withheld from the public and the suspect. This represents a good form of corroboration, unless the innocent suspect was somehow provided the inside information.

Rationale - the suspect's confession offered mundane details or other information that lends credibility to the statement. This is the weakest form of corroboration, but may be the only form of corroboration available.

Faulty corroboration - the suspect's confession contains details that do not match known facts. This may result from an innocent person not knowing the details of the crime or a guilty person not wanting to reveal certain aspects of his crime.

The second variable to consider is the interrogation that led to the confession. The law prohibits interrogation tactics that "overbear a suspect's free will". These can be divided into two broad categories:

1. For Leo's interpretation of required corroboration within a confession see Leo, R. "False Confessions and the Constitution: Problems, Possibilities and Solutions" (2012) Electronic copy available at
Duress - An interrogation environment which is intolerable because of length, temperature, deprivation of biological needs, etc.

Coercion - Physical abuse and/or threats of physical harm or severe consequences combined with promises of leniency.

The final variable that influences the trustworthiness of a confession is the suspect. It is well documented that intrinsic characteristics within some suspects make them more susceptible to falsely confessing. Consequently, we can define high and low risk suspects within our checklist:

Low risk: Above the age of 15, an IQ above 64, no mental illness or personality disorders.

High risk: Below the age of 16, an IQ below 65, presence of mental illness or personality disorders.

The final checklist might look something like this, where the presence or absence of certain variables would be weighted as positive (true confession) or negative (false confession). If the total score is negative, this would indicate a strong possibility of a false confession. Here are the weighted parameters that could be used for each variable:
False Confession Checklist
Yes No
Independent corroboration 3 -1
Dependent corroboration 2 -1
Rationale corroboration 1 -2
Faulty Corroboration -3 0
Duress -2 2
Coercion -2 2
High risk suspect -2 2

To analyze the feasibility the False Confession Checklist consider the following interrogation: An 18- year-old-suspect is arrested for the murder of his girlfriend. The two were seen arguing at a party at around 11:00 pm and her body was discovered in a parking lot near the party the next morning. The suspect has attention deficit disorder and is on probation for a previous battery.

The victim was stabbed 12 times and there was no weapon recovered from the crime scene. The suspect was interviewed at his home and consented to a search of his bedroom. A bloody shirt and knife were discovered under the suspect's bed and the suspect was taken into custody. An inspection of the sink in the suspect's bathroom revealed bloody water in the trap.

The suspect waived his Miranda rights and was interrogated. The suspect initially maintained that he cut himself with the knife and that is how his shirt got bloody. The suspect did have a cut across his left index and middle finger. He stated that he hid the shirt and knife under his bed because he did not want his father to know that he was careless with the knife. During the interrogation the investigator suggested that the victim was somewhat to blame for causing the suspect to lose emotional control. In addition, the suspect was falsely told that the blood on his shirt was determined to be from his girlfriend. After three hours of interrogation, the suspect confessed to killing the victim.

When the suspect confessed, he acknowledged stabbing his girlfriend multiple times out of anger using a jack knife that he had in his pocket. The suspect explained that after he left the victim's body in the parking lot he drove home to his parents' house and hid the shirt and knife under his bed with the intention of burying them in his back yard the next day. When asked how he got the blood off of his hands he said he wiped his hands on his shirt. The suspect's confession did not mention washing blood off his hands in his bathroom.

Using the False Confession Checklist, this confession contains no independent or dependent corroboration (-2). The primary reason for this is that the suspect was told during the interview that his girlfriend had been stabbed to death and found in a parking lot near the party they attended. The confession does contain rationale corroboration (+1). Because the suspect failed to mention washing his hands in his bathroom, some evaluators may claim that this is faulty corroboration (-3).

Certainly a three hour interrogation does not represent duress (+2) but because the investigator reduced the moral seriousness of the crime by blaming the victim and also lied to the suspect about incriminating evidence, some evaluators would describe this as coercion (-2). The suspect's diagnosis for ADD might cause some evaluators to score him as a high risk suspect (-2).

The confession's total score is minus six, rendering it false. Yet any investigator, judge or juror using common sense would recognize this as a true confession. The above checklist could be altered by having each item judged on a five-point scale, or the values of each item could be tweaked. However, a fundamental problem still exists -- none of these criteria, combined or taken individually, has the potential to predict whether a particular confession is false.

Occasionally, information is developed that positively proves a confession to be false. This may be in the form of finding the actual perpetrator through DNA evidence or by establishing an air-tight alibi for the suspect, e.g., the suspect was serving a jail sentence at the time of the crime. In reviewing these confessions, interrogations and suspect characteristics there often are common denominators. But in evaluating hundreds of thousands of confessions that are certainly true, some of these same common denominators are present in many of those confessions.

The concept of a "false confession checklist" is attractive to the defense. However, as this web tip reveals, the feasibility of developing a reliable and valid diagnostic checklist to identify false confessions is impractical. Interrogation and confessions concern human behavior which is far more complex and unpredictable than a therapist's evaluation of a patient's range of motion or a psychiatrist's diagnosis of clinical depression.

Recognizing the difficulties of assessing confession evidence, the Supreme Court established the concept of evaluating the "totality of circumstances" when determining the admissibility of a confession. Expert witnesses may be useful in educating the court about factors that influence the trustworthiness of a confession. However, the final weight a jury places on confession evidence should not be based on an unscientific, unreliable checklist. It should be determined by an objective and unbiased assessment of the corroboration within the confession, the presence of improper interrogation tactics and identifying suspect characteristics that may have resulted in a false confession.
Permission is hereby granted to those who wish to share or copy this article. In those instances, the following Credit Statement must be included "This Investigator Tip was developed by John E. Reid and Associates Inc. 800-255-5747 /" Inquiries regarding Investigator Tips should be directed to Toni Overman