Cadavers and Cognitive Bias

A few weeks ago I attended a continuing legal education webinar on strangulation. The presenter shared a book she recommended on forensic pathology, Working Stiff: Two Years, 262 Bodies, and the Making of a Medical Examiner, by Dr. Judy Melinek and T.J. Mitchell. At some point in a criminal defense lawyer’s career, we all confront gruesome autopsy photos and the report of a forensic pathologist full of medical terms completely foreign to laypeople.

For years I assumed forensic pathologists simply examined the body and made a determination of the cause and manner of death based solely on their examination. It was not until I read Dr. Melinek’s memoir that I realized how much nonmedical contextual information played a role in the determination as to what caused a person’s death and by what manner.

In the vast majority of criminal cases, this contextual information is provided by - you guessed it - the police. While this information could have some relevance to the determination of the cause and manner of death, what about nonmedical contextual information that is not relevant to the cause and manner of death? What role, if any, does it play?

Just a couple of weeks ago, a study was published that sought to answer that very question. Dr. Itiel Dror and several others created a study to determine whether cognitive bias plays a role in forensic pathology decision-making. The article detailing the study and its results, Cognitive Bias in Forensic Pathology Decisions, published in the Journal of Forensic Sciences, begins by explaining that forensic pathologists must consider nonmedical contextual information to determine the manner of death. But can forensic pathologists effectively remove irrelevant contextual information from their decision-making?

The study presented a hypothetical death case of a young child to a sample of qualified forensic pathologists. All received identical medical information. But one group was told the child was black and the mother’s boyfriend was the caretaker, while the other group was told the child was white and the child’s grandmother was the caretaker. The child’s race and the familial role of the caretaker was irrelevant to the determination of the manner of death. And yet the authors of the study discovered the forensic pathologists’ decisions were “noticeably affected” by the medically irrelevant contextual information.

The authors acknowledge several limitations with the study but note that the study still represents the first of its kind to demonstrate that irrelevant nonmedical contextual information does bias a forensic pathologist’s decision-making. They recommend that the forensic pathology community adopt procedures to minimize bias. Moreover, they recommend greater transparency regarding what contextual information the forensic pathologist relies upon. This could be achieved by forensic pathologists explicitly stating in their reports what nonmedical contextual information, if any, they were given and how it may have impacted their conclusions.

In criminal cases especially, juries rely heavily on experts when considering whether a death was intentional or accidental, for example. The more a determination as to the manner of death is dependent on nonmedical contextual information, the weaker the determination is. Factfinders tasked with deciding the weight to give to certain evidence must be made aware of this information.

In preparing for trial, defense attorneys should learn what information the forensic pathologist was provided about the decedent, who provided the information, and what effect, if any, the information had on the forensic pathologist’s determination of the cause and manner of death.