Please provide a brief introduction about yourself.
My name is Julian Marchionda, and I’m a recent graduate of the University of Michigan where I received a B.A. in Psychology. During my time in Ann Arbor, I became interested in learning about processes, heuristics, and biases that affect decision-making and influence important outcomes downstream in contexts like medicine, law, and school. Most recently, I completed my honors thesis under the guidance of Dr. Priti Shah and Dr. Madison Fansher.
Tell us about your research project.
My research project focuses on a type of heuristic thinking we’ve dubbed numerical category bias, where people judge continuous data differently when it’s sorted into categories. For instance, on a road with a 55 mph speed limit, the difference between 54 and 56 mph may be judged as larger than 56 and 58 or 52 and 54, though the raw differences are identical (2 mph).
Specifically, my thesis investigates this bias in three types of legal scenarios: (1) involving blood alcohol content and outcomes such as impairment and responsibility for an accident, (2) parole risk and an associated likelihood of reoffending if granted parole, and (3) IQ scores and outcomes such as an appropriate punishment for a crime and degree of control that a defendant had over their actions.
Although it was not consistent across all experiments, we found evidence that people judge equal differences between numbers as significantly larger (via a proxy variable) when those values straddle legal or categorical cutoffs (e.g., .08% for blood alcohol content, and “low,” “moderate” and “high” risk of reoffending).
What inspired your interest in this topic?
Last summer, I burned through Misbehaving, a book by Richard Thaler that outlines ways in which people make decisions that deviate from the economically “rational” agent. That previous spring, I also took courses in cognitive science and ethics where we read articles by legal scholars such as John Mikhail. Reading about how our psychology interacts with decisions about culpability, morality, and interferes with rational decision-making made me excited about completing my own project investigating something similar.
I had already worked on a project about a year prior where a numerical category bias was demonstrated, and felt that it could be easily applied to legal scenarios as well. Given that categories and thresholds are frequently defined in law, it felt like a natural fit.
This award recognizes the broader impact of your research. What are the societal implications of your work?
While I was not intending to evaluate the legitimacy of any kind of legal statute or policy in this project, I believe these findings could be relevant for understanding ways in which juries, judges, and other evaluators interpret quantitative evidence, even using numerical cutoffs as much stronger pieces of information than the rest of the evidence may warrant. Given that legal decisions are high stakes, I believe interpreting numbers as unbiased as possible is an important concern.
What are your next steps academically/professionally?
Beginning in June, I’ll be working as a paralegal at a private credit firm in Chicago. After a few years of work, I plan to apply to law school.
