Date/Time
Date(s) - 11/03/2025
3:30 pm - 4:30 pm

Speaker: William Marshall (Brock University)
Dr. William Marshall is a transdisciplinary researcher, specializing in statistical analysis of neuroimaging data to advance neuroscience research problems. After completing his PhD training in Statistics at the University of Waterloo, Dr. Marshall held a post-doctoral position in the Center for Sleep and Consciousness at the University of Wisconsin – Madison and is currently an Associate Professor in the Department of Mathematics and Statistics at Brock University.

Location: BSB 121

Title: Regression to the Mean with Examples in Neuroscience Research

Abstract: Regression to the mean is a statistical phenomenon whereby extreme measurements tend to be followed by less extreme measurements. Although it has been known since it was studied by Galton in the 1870’s, it remains an often misunderstood or overlooked aspect of experiment design and analysis interpretation. The first part of the talk will introduce the concept of regression to the mean with some intuitive examples and then provide a formal statistical treatment. The second part of the talk will explore regression to the mean for two contemporary neuroscience problems: (1) evaluating the synaptic homeostasis hypothesis, which proposes that the largest synapses are protected from homeostatic effects, and (2) assessing evidence for unconscious processing, where experimental results often rely on changes in performance following threshold-based selection. In both cases, regression to the mean can obscure true effects, and lead to misinterpretation of the results.