Photo: VieveTru, CC BY-SA 4.0, Wikimedia Commons
Russell A. Poldrack
Stanford University
- Research field
- Cognitive neuroscience, the methods and reproducibility of neuroimaging
Russell Poldrack studies how the brain supports learning and decision making, and he has spent much of the last two decades asking how trustworthy the methods of brain imaging are. He named the problem of reverse inference in 2006, founded the OpenfMRI database, now OpenNeuro, and co-authored Neurosynth, two databases that make imaging data open, and in 2017 led a review that measured the sample sizes of fMRI studies, catalogued the ways in which an analysis can produce a false result, and set out the remedies: power calculations, preregistration, shared data and code, and validated corrections for multiple comparisons. The unit presents this review after its sections on the statistics of a brain image and the 70-team NARPS study, which he later helped to lead.
In the masterclass
Section 5, after the NARPS study, presents his 2017 review of what makes an imaging result trustworthy.
The research is presented in Unit 13: Brain Imaging and Stimulation, in a research spotlight box that explains the result at the level of the unit and links it to what you have just learned.