Jonathan Pillow

April 27, 2018

Official Story

Jonathan Pillow received his Ph.D. from NYU, where he worked in the laboratory of Eero Simoncelli on statistical modeling of spike trains in the early visual system. He then moved to London for a postdoctoral fellowship at the Gatsby Computational Neuroscience Unit at UCL, and in 2009 became an assistant professor at the University of Texas at Austin in the department of Psychology. In 2014, Jonathan moved to Princeton University, where he is currently an associate professor in the Princeton Neuroscience Institute, Psychology department, and the Center for Statistics and Machine Learning. Jonathan's current research sits at the border between neuroscience and statistical machine learning, focusing on computational and statistical methods for understanding how large populations of neurons transmit and process information.

Unofficial Story

Jonathan grew up in Phoenix, Arizona, the son of a librarian and a preschool teacher. His principal childhood loves were sports and arguing, and he overcame the traumatic experience of failing his third grade multiplication test to find that he also enjoyed math. During the summer before his senior year in high school, Jonathan attended a 5-day summer camp in science and engineering and returned to announce, to the dismay of his parents, that he intended to major in philosophy. He attended college at the University of Arizona in Tucson, where he ran on the track and cross country teams and double-majored in philosophy and (in a kind of peace offering to his parents, who urged him to think more pragmatically about his future) math. During the fall semester of his senior year, he was consumed by a rising sense of panic as it slowly dawned on him that college would not last forever. Casting about for a senior thesis advisor, Jonathan had the good fortune to meet Rich Zemel, an assistant professor in the Psychology department who worked on neural population coding. Rich gave Jonathan a paper to peruse and suggested they meet up the following week to discuss ideas; Jonathan spent that week mostly not reading the paper, then went to the meeting to tell Rich that, thank you very much for the offer, but he hadn't been able to understand the paper, and he'd decided to do a senior thesis with a philosophy professor instead. Luckily for Jonathan, Rich clarified that he hadn't meant for Jonathan to actually understand the paper---just get the gist of ideas Rich was thinking about, and that he thought that Jonathan had the right background for that kind of work if Jonathan would only apply himself. Grateful for the vote of confidence, Jonathan shut the door to philosophy once and for all and, with Rich's guidance, applied to Ph.D. programs in neuroscience. Despite a mere three months of research experience and an application essay that focused largely on how electrified he was by Roger Penrose's writings on quantum theories of consciousness, he was accepted. Although Jonathan believed that his future lay in neuroscience, he --- in yet another decision that he struggled to explain to his parents --- deferred graduate school to spend a year studying arabic and North African literature in Morocco. When at last he arrived at NYU, Jonathan struggled to overcome a total lack of science coursework during college, and was grateful to classmates who took the time to explain to him basic concepts from biology like what phosphorylation was and why he should care. Jonathan spent a long time in lab rotations before joining a thesis lab, studying illusory contour perception with Nava Rubin and learning to patch neurons with Alex Reyes, an experience that convinced him (as if there was ever any doubt) that he was not cut out for experimental work. Toward the end of his second year of graduate school, Eero Simoncelli handed Jonathan a book on Bayesian graphical models, and Jonathan decided to tackle the problem of understanding the computational role of feedback projections in the brain from first principles. Meanwhile, Odelia Schwartz (a 4th-year student working on spike-triggered analysis methods) inspired Jonathan to undertake a side project applying some of Odelia's ideas to integrate-and-fire neurons. Jonathan pursued these ideas in collaboration with Liam Paninski, a 1st year student who (to Jonathan's considerable consternation) knew a lot more math and was far more skilled at having good ideas and writing them up into scientific papers. Jonathan has been working on the offshoots of these ideas ever since, though hopes to one day return to the original topic of his thesis, and occasionally still attempts to bust out philosophical insights while his wife and two young children run for cover.