Hey!

My name is John R. Starr.¹ I’m a linguist interested in human cognition. Most of my work focuses on 1) the underlying processes and representations that influence how people process linguistic information as they encounter it, and 2) building, comparing, and testing cognitively-informed language models (large or not) on human data collected by both myself and others. I am also the developer of GRIS, a novel and flexible experimental paradigm where participants manipulate objects in space; the tl;dr of GRIS is that you can approximate and compare mental representations of humans for all kinds of things. You can read more about my work by checking my CV, which I update pretty frequently.

I am an incoming Postdoctoral Fellow in the Department of Cognitive Science at Johns Hopkins University, where I will advised by Dr. Jennifer Hu and affiliated with glint. I completed my PhD in the Department of Linguistics at Cornell University, where I was advised by Dr. Marten van Schijndel. At Cornell, I was affiliated with the Computational Psycholinguistics Discussion Group (C.Psyd), Cornell NLP, the Cornell Phonetics Lab (PLab), and the Linguistic Meaning Lab (LiMe). Additionally, I served as a Student Representative on the Executive Committee for the Linguistic Society of America. Before Cornell, I completed two BAs, one in English Writing (Poetry) and one in Linguistics, from the University of Pittsburgh.

In my time off, I like to skip stones in local creeks, lakes, and other waterways; rock climb (particularly bouldering, of which I have significant routesetting experience as the Head Routesetter at Cayuga Climbs); and nap with my cat Magnolia.

Think you’re good at data manipulation? Try these pandas puzzles..

I’m originally from the suburbs of Philadelphia, Pennsylvania, meaning \(John \in \{ \,x\, \mid \, x\, is\, a\, jaded\, and\, delusional\, Philly\, sports\, fan \} \).