cogsci phd student at brown university
nsf graduate research fellow

My name is Laila Johnston (she/her). I am a Cognitive Science PhD Student at Brown University advised by Roman Feiman and Ellie Pavlick. I received my B.S. in Mathematics with minors in Computer Science and Philosophy from the University of Central Florida.

My research focuses on the mechanisms of how humans use ad-hoc contextual information to determine what is revelant (i.e. the frame problem or the relevance problem). For any task, the search space is practically infinite. An exhaustive search would be computationally explosive and take too much time. For example, when looking for a cup in a kitchen we do not first consider the light on the faucet and then decide it’s irrelevant to finding a cup. How does the cognitive system know what is relevant without first computing and discarding all the irrelevant information? Currently, I am investiagting this problem in communication and language processing (next word predictions, focus alternatives generation, and question asking [see my NSF GRFP research proposal for more details on question asking]).

Before starting my PhD, I worked in the CoCoSci Group at MIT with Max Siegel, Josh Tenenbaum, and Tobi Gerstenberg researching conceptual reasoning within probabilistic programs and large language models, in the domain of tug-of-war games. Before that, I worked with David Danks researching individual differences in causal learning and building models of causal inference learning (see our cogsci paper!).

On my free time, I enjoy playing beach tennis (I won 2 medals at the IFBT 2025 World Championships!), dancing salsa, bachata, and merengue, hiking/backpacking, film photography, and going to jazz concerts. I also enjoy spending too much money on vinyl (see my collection) and buying more books than I could ever read.