We’re really excited to have gotten to interview Lisa Pearl recently! Dr. Pearl is an Associate Professor at the University of California, Irvine, and the director of their Computation of Language Laboratory. She’s published numerous articles on how to use statistical models to check different hypotheses about what kids do to learn language, as well as about natural language processing and textual analysis. We’ve been talking about her work since our very first episode!

We got to ask her about a lot of great topics, including:
– what statistical models can tell us about how kids acquire language
– what’s under our control in our writing, and what we unconsciously show as our write-print
– why computers are so bad at detecting tone and picking out the right meanings of words
– how statistical models and Universal Grammar interact
– a question from one of our viewers about how to approach modeling for second language acquisition

And more! Hope you all enjoy it, and thanks to Dr. Lisa Travis and the Department of Linguistics at McGill University for letting us film there.

This discussion of “write-prints” (like a fingerprint, but for writing) and epistolary novels makes me wonder if Lisa Pearl could be the linguist who could do the study I’ve always wanted to see, which would compare the linguistic characteristics of the three types of text in Rainbow Rowell’s Fangirl: the Gemma T. Leslie Simon Snow text, Cath’s fanfic of the Simon Snow books, and the body tex. Which, for example, would be most similar to Carry On, the not-quite-fanfic of the Simon Snow portions of Fangirl?

ommggg Moti can we make this happen?