PPE Speaker Series: “Ideological Bayesians” with Kevin Dorst (MIT)

Abstract:
It’s often said that Bayesian updating is unbiased and converges to the truth—and, therefore, that biases must emerge from non-Bayesian sources. That’s wrong. For agents like us, updating on our total evidence is impossible—instead, we must selectively attend to certain questions, ignoring others. Yet correlations between what we see and what questions we ask—“ideological” Bayesian updating—can lead to predictable biases and polarization. I explore what this model might teach us about (the lack of) deference and (the ubiquity of) disagreement in politics.

Join us for this free, open lecture with Q&A and pizza to follow!

Kevin Dorst is an Associate Professor in the Department of Linguistics and Philosophy at MIT. He works at the border between philosophy and social science, focusing on rationality.

How rational are people in general? How about your political opponents? How about yourself? Many popular and social-scientific answers to these questions are unflattering. Yet most people think and act as if they are exceptions.

​Kevin’s work tries to square this circle. The case for irrationalism is built on the empirical fact that people predictably deviate from classical theories of rationality. His work is built on the normative fact that such theories are elegant and useful—but wrong. Dorst develops and deploys better models of rationality to help refine our interpretation of the empirical results, and square them with our own experiences of (ir)rationality. 

Short story: people are more rational than you think—but rationality buys you less than you thought. For a summary of the project and its connection to polarization in politics, see my blog post: “A Plea for Political Empathy“. For an extended argument that polarization has rational causes, see the series Reasonably Polarized: Why politics is more rational than you think.  For the latest, see the blog’s feed (now on Substack)—sign up for posts every few weeks.