Spring 2014: Meetings 1-3

The plan for the spring quarter and possibly early summer is to read and work through the following two books:

• Bayesian Cognitive Modeling: A Practical Course (Lee & Wagenmakers 2014): see this pdf for the first two parts of the book and also this website for additional materials (code, answers to exercises, etc.)
• Bayesian Rationality: The Probabilistic Approach to Human Reasoning (Oaksford & Chater 2007); see this overview article for more information

Our ultimate goal is to get started on a cognitive modeling project together, starting with a specific data set and using the computational tools from the first (very practical) book and the general cognitive framework introduced in the second book. Nate has kindly offered to get us started with a data set he collected — more on this later.

The first 3 meetings (May 1, May 8, and May 15) were just introductory: we reviewed the basics of probability theory, the differences between frequentist and Bayesian inference, the installation of R and JAGS, and a discussion of the Bernoulli likelihood function and basic Bayesian updates for coin-flip / Bernoulli distributed data with Beta priors (supported by an R script). The actual Bernoulli-Beta models are very simple, but instantiate very clearly how Bayesian inference works by combining a prior probability distribution over theories / hypotheses and the likelihood of the actual data given those theories / hypotheses, to obtain a posterior probability distribution over theories / hypotheses that encapsulates all the probabilistic inferences and conclusions we need.

For the participants, there is a Dropbox folder that makes available various materials (just like last quarter).

Once the necessary introductory discussion is completed (we’re almost there), we’ll start reading the first 4 chapters from each of the above 2 books and we’ll probably discuss them in a more or less interleaved manner.