This year we organize two summer workshops in Amsterdam – the first focuses on Bayesian cognitive modeling, and the second focuses on Bayesian statistics with JASP. Before I outline the content of these workshops, let me recount how it was that bitter envy drove me to practical Bayesian modeling.
A little over 10 years ago, in a small guest office at UC Irvine, my long-term collaborator Michael Lee and I discussed an interesting data set on patients with dissociative identity disorder. Michael and I quickly decided that we wanted to fit a hierarchical mixture model; but how this should be accomplished was another matter, at least to me. I was first surprised, then impressed, and ultimately envious when Michael was able to fit the models to data virtually at the same speed with which they were proposed. The end result was published only recently:
Lee, M. D., Lodewyckx, T., & Wagenmakers, E.-J. (2015). Three Bayesian analyses of memory deficits in patients with dissociative identity disorder. In J. R. Raaijmakers, A. Criss, R. Goldstone, R. Nosofsky, & M. Steyvers (Eds.), Cognitive Modeling in Perception and Memory: A Festschrift for Richard M. Shiffrin, pp. 189-200. Psychology Press.
At the time Michael had already written a few chapters explaining how the BUGS language could be applied to problems in cognitive modeling. As soon as I returned to the University of Amsterdam my students and I studied the chapters and started to contribute more material. In the end, Michael and I co-authored a course book that provides a relatively comprehensive introduction to Bayesian cognitive modeling:
Lee, M. D., & Wagenmakers, E.-J. (2013). Bayesian cognitive modeling: A practical course. Cambridge University Press.
The bottom line is that the BUGS language (and similar probabilistic programming languages such as JAGS and Stan) allows anybody to implement realistic models with great efficiency. Best of all, it is fun to do.
Workshop 1: Bayesian Cognitive Modeling
The first workshop is on for five days, from August 21-25, 2017. Co-taught with the before-mentioned Michael Lee from UCI, this workshop provides participants with hands-on experience in constructing, applying, and evaluating Bayesian models.
Each morning/afternoon starts with a plenary session on Bayesian theory, followed by an intensive practical. For the practical, we use the course book Bayesian cognitive modeling: A practical course, which will be provided to all participants upon arrival. As the workshop covers a variety of topics within cognitive science and exercises of varying difficulty, the course material is appropriate for researchers with a wide range of prior knowledge and interests. Moreover, the practicals are supported by an expert team of students, postdocs, and professors – this way, participants can get quick individual feedback.
At the end of this workshop, participants are familiar with key Bayesian concepts such as prior and posterior distributions, prior and posterior predictives, MCMC sampling, and Bayes factor hypothesis tests. Participants have completed exercises on mixture models, hierarchical models, and have acquired the skills to apply the techniques to their own problems. Although many of the examples concern cognitive modeling (i.e., modeling performance in tasks that tap processes such as memory, perception, risk-taking, decision-making, and reasoning), the knowledge obtained in this workshop generalizes to a broad range of other phenomena.
This is the seventh installment of this workshop. For more information and the option to register click here
Workshop 2: Theory and Practice of Bayesian Hypothesis Testing: A JASP Workshop
The second workshop is on for two days, from August 28-29, 2017. Co-taught with Richard Morey from Cardiff University, this workshop provides participants with the theory and practice of Bayesian inference using JASP.
In contrast to the cognitive modeling workshop, this workshop focuses on standard Bayesian statistical procedures such as the binomial test, t-test, ANOVA, correlation, and regression. We alternate between plenary lectures and practical exercises. The fact thet JASP program is so easy to use allows us to spend more time explaining key Bayesian concepts. Particular attention will go out to model specification and the interpretation of output: Bayes factors, model-averaging, sequential testing, robustness checks, and more.
Throughtout the workshop, important ideas are illustrated with real data sets. At the end of the workshop, participants are familiar with core Bayesian concepts, and they are able to execute and interpret Bayesian tests in JASP. the course material is appropriate for researchers with a wide range of prior knowledge and interests. This workshop is also supported by an expert team of students, postdocs, and professors, which means that participants can get quick individual feedback.
This is the third installement of this workshop. For more information and the option to register click here.
Like this post?
Subscribe to our newsletter to receive regular updates about JASP including our latest blog posts, JASP articles, example analyses, new features, interviews with team members, and more! You can unsubscribe at any time.