Bayesian Data Analysis and Cognitive Modeling

Objectives

In the course, students will learn to understand the key Bayesian concepts of data analysis, such as the updating of a prior into a posterior distribution of beliefs, Bayes factors, and how Bayesian statistics helps to overcome limitations of the classical frequentists approach to data analysis. The course also provides an introduction to cognitive modeling, including the development of a model and how to interpret the results of Bayesian parameter estimation with JAGS. 

Students will acquire the skill to implement common statistical tests (t-tests, ANOVA, correlation, regression) with the analysis program JASP (that can implement both frequentist and Bayesian implementations of these tests), learn how to interpret the results of Bayesian data analyses, to write code for Bayesian data analysis in JAGS, to implement cognitive models (e.g., memory models, prospect theory) in JAGS, and calling JAGS from R.

Description

The advantages of a Bayesian approach to data analysis have been known for a long time (e.g., Edwards, Lindman, & Savage, 1963). Recent developments in computer science have made the practical application of these approaches accessible. Bayesian methods avoid many of the problems of frequentist methods—such as not being able to confirm the null hypothesis or that p-values depend on the goals during participant recruitment (e.g., Wagenmakers et al., 2017; Kruschke, 2014). In addition, Bayesian statistics is more intuitive than frequentist statistics, in that it evaluates a hypothesis given the data rather than the other way round. Bayesian approaches are therefore likely to eventually replace frequentist approaches in data analysis. The goal of this course is to introduce students to the logic and practice of Bayesian statistics as well as to provide an introduction to cognitive modeling

Teaching and learning methods

The course will involve presentations from the instructor as well as many practical elements in which students get hands-on experience with the JASP, JAGS, and R. In addition, there will be group exercises. All students should bring a laptop with preinstalled free software (more information on the software that needs to be installed beforehand will be provided two weeks before the beginning of the course).

Recommended Literature

Dienes, Z. (2008). Bayes and the probability of hypotheses. In Dienes, Z., Understanding psychology as a science: An introduction to scientific and statistical inference (p. 82-120). Palgrave Macmillan.

Kruschke, J. K. & Liddell, T. M. (2018). Bayesian data analysis for newcomers. Psychonomic Bulletin & Review, 25, 155-177.

Wagenmakers, E.-J. et al. (2018). Bayesian inference for psychology. Part I: Theoretical advantages and practical ramifications. Psychonomic Bulletin & Review, 25, 35-57.

Shiffrin, R. M., Lee, M. D., Wagenmakers, E.-J., & Kim, W. J. (2008). A survey of model evaluation approaches with a focus on hierarchical Bayesian methods. Cognitive Science, 32, 1248-1284.