Conference Events

Prior to the conference, on the 6th of February, we will be hosting two workshops. These workshops are excellent opportunities to expand your methodological toolkit. We will also be hosting a pre-conference welcome event in the evening.

Please click here register your attendance for the workshops and/or welcome event.

Both workshops will be held in the EZONE building, Room 110. The welcome will be held at 5-7PM in the Science Common Room, Room 405 (see location).

The morning workshop (9AM-12PM) is by Professor Michel Regenwetter:

Heterogeneity of Behavior: Order-Constrained Modeling and Data Analytics

Synopsis: Conceptual, mathematical, and statistical framework to model heterogeneity of behavior and better understand the scope of psychological theory.  Workshop participants will learn to move beyond a psychology of averages and think of variability of behavior as a source of information for scientific inquiry rather than mere noise. This workshop provides a basic introduction to order-constrained models and associated statistical inference methods, including frequentist and Bayesian approaches. No advanced mathematical modeling or quantitative analytics skills are required. The workshop aims to speak to a broad audience with a broad range of scientific interests. If you are willing to think deeply about variability and heterogeneity, this workshop should have something to offer.

Requirements: You may wish to bring a laptop to follow along with demonstrations on an online app.

The afternoon workshop (2PM-5PM)  is by Professor Andrew Heathcote:

Extended Models of Choice in R

Synopsis: This workshop will provide an overview of the EMC2 package, which supports Bayesian hierarchical fitting of a range of cognitive models of choice. The workshop will focus on four basic Evidence Accumulation models, the Diffusion Decision Model (DDM), the Linear Ballistic Accumulator (LBA), the Racing Diffusion Model (RDM) and the Lognormal Race (LNR). It will demonstrate how to estimate these models using R's linear model language to specify the mapping of design factors to each type of model parameter, with fits to both individuals and to groups of participants with a population model allowing for correlations among model parameters. Methods of model checking, posterior inference and model comparison will be demonstrated, with the latter using DIC, BPIC, and marginal likelihoods (i.e., Bayes Factors) estimated by the IS2 and bridge sampling methods.

Requirements: Before the workshop we recommend you read this paper: https://osf.io/preprints/psyarxiv/2e4dq

As described in the paper you can download the EMC2 package from https://github.com/ampl-psych/EMC2 (see the README file there for installation instructions). Please make sure EMC2 and all of its dependencies are installed on your laptop before the workshop.

We recommend EMC2 be used with RStudio (https://posit.co/download/rstudio-desktop/). The script and accompanying results for the analysis described in the paper can be downloaded from https://osf.io/7euya/ 

Copy all files into a directory, double click on the EMC_paper.Rproj file (this requires RStudio), and open the EMC_paper.R script. The workshop will work through the analyses in this script. The directory also contains RData files with completed analyses so they do not have to be run during the workshop (as they can be time consuming) but you are welcome to also run your own versions following the instructions in the script.

As the workshop is brief it will assume some background knowledge of 1) Evidence Accumulation Models, 2) Bayesian analysis and 3) Hierarchical models. Recommended readings:

https://journals.sagepub.com/doi/pdf/10.1177/09637214221095852

https://link.springer.com/article/10.3758/s13423-017-1343-3

https://sites.socsci.uci.edu/~mdlee/Lee2010.pdf