In the workshop we will discuss why it is beneficial to work with computational process models (session 1), how we can plan optimal studies and compare competitor models based on multiple process and outcome measures (session 2), and how this can be accomplished using the statistical software R (session 3).
Session 1: I will reflect on criteria for evaluating models before data collection (Jekel, in press; Glöckner & Betsch, 2011). We will also discuss why process models are especially helpful for comparing competitor models: Whereas competitor models often predict similar outcomes for (e.g.) decisions between options in a probabilistic inference task, they may differ fundamentally in their assumptions about the process that leads to a decision. I will use the Parallel-constraint Satisfaction Model of Decision Making (Glöckner, Hilbig, & Jekel, 2014) and iCodes (Jekel, Glöckner, & Bröder, 2018) for illustration.
Session 2: I will introduce you to the Euclidian Diagnostic Task Selection (EDTS; Jekel, Fiedler, & Glöckner, 2011). EDTS allows researches to find tasks that optimally differentiate between competitor models by taking process measures into account. I will also introduce you to the Multiple-Measure Maximum Likelihood strategy classification method (MM-ML; Jekel, Nicklisch, & Glöckner, 2010; Glöckner & Jekel, in press). MM-ML allows researchers to integrate outcome and process measures in a single measure of model fit. Extensions, recent developments, and open challenges in testing process models will be briefly discussed.
Session 3: We will apply EDTS and MM-ML in R using data from our lab.
Glöckner, A., & Betsch, T. (2011) The empirical content of theories in judgment and decision making: Shortcomings and remedies. Judgment and Decision Making, 6, 711–721.
Glöckner, A., Hilbig, B. E., & Jekel, M. (2014). What is adaptive about adaptive decision making? A parallel constraint satisfaction account. Cognition, 133, 641-666.
Glöckner, A., & Jekel, M. (in press). Testing cognitive models by a joint analysis of multiple dependent measures. In M. Schulte-Mecklenbeck et al. (Eds.), Handbook of Process Tracing.
Jekel, M. (in press). Empirical content as a criterion for evaluating models. Cognitive Processing.
Jekel, M., Fiedler, S., & Glöckner, A. (2011). Diagnostic task selection for strategy classification in judgment and decision making: Theory, validation, and implementation in R. Judgment and Decision Making, 6, 782–799.
Jekel, M., Glöckner, A., & Bröder, A. (2018). A new and unique prediction for cue-search in a parallel-constraint satisfaction network model: The attraction search effect. Psychological Review, 125, 744–768.
Jekel, M., Nicklisch, A., & Glöckner, A. (2010). Implementation of the Multiple-Measure Maximum Likelihood strategy classification method in R: Addendum to Glöckner (2009) and practical guide for application. Judgment and Decision Making, 5, 54–63.