A principled Bayesian workflow to study robust individual differences in cognitive science

• Applicant: Professorin Julia M. Haaf, Ph.D.
• Subject Area: General, Cognitive and Mathematical Psychology ; General and Comparative Linguistics, Experimental Linguistics, Typology, Non-European Languages ; Personality Psychology, Clinical and Medical Psychology, Methodology
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 546243727

While cognitive scientists have historically been focused on experimental effects, individual differences have more recently become a target of interest as well. However, it is unclear whether cognitive experiments are suitable instruments to reliably assess individual differences. And if researchers wish to better understand or even improve the reliability of their tasks, they are faced with many obstacles. For example, it is not trivial to translate psychometric models underlying common reliability measures into models that are appropriate for cognitive tasks. Additionally, while it may seem straight-forward to use power analysis to assess how many participants are needed to robustly estimate correlations, it is considerably more complicated to also plan how many observations per participant are needed. With this project, we aim to provide a principled Bayesian workflow to study individual differences across cognitive science. First, we will develop a tool to develop Bayesian statistical models of individual differences. Second, we will design a Bayesian study planning tool that allows us to plan the number of observations and participants ahead of time, but also allows us to continuously adjust the sample sizes during data collection. In addition, we will conduct an empirical study to better understand agreement attraction in antecedent-reflexive constructions, and to apply and validate the developed statistical tools.

• DFG Programme: Priority Programmes
• Subproject of: SPP 2317: META-REP: A Meta-scientific Programme to Analyse and Optimise Replicability in the Behavioural, Social, and Cognitive Sciences
• Co-Investigators: Professorin Audrey Bürki, Ph.D.; Professor Dr. Shravan Vasishth