dockerHDDM: A User-Friendly Environment for Bayesian Hierarchical Drift-Diffusion Modeling
Wanke Pan et al.
What the paper says
Drift-diffusion models (DDMs) are pivotal in understanding evidence-accumulation processes during decision-making across psychology, behavioral economics, neuroscience, and psychiatry. Hierarchical DDMs (HDDMs), a Python library for hierarchical Bayesian estimation of DDMs, has been widely used among researchers, including researchers with limited coding proficiency, in fitting DDMs and other sequential sampling models. However, issues of compatibility in installation and lack of support for more recent Bayesian-modeling functionalities pose serious challenges for new users, limiting broader adaptation and reproducibility of HDDMs. To address these issues, we created dockerHDDM, a user-friendly computational environment for HDDMs with new features. dockerHDDM brings three improvements: (a) easy to install once docker is installed, ensuring reproducibility and saving time for researchers; (b) compatible with machines with Apple chips; (c) seamless integration with ArviZ, a state-of-the-art Bayesian-modeling library. This tutorial serves as a practical, hands-on guide for researchers to leverage dockerHDDM’s capabilities in conducting efficient Bayesian hierarchical analysis of DDMs. The notebook presented here and in the docker image will enable researchers with various programming levels to model their data with HDDMs.
23 citations
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.72 × 0.4 = 0.29 |
| M · momentum | 1.00 × 0.15 = 0.15 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.