Dynamic models of choice
DOI: 10.3758/s13428-018-1067-y
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper introduces the Dynamic Models of Choice (DMC), an open-source R-based software package designed to facilitate the Bayesian parameter estimation of evidence-accumulation models. The authors address the significant computational and methodological challenges associated with fitting these models, which are widely used in psychology and neuroscience to quantify latent cognitive processes such as evidence quality and response caution. Specifically, the paper focuses on two prominent models: the diffusion decision model (DDM), which assumes continuous stochastic evidence accumulation, and the linear ballistic accumulator (LBA), a deterministic race model that offers greater computational tractability. The motivation for DMC is to provide a flexible, user-friendly framework that implements best practices in cognitive modeling, including rigorous assessment of parameter recovery and model fit. The authors demonstrate the use of DMC through a series of tutorials and case studies. The methodology relies on Bayesian inference, where prior distributions representing pre-data knowledge are updated by observed data to yield posterior distributions. Because these posteriors cannot be derived analytically for complex models, DMC employs Differential-Evolution Markov chain Monte Carlo (DE-MCMC) sampling. This approach uses multiple chains and a crossover step to handle the "sloppiness" (high parameter correlation) inherent in race models, ensuring efficient exploration of the parameter space. The paper details the process of fitting models to single subjects and hierarchical groups, emphasizing the importance of convergence diagnostics, such as the $\hat{R}$ statistic and visual inspection of MCMC chains. Additionally, the authors illustrate the software's flexibility by modeling complex cognitive processes, specifically extending a parametric stop-signal race model to account for attention failures. Key findings presented in the paper include the successful application of DMC to recover true parameter values from simulated data, demonstrating the reliability of the DE-MCMC sampler. The authors show that while convergence can be challenging, techniques such as "migration" steps during the burn-in phase and chain thinning help achieve stable posterior estimates. The paper highlights that Bayesian estimation provides more than point estimates; it quantifies uncertainty through credible intervals and allows for the assessment of how well data update prior beliefs. In the stop-signal paradigm example, DMC effectively modeled mixtures of cognitive processes, illustrating its capacity to handle complex experimental designs beyond simple binary choices. The significance of this work lies in its contribution to making rigorous cognitive modeling accessible and reproducible. By providing a comprehensive toolkit for Bayesian estimation, DMC encourages the adoption of best practices such as parameter-recovery simulations, which are essential for validating model assumptions in specific experimental contexts. Furthermore, the paper advocates for a quantitatively cumulative science, demonstrating how posterior distributions from one study can serve as informative priors for future experiments. This approach enhances the precision of parameter estimates over time and facilitates a deeper understanding of cognitive mechanisms across diverse domains, from basic perception to complex decision-making and clinical conditions.
Key finding
The Dynamic Models of Choice software provides a robust, flexible framework for performing Bayesian parameter estimation on evidence-accumulation models, enabling researchers to accurately fit complex cognitive processes and facilitate cumulative scientific inquiry.
Methodology
simulation_modeling
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via author_sweep_intake on 2026-05-28.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | author_sweep | — | — | 2 | 2026-05-28 |
| archive | success | canonical_url | — | — | 1 | 2026-06-04 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-04 |
| chunk | success | chunk | — | — | 1 | 2026-06-04 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-04 |
| enrich | success | — | — | — | 1 | 2026-05-28 |
| promote | success | — | — | — | 1 | 2026-06-04 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Theoretical Contribution: computational model