Drawing conclusions from choice response time models: A tutorial using the linear ballistic accumulator
DOI: 10.1016/j.jmp.2010.10.001
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This tutorial addresses the challenge of interpreting cognitive models of choice and response time (RT), specifically bridging the gap between parameter estimation and drawing psychologically meaningful conclusions. While evidence accumulation models, such as the linear ballistic accumulator (LBA), offer deeper insights into decision processes than traditional accuracy and mean RT analyses, their application has historically been limited to model developers. The authors aim to provide accessible guidance for researchers on selecting appropriate parametric characterizations for datasets and evaluating model-data agreement. The paper uses the LBA model as a primary example, which assumes that evidence accumulates linearly for competing choices until a threshold is reached. Key parameters include drift rate (evidence accumulation speed), threshold and starting point variability (response caution and bias), and non-decision time (sensory/motor processes). The tutorial demonstrates the modeling process using data from Forstmann et al. (2008), where participants performed a perceptual motion discrimination task under speed, neutral, or accuracy emphasis conditions. The authors outline a step-by-step approach to model selection, beginning with *a priori* constraints based on theoretical plausibility. For instance, they argue that non-decision time and between-trial variability in drift rates should typically remain constant across conditions, while drift rates and caution parameters may vary. This reduces the potential parameter space from 60 to 26 free parameters. To determine which parameters actually need to vary to fit the data, the authors compare two methods: fitting the most complex model to individual participants and examining average parameter estimates, and using sequential model building with information criteria like the Bayesian Information Criterion (BIC). Analysis of the Forstmann et al. data revealed that response caution parameters (threshold and starting point) varied significantly across emphasis conditions, consistent with the manipulation of speed-accuracy trade-offs. In contrast, non-decision time and drift rates showed minimal variation across these conditions, though drift rates differed between stimulus types. The authors emphasize that relying solely on average parameter estimates can be misleading due to individual differences, advocating for careful model comparison techniques. The significance of this work lies in democratizing the use of choice RT models by providing clear conventions and assumptions often omitted in published applications. By detailing how to constrain parameters and evaluate model fits, the tutorial enables researchers to infer which specific cognitive processes—such as evidence accumulation rate versus response caution—are influenced by experimental manipulations. This facilitates more precise theoretical interpretations of decision-making data, moving beyond descriptive statistics to mechanistic explanations of behavior.
Key finding
Re-analysis of perceptual decision data using the linear ballistic accumulator model demonstrates that experimental manipulations of response caution primarily affect evidence threshold and starting point parameters, while non-decision time remains largely invariant across conditions.
Methodology
modeling
Sample size: 20
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 | — | — | 11 | 2026-06-06 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-07 |
| chunk | success | chunk | — | — | 1 | 2026-06-07 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-07 |
| enrich | failed | — | — | — | 4 | 2026-07-02 |
| 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