Winner Takes All! What Are Race Models, and Why and How Should Psychologists Use Them?
DOI: 10.1177/09637214221095852
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This review article by Heathcote and Matzke (2022) addresses the role of race models in psychological science, specifically focusing on how these quantitative frameworks allow researchers to infer unobservable cognitive processes from observed response choices and response times (RTs). The authors argue that race models are essential for rigorously testing psychological theories, as they provide a mechanism to disentangle decision-making processes from non-decision factors like encoding and motor execution. The paper aims to showcase the state of the art in race modeling, explaining their theoretical foundations, variations, and practical applications in understanding complex cognitive tasks. The authors define race models by three core characteristics: they contain one or more "runners" representing cognitive processes that take time to complete; these runners may interact or remain independent; and a "winner-takes-all" rule determines the outcome based on which runner finishes first. The review details several specific evidence-accumulation models (EAMs), a prominent subclass of race models. These include the Linear Ballistic Accumulator (LBA), where accumulation is deterministic with trial-to-trial variability; the Racing Diffusion Model (RDM), which features diffusive, moment-to-moment variability; and the Wiener Diffusion Model (WDM), a special case where accumulators are negatively correlated, effectively reducing the race to a single accumulator with two thresholds. The authors illustrate how these basic models serve as building blocks for more complex architectures, such as the Prospective Memory Decision-Control (PMDC) model, the Timed Racing Diffusion Model (TRDM), and the Advantage LBA (ALBA), which handle tasks involving prospective memory, time perception, and multi-alternative choices, respectively. The paper highlights several key findings regarding the utility and advantages of race models. First, they prevent mistaken inferences by accounting for speed-accuracy trade-offs; for instance, the authors cite evidence that performance correlations previously attributed to cognitive ability were actually driven by heritable differences in response caution. Second, race models enable the separation of decision-making effects from non-decision phenomena, such as task-switching costs or attentional selection, by modeling these as distinct stages. Third, incorporating RT data provides greater measurement precision than choice-only models, effectively doubling estimation precision in some contexts. The authors also demonstrate that race models can unify theoretical constructs, such as linking signal detection theory parameters with RT-informed estimates. In terms of methodology, the authors emphasize the importance of hierarchical Bayesian estimation to ensure parameter identifiability, particularly when data per participant is limited. They argue that fitting models to simulated data allows researchers to assess estimation accuracy and precision. The significance of this work lies in its demonstration that race models are not merely descriptive tools but powerful instruments for uncovering the mechanistic causes of behavior. By providing a rigorous framework for modeling both simple and complex cognitive tasks, race models facilitate deeper insights into psychological processes ranging from inhibitory control and working memory to clinical conditions like schizophrenia and ADHD, thereby advancing the precision and theoretical clarity of psychological science.
Key finding
Race models provide a rigorous quantitative framework for inferring unobservable cognitive processes by jointly modeling response choices and response times.
Methodology
review
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 | — | — | 5 | 2026-06-06 |
| 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.