A hybrid approach to dynamic cognitive psychometrics
DOI: 10.3758/s13428-023-02295-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 addresses the challenge of maintaining robust measurement properties in dynamic cognitive psychometrics, specifically within the stop-signal paradigm used to assess response inhibition. While process-based models grounded in experimental psychology and neuroscience offer valid interpretations of mental capacities, they often suffer from poor estimation performance. The authors identify a specific statistical issue: estimating a parameter-dependent lower bound (non-decision time) for reaction time distributions violates regularity conditions required for consistent likelihood-based estimation. This "shift parameter" problem is particularly severe in stop-signal tasks because the key index of inhibition, stop-signal reaction time (SSRT), is latent and indirectly observed. Previous attempts to model both the choice (go) and inhibition (stop) processes using psychologically realistic evidence-accumulation models resulted in inconsistent parameter estimates and poor recovery. To resolve this, the authors propose a "hybrid" modeling approach called the racing-diffusion ex-Gaussian (RDEX) model. This method blends process-based and descriptive components. For the choice process, the model retains a psychologically realistic independent racing-diffusion architecture, where evidence accumulates to a threshold, allowing for meaningful interpretation of parameters related to attention and decision thresholds. For the stop process, however, the model replaces the complex accumulation process with a descriptive ex-Gaussian distribution. This descriptive component provides a flexible statistical fit for the SSRT distribution without requiring the estimation of problematic shift parameters associated with evidence accumulation. The model also incorporates a dynamic multinomial processing tree structure to account for "trigger failures," representing instances where participants fail to initiate the go or stop processes due to attention lapses. The authors validate the RDEX model through empirical application and simulation studies. They apply the model to data from a complex experimental design involving manipulations that selectively influence specific psychological parameters of the choice process. The results demonstrate that the hybrid model accurately captures these selective influences, confirming that the process-based components correctly reflect underlying psychological mechanisms. Furthermore, parameter-recovery simulations show that hierarchical Bayesian estimation of the RDEX model performs effectively even with a modest number of test trials, a significant advantage over purely process-based models which require impractically large datasets to achieve similar accuracy. The significance of this work lies in providing a practical solution to the trade-off between psychological validity and statistical measurability in cognitive modeling. By demonstrating that a hybrid approach can simultaneously yield accurate estimates of latent inhibitory capacity (SSRT) and interpretable parameters of choice behavior, the authors expand the applicability of dynamic cognitive psychometrics. This methodology allows researchers to apply sophisticated process models to complex tasks in clinical, developmental, and neuroscience contexts without succumbing to the estimation inconsistencies that have previously limited the use of fully process-based stop-signal models.
Key finding
A hybrid modeling approach that blends process-based components for choice behavior with descriptive components for the stop process achieves accurate parameter estimation in the stop-signal paradigm, overcoming the measurement issues associated with purely process-based models.
Methodology
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