Modeling the dynamics of recognition memory testing with an integrated model of retrieval and decision making
DOI: 10.1016/j.cogpsych.2018.04.002
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 "test position effect" (TPE), a robust phenomenon in recognition memory where performance declines monotonically across test trials. Despite its prevalence, the underlying mechanism remains debated, with three primary hypotheses proposed: (1) interference from learning test items (item noise), (2) a shift in the context representation used to cue memory (context drift), and (3) changes in participants' speed-accuracy thresholds. Previous models addressed either retrieval or decision-making processes but not both, limiting their ability to decompose performance changes into specific cognitive components. To resolve this, the authors developed an integrated model combining the Osth and Dennis (2015) global matching model for memory retrieval with the Diffusion Decision Model (DDM) for decision making. The retrieval component generates memory strength estimates based on item-context bindings, accounting for item noise, context noise, and context drift. These strength estimates determine the drift rate distributions in the DDM, which models choice probabilities and response times. The model allows for parametric estimation of item noise, context drift, and dynamic changes in decision boundaries and bias across trials. The model was applied to four datasets representing three specific empirical challenges: the relative impact of tested versus studied items, the differential decline of strong versus weak items in pure versus mixed lists, and the effect of interspersed lexical decision trials on performance decline. Parameter estimation revealed that item interference plays a weak role in explaining the TPE. In contrast, context drift—where the context representation changes as a direct consequence of retrieval—played a very large role. The model successfully captured the finding that the number of test items influences performance more than the number of studied items, and that performance declines less for strong items in pure lists but not in mixed lists. Furthermore, the model accounted for the lack of increased decline rates when lexical decisions were interspersed with recognition trials. These findings suggest that the decline in recognition memory performance during testing is primarily driven by changes in the episodic context rather than interference from newly learned items. This aligns with prior work indicating a minimal role for item noise in recognition memory and supports the broader episodic memory literature showing that retrieval causes significant context change. By integrating retrieval and decision processes, the study provides a unified account of how both memory strength and decision dynamics contribute to the test position effect, offering a more precise theoretical framework for understanding recognition memory dynamics.
Key finding
Context drift during the testing phase is the dominant mechanism causing the decline in recognition memory performance, while item interference from learned test items plays a negligible role.
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 | 2026-06-06 |
| 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 | success | semantic_scholar | — | — | 4 | 2026-06-15 |
| 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.