An integrated model of choices and response times in absolute identification.
DOI: 10.1037/0033-295x.115.2.396
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 paradox of absolute identification, where participants struggle to accurately identify stimuli from a small set despite being able to make accurate comparative judgments. The authors developed SAMBA (Selective Attention, Mapping, and Ballistic Accumulation), an integrated model designed to account for both choice accuracy and response time (RT) distributions, including sequential effects like assimilation and contrast. Previous models typically focused on either relative processes (short-term context) or absolute processes (long-term context) and often failed to simultaneously explain choices, RTs, and sequential dependencies. SAMBA aims to provide a unified, constrained account of these phenomena by integrating three core components: selective attention for stimulus representation, mapping for transforming estimates into response strengths, and ballistic accumulation for decision making. The model operates in three stages. First, the selective attention stage uses a spatial psychophysical representation where stimulus magnitude is estimated by summing activity between the selected stimulus location and the upper and lower anchors of the stimulus range. This process generates a magnitude estimate that varies trial-to-trial, naturally producing bow effects in accuracy. Second, the mapping stage transforms this magnitude estimate into $N$ response strengths using long-term memory of average magnitude estimates for each stimulus, avoiding the need for arbitrary parameter estimation for each response alternative. Third, the ballistic accumulator stage assigns these response strengths to separate accumulators that race to a threshold; the first to reach the threshold determines the choice and RT. The model incorporates short-term memory effects by allowing the previous stimulus's estimated magnitude to influence the current trial's anchors, thereby accounting for sequential effects. The authors validated SAMBA by fitting it to benchmark data sets, including Lacouture’s (1997) comprehensive choice and RT data, and Kent and Lamberts’s (2005) and Lacouture and Marley’s (2004) individual subject data. SAMBA successfully accounted for global effects such as stimulus range and set size limitations (asymptoting at 2–3 bits of information), as well as local sequential effects like assimilation (errors toward the previous stimulus) and contrast (errors away from stimuli two trials prior). The model required only a few parameters, which remained consistent across diverse data sets. Unlike previous models that treated RTs as an afterthought or failed to explain sequential effects, SAMBA provided a quantitative account of both choice phenomena and RT distributions, including distribution shapes for correct and error responses. The significance of this work lies in providing the first comprehensive model of absolute identification that integrates choice and RT data while explaining sequential effects through a mechanistic architecture. By deriving predictions from its structure rather than arbitrary parameter adjustments, SAMBA offers a more theoretically satisfying explanation of how magnitude estimates arise and how they influence decision processes. This approach resolves limitations of prior relative and absolute models, demonstrating that a unified framework can capture the complex interplay between short-term memory, long-term referents, and decision dynamics in absolute identification tasks.
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 | — | — | 23 | 2026-06-09 |
| extract | success | cached | — | — | 2 | 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 | 1 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | success | — | — | — | 1 | 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.