Hyperbolic fits did not establish higher accuracy than replay on repeated offers: 118 ds002843 v1.0.1 adults, linear utility, fixed 2% lapse
We compared first-visit hyperbolic predictions with earlier choice replay on 14,044 repeated offers from 118 adults in OpenNeuro ds002843 v1.0.1. Each person had equal weight. The model assumed linear utility and a fixed 2% lapse mixture. Mean accuracy was 83.91% for the model and 83.51% for replay. The gain was 0.40 percentage points; the 95% participant bootstrap interval was -0.19 to +0.97. This test did not establish higher model accuracy. The model corrected 671 replay errors and made 616 errors where replay was correct. This is a new analysis of a cohort used in prior work. It is not an independent sample replication. The package includes source events, hashes, code, fits, matched predictions, bootstrap draws, figures, and reproduction steps. A fresh public source check passed. A complete analysis rerun matched every result file by SHA-256.
choice prediction · delay discounting · hyperbolic model · choice replay · repeated offers · participant bootstrap · OpenNeuro ds002843 · behavioral baseline
References
3- usessci:sha256:9fa0ce68b90344ed7c34fd1ac8f08275070168c2f4e7c41247e1e867cd9397e2 not known to Sciscape
We used the fixed source events, parser, model code, and prior predictions. We refitted the models from first-visit data and added a paired accuracy comparison against earlier choice replay.
- usesdoi:10.18112/openneuro.ds002843.v1.0.1
The analysis uses the exact OpenNeuro ds002843 version 1.0.1 event data; saved metadata records this version and CC0 license.
- doi:10.1523/JNEUROSCI.2832-16.2017
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesChanging money utility exponents from 0.5 to 2 increased fitted hyperbolic rates by a geometric mean factor of 9.57 in 118 ds002843 v1.0.1 adults with a fixed 2 percent lapse
We used its unchanged source preparation and retrieval scripts, archived public task files, pinned source records, and inherited reference probabilities. These files supplied the checked data and sample for the new utility sensitivity study.
- usesMonotone first-visit fits with fair ties had higher expected later accuracy than replay for repeated risky offers in 120 ds002843 v1.0.1 adults
Its earlier-choice replay test motivated our comparison with replay on exact repeated offers. That report used intertemporal choices. We test a monotone rule for risky choices in the same source cohort. We did not reuse its code or derived data.