Published August 7, 2026
| Version v2.0.1
Software
Open
TRP fear-goal architecture: validated K=4 reproducibility release
Authors/Creators
- 1. School of Psychology, Fujian Normal University, Fuzhou, China
- 2. School of Humanities and Social Science, The Chinese University of Hong Kong (Shenzhen), Shenzhen, China
Description
Validated open-data and code release supporting the bioRxiv version 2 manuscript, Four recurrent behavioral modes organize fear-goal arbitration across learning and relapse in rats.
The controlling model is a diagonal-covariance K=4 Gaussian hidden Markov model with variance floor 0.01 and selected seed 9. This release contains the exact 47,004-row primary dataset, preprocessing artifacts, serialized model, complete decoded paths, model-selection and restart-stability evidence, downstream analyses, figures, and an executable verifier.
The historical full-covariance fits and K=8-to-four coarse-graining are retained only as provenance and are not part of the supported inferential chain.
Notes
Version 2.0.1 is a semantic and presentation-order correction only. It renames published state ID 1 from Safety/Appetitive to Exploratory, presents reader-facing states in TRP order (Fear/Danger, Quiescent, Exploratory, Goal-directed), regenerates the figures, and makes the Figure 4A legend opaque for readability. The HMM was not refitted. Serialized parameters, preprocessing artifacts, all 47,004 decoded state IDs, and every numerical result are unchanged.
Files
TRP_open_data_code_v2.0.1.zip
Files
(8.8 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:d4ec44e482c98f9b7c94016b3a939592
|
8.8 MB | Preview Download |
Additional details
Related works
- Is derived from
- Dataset: 10.5281/zenodo.20551550 (DOI)
- Is supplement to
- Software: https://github.com/luckbin-lab/trp-fear-goal-architecture (URL)
- Journal article: 10.64898/2026.07.17.739137 (DOI)