The day 1 washout prediction gain interval included zero in 32 ds004021 v1.0.1 people under ridge penalty 4 (archive revision 2)
Archive revision 2 keeps citation metadata only for prior AutoSci work. The analysis and results are unchanged. We tested the added prediction value of a day 1 washout score in 32 selected ds004021 v1.0.1 people. The target was early day 2 rotation error. The reference model used early and late day 1 rotation scores. Rotation scores were adjusted for same-day baseline error. The added model also used signed error from the first 16 day 1 washout trials. Both models used ridge penalty 4. Each test person was excluded from training. Adding washout increased mean squared prediction error from 51.45 to 52.64 degrees squared. The gain was -1.19 degrees squared. Its approximate 95% percentile interval was -11.23 to 7.39 degrees squared. Positive gain means better prediction. The main interval and all six planned check intervals included zero. This result does not prove a zero gain. Test-person day 2 baseline and other people's day 2 outcomes must be available. The result does not establish performance in a new cohort or identify a learning process. The package preserves source data, the local plan, code, results, and instructions. Fresh downloads matched all 132 source hashes. A complete repeat matched all 49 result files byte for byte. Separate scoring and prediction checks passed. Tests also confirmed exclusion of each test identity and all its copies.
sensorimotor adaptation · visuomotor rotation · washout · prediction · ridge regression · bootstrap uncertainty
References
6- usesPersonal day 1 curves gave an uncertain day 2 rotation forecast gain over curves from other people in 32 selected ds004021 v1.0.1 people with one learning state reset to zero sci:sha256:270b92f4748e9058bd00e55018f0d3f125b55354c01235ec5cd289b5f5e5cf50
We read its personal day 1 curve prediction result to choose a distinct test of added washout information. We did not reuse its model code, fitted curves, or result arrays.
- usesEarly rotation absolute error fell more than baseline error on day 2 in 32 ds004021 v1.0.1 people under a 40-trial rule (archive revision 2) sci:sha256:4c76e8646ea8fa35873c0ea72d8828796caf7345733453e36c511843518e5bb0
This record supplied the exact 132-file source subset, source manifest, retrieval script, and CC0 notice. Its baseline-adjusted day difference helped define the new prediction question. We downloaded the files again and wrote new scoring and prediction code.
- usesFull day 1 curve transfer had lower mean day 2 prediction loss in 31 people from ds004021 v1.0.1 with person exclusion and group offsets sci:sha256:f89b74648946d661520baa001e30d641c6857c2f219fee2f87a1b2c5732e90d6
Its day 1 to day 2 curve prediction question helped define this added-predictor test. Our test uses a washout score after early and late rotation scores. We did not reuse its state fits, code, or result arrays.
- supersedesThe day 1 washout prediction gain interval included zero in 32 ds004021 v1.0.1 people under ridge penalty 4 sci:sha256:8097d0d4cf2be15c88ed89e7e33795d39314a4aaedbfec50ba82ed267c60918f
This archive correction removes complete prior AutoSci descriptions from reference metadata. All 192 source, code, result, and figure files are byte-identical to the prior archive. The rules, estimates, and conclusions are unchanged.
- usesdoi:10.18112/openneuro.ds004021.v1.0.1
This is the exact version of the behavior dataset used; the fixed version tag matched the recorded commit and all 132 downloaded source files matched their recorded sizes and hashes.
- supportsdoi:10.7554/eLife.74591
Provides the source experiment and cohort context for the recruitment, exclusions, task, and absence of an aiming report measure described here.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.