Mismatch-Gated Recalibration and the Time Course of Release Hallucinations in a Deafferented Cortical Model
A paper from the School of Settled Matters and the School of Underlying Conditions builds a two-area cortical rate model with homeostatic gain and finds that, with the set point fixed, unsupported activity after input loss never clears. When the set point may drift only against an unbroken record of mismatch, residual input that occasionally confirms the old expectation stalls recalibration, so across 920 simulated runs clearance takes longest at 80% loss (median 112 days) and is quicker at complete loss (47 days).
| Authors | Clemency Tredinnick , Brynjar Obiora |
|---|---|
| School | School of Settled Matters |
| Output type | Research paper |
| Published | |
| DOI | 10.5555/slop.r1mf7y |
| Pages | 5 |
| Version | 1.0 |
| Licence | CC BY 4.0 |
Cites
Cite as
@misc{slop_r1mf7y,
author = {Clemency Tredinnick and Brynjar Obiora},
title = {The Set Point Outlives the Input: Mismatch-Gated Recalibration and the Time Course of Release Hallucinations in a Deafferented Cortical Model},
year = {2026},
publisher = {Slop University},
doi = {10.5555/slop.r1mf7y},
url = {https://slop.university/outputs/slop-paper-a-computational-model-r1mf7y/},
version = {1.0},
note = {Research paper},
}