Training Transfers the Limits on Perception from Parietal to Ventral Cortex
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Visually guided behavior depends on (1) extracting and (2) discriminating signals from complex retinal inputs, and these perceptual skills improve with practice [1Dosher B.A. Lu Z.L. Perceptual learning reflects external noise filtering and internal noise reduction through channel reweighting.Proc. Natl. Acad. Sci. USA. 1998; 95: 13988-13993Crossref PubMed Scopus (384) Google Scholar]. For instance, training on aerial reconnaissance facilitated World War II Allied military operations [2Downing T. Spies in the Sky. Hachette UK, London2011Google Scholar]; analysts pored over stereoscopic photographs, becoming expert at (1) segmenting pictures into meaningful items to break camouflage from (noisy) backgrounds, and (2) discriminating fine details to distinguish V-weapons from innocuous pylons. Training is understood to optimize neural circuits that process scene features (e.g., orientation) for particular purposes (e.g., judging position) [3Gilbert C.D. Sigman M. Crist R.E. The neural basis of perceptual learning.Neuron. 2001; 31: 681-697Abstract Full Text Full Text PDF PubMed Scopus (532) Google Scholar, 4Poggio T. Fahle M. Edelman S. Fast perceptual learning in visual hyperacuity.Science. 1992; 256: 1018-1021Crossref PubMed Scopus (453) Google Scholar, 5Karni A. Sagi D. Where practice makes perfect in texture discrimination: evidence for primary visual cortex plasticity.Proc. Natl. Acad. Sci. USA. 1991; 88: 4966-4970Crossref PubMed Scopus (871) Google Scholar, 6Schoups A.A. Vogels R. Orban G.A. Human perceptual learning in identifying the oblique orientation: retinotopy, orientation specificity and monocularity.J. Physiol. 1995; 483: 797-810PubMed Google Scholar]. Yet learning is most beneficial when it generalizes to other settings [7Dosher B.A. Lu Z.L. Perceptual learning in clear displays optimizes perceptual expertise: learning the limiting process.Proc. Natl. Acad. Sci. USA. 2005; 102: 5286-5290Crossref PubMed Scopus (71) Google Scholar, 8Xiao L.-Q. Zhang J.-Y. Wang R. Klein S.A. Levi D.M. Yu C. Complete transfer of perceptual learning across retinal locations enabled by double training.Curr. Biol. 2008; 18: 1922-1926Abstract Full Text Full Text PDF PubMed Scopus (295) Google Scholar] and is critical in recovery after adversity [9Ding J. Levi D.M. Recovery of stereopsis through perceptual learning in human adults with abnormal binocular vision.Proc. Natl. Acad. Sci. USA. 2011; 108: E733-E741Crossref PubMed Scopus (103) Google Scholar], challenging understanding of the circuitry involved. Here we used repetitive transcranial magnetic stimulation (rTMS) to infer the functional organization supporting learning generalization in the human brain. First, we show dissociable contributions of the posterior parietal cortex (PPC) versus lateral occipital (LO) circuits: extracting targets from noise is disrupted by PPC stimulation, in contrast to judging feature differences, which is affected by LO rTMS. Then, we demonstrate that training causes striking changes in this circuit: after feature training, identifying a target in noise is not disrupted by PPC stimulation but instead by LO stimulation. This indicates that training shifts the limits on perception from parietal to ventral brain regions and identifies a critical neural circuit for visual learning. We suggest that generalization is implemented by supplanting dynamic processing conducted in the PPC with specific feature templates stored in the ventral cortex.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it