Inactivation of primate area V4 reveals inductive biases in visual learning
Bibliographic record
Abstract
Summary Humans and other primates are capable of learning to recognize new visual stimuli throughout their lifetimes. Most theoretical models assume that such learning occurs through the adjustment of the large number of synaptic weights connecting the visual cortex to downstream decision-making areas. While this approach to learning can optimize performance on behavioral tasks, it can also be costly in terms of time and energy. An alternative hypothesis is that the brain favors simpler learning rules that do not necessarily optimize the readout of information from visual cortical neurons. Here we have examined these hypotheses by reversibly inactivating visual area V4 in non-human primates at different stages of training on shape discrimination tasks. We find that V4 inactivation generally has a behavioral effect for only a subset of the stimuli that are encoded in the V4 population activity, specifically those that can be represented efficiently in the population firing rate. As a result, there is little relationship between neural measures of discriminability and the causal contribution of V4 neurons to a particular task. This pattern of results can be explained by incorporating a strong inductive bias for simpler perceptual readouts into existing theoretical frameworks. Such a simplicity bias is suboptimal in the sense that it ignores information that could theoretically be extracted from the neural population, but it has the likely advantage of facilitating efficient learning on ecologically-relevant timescales.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| 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 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".