Top-down modulation of spatial frequency extraction
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
According to prominent models of object recognition, the early extraction of low spatial frequencies (SF) modulates in a top-down fashion the later extraction of high SFs. In the present study, we investigated the precise time course of SF extraction during object recognition in 49 healthy adults. On each trial, a short video (333 ms), in which the SFs of an object were randomly sampled across time, was presented. An object name followed and subjects had to indicate if it matched the object. We then performed multiple linear regressions between SF x time sampling planes and accuracy. We observed a continuous extraction of low SFs (1-21.5 cycles per image, cpi) with an extraction of higher SFs (up to 36 cpi) afterwards (t > 4.00, p < .05). This means that some information was extracted at specific moments regardless of what was seen before (i.e., ballistically). Next, we performed the regressions after having weighted trials according to the quantity of low SFs (1-20 cpi) shown in the first 167 ms. We observed that high SFs (up to 35 cpi), but also lower SFs (as low as 3 cpi), led to more accurate responses when they were preceded by low SFs (t > 3.78, p < .05). These results indicate that SF extraction is modulated by the earlier extraction of low SFs (i.e., adaptively). To disentangle adaptive and ballistic aspects of visual processing, we analyzed the modulation of SFs in every frame by low SFs in every preceding frame. Information around 150-242 ms was exclusively modulated by low SFs around 80-96 ms (t > 3.96, p < .05). Altogether, these results suggest a top-down modulation of SF extraction, but not limited to high SFs, and occurring at specific moments. Meeting abstract presented at VSS 2016
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) | 0.001 | 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 it