Size constancy in face discrimination: synthetic faces & principal componentss
Bibliographic record
Abstract
There is some evidence that faces are harder to recognize across a range of sizes using face photographs (Kolers et al., 1985). This may be because the face photographs would have provided more detailed information at larger sizes. We investigate whether thresholds for face discrimination exhibit size constancy using synthetic faces which all have the same information content for face identification. Thresholds for face discrimination were measured using 4-D synthetic face cubes (Wilson et al., ARVO, 2001). The experiment consisted of a 110 ms presentation of one face followed by a noise mask for 110 ms. Then two comparison faces appeared on the screen and a subject selected the face matched to the flashed face in a 2 AFC paradigm. We varied the linear size ratio (1:1, 2:1, or 4:1) between the target face and the comparison faces. When constructing regular face cubes, thresholds for face discrimination tended to remain constant across the size changes between target and comparison faces for both front and 20 deg side views of faces. When constructing face cubes based on principal components of our entire face population, thresholds for face discrimination tended to be unaffected by the size change between the flashed face and the comparison faces over a considerable range of the principal components. Only when the small face was flashed, the thresholds in higher and lower order principal components were affected by the size change. A control experiment in which small faces were flashed for 250 ms restored size constancy. These findings indicate that thresholds for face discrimination exhibit size constancy over a 4:1 size range for most principal components and that size constancy breaks down only when a small face defined by high spatial frequencies is briefly flashed. A longer processing time for small high frequency faces restores size constancy.
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 | 0.003 |
| 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 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".