Gender- and age-contingent face aftereffects and the Hebbian normalization model
Bibliographic record
Abstract
Classical adaptation phenomena (e.g., tilt after-effect, waterfall illusion) are a response to a biased distribution of stimuli (e.g., more right-tilted adapters) and led to models involving gain adaptation (highly stimulated neurons reduce their gain). But, neurons also adapt to 2nd-order statistics (stimulus feature co-occurrence, “contingent adaptation”, Benucci et al., 2013; Aschner et al., 2018), consistent with the Hebbian normalization model (Westrick et al., 2016) in which strong co-firing of neuron pairs leads to increased mutual inhibition. This model predicts analogous behavioral effects: Co-occurring stimulus elements in an adapter lead to inhibitory effects of one on the other in perception as first tested using pairs of grating stimuli (Yiltiz et al., VSS 2018 and in press). Here, we test whether contingent adaptation, as predicted by the model, applies to adaptation to high-level perceptual features. We adapted observers (N=19) to a series of alternating old-male and young-female faces (or old-female/young-male) followed by an androgynous or a middle-aged test face. Despite the fact that there was no net first-order gender or age adaptation, we found significant age-contingent gender aftereffects (d=0.77, p=.02) biasing perception away from the adapting gender (i.e., repulsive): test faces were perceived as more masculine or feminine depending on the age of the face with which they were paired during adaptation, and analogous gender-contingent age aftereffects (d=0.63 , p=.02). Prior face adaptation work has shown figural aftereffects (e.g., eye-spacing, internal features distortion) contingent on some discrete facial attributes such as race and orientation (Rhodes et al., 2004; Jacquet, et al., 2008). Our results represent the first report of contingent aftereffects on natural face categories (i.e., age and gender) that are plausibly represented by a continuum of neurons coding for a range of values along these dimensions. The Hebbian normalization model provides an account for these contingent adaptation aftereffects in face perception.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 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 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 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".