Decoding representations of food images within the ventral visual stream
Bibliographic record
Abstract
Food represents one of the most rewarding stimuli present in nature since it is necessary for our survival. Neuroimaging studies have revealed food-selective activation across a broad network of human brain regions, including areas within the ventral visual stream implicated in recognition of visual stimulus categories, including objects and bodies. However, the nature of visual food representations has not been systematically explored. We used representational similarity analysis (RSA) on multivoxel pattern data from functional magnetic resonance imaging (fMRI at 3 Tesla) to investigate whether representations of food stimuli are distinct from those for other visual categories. Moreover, we examined the degree to which representations of food and other stimuli could be accounted for by low-level properties (i.e., similarity in retinal size, luminance, elongation, texture, or silhouette overlap). Healthy normal-weight individuals (n=22) performed a one-back task in the scanner while viewing colored pictures of different object categories (food, body parts, utensils, objects, scrambled images), matched for retinal size and where possible, familiar size in the real world. RSA was applied to regions of interest (ROIs) within the visual system (including early visual cortex, ventral occipito-temporal cortex and lateral occipito-temporal cortex LOTC), and within a food-selective region (orbitofrontal cortex, OFC). RSA revealed distinct activation patterns for food stimuli in each ROI; that is, food images evoked activation patterns similar to other food images but distinct from other object categories. Visual areas also showed distinctions between different categories of non-food stimuli (e.g., bodies vs. tools); whereas food-selective OFC did not, showing only a difference between food and non-food stimuli. Statistical evaluation of competing models suggested that the representation of food images was not simply related to low-level visual properties. Taken together, our results suggest distinct neural representations of food stimuli that warrant further study, ideally with real food rather than images.
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 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 | 0.000 |
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 teacher head, 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".