Information transfer during goal-directed viewing of everyday scenes
Bibliographic record
Abstract
One of the fundamental issues in visual perception is how and how much visual information is transferred across fixations. Here, we examine how humans actively scan the visual environment when performing goal-directed visual search on photographs of complex everyday scenes. We analyze data from two open datasets of eye-movements made by participants performing either category-search with 18 target categories (COCO-Search18) or free-viewing (COCO-FreeView) on over 4000 unique naturalistic images from the MS-COCO dataset. We focus specifically on the evidence for information transfer across saccades as revealed by saccades made after short inter-saccadic intervals (< 125 ms, short-latency saccades). When the target is present in the scene, participants (n = 10) fixate it after predominantly one saccade (45% of trials) or two saccades (36% of trials). Short-latency second saccades occur frequently (45% of second saccades on average) in goal-directed visual search. These saccades foveate the search target more often than saccades executed after longer intersaccadic intervals (regular-latency saccades). Short-latency second saccades are not small-amplitude corrective saccades: they are both more common and more likely to foveate the target when they follow first saccades that end further away from the target. Further, they are much more frequent during goal-directed visual search with the search target present than when the search target is absent or during free-viewing: active searching, and the top-down salience of the search-target contribute to increasing the frequency of short-latency saccades. The results show that human searchers use a satisficing strategy when actively searching complex everyday scenes for a categorically defined target. Short-latency saccades and information transfer across saccades work towards ensuring that the cost of making additional saccades to distractor stimuli is minimal; this would not be the case if perception began anew at each fixation. Information integration and transfer across saccades plays a prominent role during naturalistic vision.
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 | 0.002 |
| 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".