The semantic distance between a linguistic prime and a natural scene target predicts reaction times in a visual search experiment
Bibliographic record
Abstract
Does reading a semantically similar sentence description of a scene make you faster at subsequently detecting that scene in a visual search task? How does this vary across individuals when everyone has different conceptual knowledge? To investigate the former, 95 subjects were asked to identify a target natural scene in a visual search experiment where the display of a target scene and five semantically related distractors was preceded by a sentence prime. Every scene from our stimuli set was sampled as a target under three conditions: the prime was either the same as the target, halfway between the target and the semantically farthest item from the target, or the semantically farthest item from the target. The semantic distances between each pair of items were averaged Euclidean distances from the multi-arrangements (MA) task collected as part of the Natural Scenes Dataset (NSD). A subset of 27 subjects also completed MA on both the linguistic primes and scene targets, allowing us to use the idiosyncratic distances as predictors of each subject’s RTs. A generalised linear mixed effects model (GLMM) revealed that after removing the zero-distance condition (prime same as the target) the averaged NSD distances (n = 84, slope = .953, t = 5.229, p < .001) do indeed predict the RTs. Strikingly, the strongest effect emerged when the idiosyncratic distances from the captions MA (n = 26, slope = 1.165, t = 3.105, p < .01) were used as the predictors. These findings are significant in at least two major respects: firstly, linguistic primes promote visual targets the more semantically related they are. Secondly, this seems to be closely linked to the subject-specific similarity judgements which extends previous knowledge of semantic priming and opens a discussion as to what level the linguistic input intertwines with one's visual representations.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| 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".