How do perceptual grouping cues affect image memorability?
Bibliographic record
Abstract
What makes an image memorable? People encounter hundreds of images throughout their days, in real-life situations, on billboards, or on their computers or cell phones. Yet, people do not remember all images equally well. Some images are intrinsically more memorable than others (Bainbridge, Isola, & Oliva, 2013; Isola, Xiao, Parikh, Torralba, & Oliva, 2014; Khosla, Xiao, Torralba, & Oliva, 2012). While image memorability (typically measured as hit rate) is only weakly affected by low-level image properties, such as color, saturation, or spatial frequencies, high-level properties, such as semantics, emotion, popularity, or aesthetics, were shown to have stronger relationships with memorability. Here we investigate the influence on memorability exerted by mid-level perceptual grouping features, such as contour curvature, contour junctions, or local symmetry. To this end, we converted scene images from the FIGRIM Dataset (Bylinskii et al., 2015) into line drawings and computed their mid-level features. Our results suggest a positive relationship between local mirror symmetry and hit rate for the old/new memory task. Interestingly, we also found a positive correlation between local mirror symmetry and false alarm rate. Following re-analysis of the data using signal detection theory, we found no connection between any of the mid-level features and d-prime. In spite of that, we found a strong positive relationship between local mirror symmetry and the decision criterion, indicating that participants were more likely to respond that they had seen the image before, irrespective of whether the image had been seen before or not. We hypothesize that local symmetry improves perceptual fluency by reducing the complexity of features in the image, thereby leading to an enhanced feeling of familiarity, which may lead to a decision bias toward reporting having seen the image before. However, symmetry did not affect the sensitivity of the memory recognition task.
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 | 0.015 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".