A flicker change detection task reveals object-in-scene memory across species
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Tests of recognition memory in macaques typically assay memory for objects or isolated images, over time spans of seconds to hours from stimulus presentation, and/or require extensive training. Here, we propose a new application of the flicker change detection task that could measure object-in-scene memory days after single-trial exposures. In three experiments, participants searched for a changing object - or "target" - embedded within a scene as their eye movements were tracked. For new targets-in-scenes, the change is difficult to detect and requires extensive search. Once the target is found, however, the change becomes obvious. We reasoned that the decreased times required to find a target in a repeated scene would indicate memory for the target. In humans, targets were found faster when the targets-and-scenes were explicitly remembered than when they were forgotten, or had never been seen before. This led to faster repeated-trial compared to novel-trial search times. Based solely on repeated-trial search times, we were able to select distributions comprised of predominantly remembered or predominantly forgotten trials. Macaques exhibited the same repetition effects as humans, suggesting that remembered trials could be dissociated from novel or forgotten trials using the same procedures we established in humans. Finally, an anterograde amnesic patient with damage that included the medial temporal lobe (MTL) showed no search time differences, suggesting that memory revealed through search times on this task requires MTL integrity. Together, these findings indicate that the time required to locate a changing object reveals object-in-scene memory over long retention intervals in humans and macaques.
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.
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 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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 it