The effect of object features on target and identity localization in multiple identity tracking
Bibliographic record
Abstract
It is debated whether multiple object tracking (MOT) and multiple identity tracking (MIT) involve distinct processes. In a standard MOT task, all items (targets and distractors) are identical while in a standard MIT task, all items are unique. Both tasks involve localization, although MOT only requires distinguishing targets from distractors (target localization) whereas MIT requires distinguishing targets from distractors as well as distinguishing targets from other targets (identity localization). To test whether the processes used in target localization and identity localization are the same, we used an MIT task with 16 unique items representing every combination of four colours and four shapes and manipulated target similarity: a variable that may have different effects on target localization and identity location. Specifically, distinguishing targets from distractors should be easier when targets share a feature that differentiates them from distractors (e.g., targets are red items) compared to when targets do not share any features with each other (e.g., targets are four different shapes with four different colours). In contrast, distinguishing targets from other targets should be more difficult when the targets are similar (e.g., targets are red items) than when they are dissimilar (e.g., targets are four different shapes with four different colours). Performance was assessed by prompting the participant to report the location of specific targets one at a time. Accuracy was scored in two ways: identity localization score – the percentage of reported objects that correctly matched the identity of the prompted target – and target localization score – the percentage of reported objects that were targets, regardless of the prompted target. If distinguishing between targets and distractors and distinguishing between targets and other targets rely on different operations, then there should be a greater advantage for target similarity in target localization compared to identity localization. Results supported this prediction.
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 | 0.003 | 0.030 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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 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".