Automated Analysis Pipeline for Extracting Saccade, Pupil, and Blink Parameters Using Video-Based Eye Tracking
Bibliographic record
Abstract
The tremendous increase in the use of video-based eye tracking has made it possible to collect eye tracking data from thousands of participants. The traditional procedures for the manual detection and classification of saccades and for trial categorization (e.g., correct vs. incorrect) are not viable for the large datasets being collected. Additionally, video-based eye trackers allow for the analysis of pupil responses and blink behaviors. Here, we present a detailed description of our pipeline for collecting, storing, and cleaning data, as well as for organizing participant codes, which are fairly lab-specific but nonetheless, are important precursory steps in establishing standardized pipelines. More importantly, we also include descriptions of the automated detection and classification of saccades, blinks, "blincades" (blinks occurring during saccades), and boomerang saccades (two nearly simultaneous saccades in opposite directions where speed-based algorithms fail to split them), This is almost entirely task-agnostic and can be used on a wide variety of data. We additionally describe novel findings regarding post-saccadic oscillations and provide a method to achieve more accurate estimates for saccade end points. Lastly, we describe the automated behavior classification for the interleaved pro/anti-saccade task (IPAST), a task that probes voluntary and inhibitory control. This pipeline was evaluated using data collected from 592 human participants between 5 and 93 years of age, making it robust enough to handle large clinical patient datasets. In summary, this pipeline has been optimized to consistently handle large datasets obtained from diverse study cohorts (i.e., developmental, aging, clinical) and collected across multiple laboratory sites.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.013 | 0.011 |
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".