A new method for comparing scanpaths based on vectors and dimensions
Bibliographic record
Abstract
We make different sequences of eye movements – or scanpaths – depending on what we are viewing and the current task we are carrying out (e.g. Land, Mennie, & Rusted, 1999). In recent years, research efforts have been very informative in identifying commonalities between scanpath pairs, allowing us to quantify, for example, the similarity in eye movement behaviour between experts and novices (Underwood, Humphrey, & Foulsham, 2008), or between encoding and recognition of the same image (Foulsham & Underwood, 2008). However, common methods for comparing scanpaths (e.g., ‘string-edit’, based on Levenshtein, 1966, or ‘positon measures’, see Mannan, Ruddock, & Wooding, 1995) fail to capture both the spatial and temporal aspects of scanpaths. Even the newest techniques (e.g., ‘Scanmatch’, Cristino, Mathôt, Theeuwes, & Gilchrist, 2010) are restricted by the fact that they rely on the division of space into Areas of Interest (AOIs), thus limiting the spatial resolution of the similarity metric produced. Here we validate a new algorithm for comparing scanpaths (Jarodzka, Holmqvist, & Nyström, 2010) with eye movement data from human observers. Instead of relying on the quantization of space into AOIs, our method represents scanpaths as geometrical vectors, which retain temporal order and spatial position. Scanpaths are then compared across several dimensions – shape, position, length, direction, and duration – and a similarity value is returned for each. Using this new multidimensional approach, our data from two experiments highlights aspects of scanpath similarity which cannot otherwise be quantified: when scanpaths are clearly similar, but are spatially downscaled, for instance. Moreover, we show how scanpath similarity changes depending on task, using our algorithm in comparison to the most popular alternatives. This data demonstrates that our vector-based multi-dimensional approach to scanpath comparison is favorable to others, and should encourage a shift away from methods which are rooted in the Levenstein principle or spatial position alone.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| 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".