The MultiMatch method proposed by Jarodzka, Holmqvist and Nyström (2010), implemented in Matlab as the MultiMatch toolbox and validated by Dewhurst and colleagues (2012) is a vector-based, multi-dimensional approach to compute scan path similarity.

For a complete overview of this software, please take a look at the Documentation

The method represents scan paths as geometrical vectors in a two-dimensional space: Any scan path is build up of a vector sequence in which the vectors represent saccades, and the start and end position of saccade vectors represent fixations. Two such sequences (which can differ in length) are compared on the five dimensions 'vector shape', 'vector length' (saccadic amplitude), 'vector position', 'vector direction' and 'fixation duration' for a multidimensional similarity evaluation (all in range [0, 1] with 0 denoting maximal dissimilarity and 1 denoting identical scan paths on the given measure). The original Matlab toolbox was kindly provided via email by Dr. Richard Dewhurst and the method was ported into Python with the intent of providing an open source alternative to the matlab toolbox.

It is recommended to use a dedicated virtualenv:

# create and enter a new virtual environment (optional)
virtualenv --python=python3 ~/env/multimatch
. ~/env/multimatch/bin/activate

multimatch-gaze can be installed via pip. To automatically install multimatch-gaze with all dependencies (pandas, numpy, and argparse), use:

# install from pyPi
pip install multimatch-gaze

Bug reports, feedback, or any other contribution are always appreciated. To report a bug, request a feature, or ask a question, please open an issue. Pull requests are always welcome. For additional information on how to contribute, checkout CONTRIBUTING.md.

required inputs:

• two tab-separated files with nx3 fixation vectors (x coordinate in px, y coordinate in px, duration)
• screensize in px (x dimension, y dimension)

multimatch-gaze data/fixvectors/segment_10_sub-19.tsv data/fixvectors/segment_10_sub-01.tsv 1280 720

optional inputs:

if scan path simplification should be performed, please specify in addition

• --amplitude-threshold (-am) in px
• --direction-threshold (-di) in degree
• --duration-threshold (-du) in seconds

Example usage with grouping:

multimatch-gaze data/fixvectors/segment_10_sub-19.tsv data/fixvectors/segment_10_sub-01.tsv 1280 720 --direction-threshold 45.0 --duration-threshold 0.3 --amplitude-threshold 147.0

REMoDNaV helper:

Eye movement event detection results produced by REMoDNaV can be read in natively by multimatch-gaze. To indicate that datafiles are REMoDNaV outputs, supply the --remodnav parameter.

multimatch-gaze data/remodnav_samples/sub-01_task-movie_run-1_events.tsv data/remodnav_samples/sub-01_task-movie_run-2_events.tsv 1280 720 --remodnav

REMoDNaV can classify smooth pursuit movements. As a consequence, when using REMoDNaV output, users need to indicate how these events should be treated. By default, multimatch-gaze will discard pursuits. In some circumstances, however, it can be useful to include pursuit information. Moving stimuli for example would evoke a pursuit movement during visual intake. When specifying the --pursuit keep parameter, the start and end points of pursuits will be included in the scan path.

multimatch-gaze data/remodnav_samples/sub-01_task-movie_run-1_events.tsv data/remodnav_samples/sub-01_task-movie_run-2_events.tsv 1280 720 --remodnav --pursuit keep

Dewhurst, R., Nyström, M., Jarodzka, H., Foulsham, T., Johansson, R. & Holmqvist, K. (2012). It depends on how you look at it: scanpath comparison in multiple dimensions with MultiMatch, a vector-based approach. Behaviour Research Methods, 44(4), 1079-1100. doi: 10.3758/s13428-012-0212-2.

Dijkstra, E. W. (1959). A note on two problems in connexion withgraphs. Numerische Mathematik, 1, 269–271. https://doi.org/10.1007/BF01386390

Jarodzka, H., Holmqvist, K., & Nyström, M. (eds.) (2010). A vector-based, multidimensional scanpath similarity measure. In Proceedings of the 2010 symposium on eye-tracking research & applications (pp. 211-218). ACM. doi: 10.1145/1743666.1743718