The binocular correlation (BC) method for extracting microsaccades is described in a paper by Anna-Katharina Hauperich, Laura K. Young and Hannah E. Smithson ('What Makes a Microsaccade? A review of 70 years of research prompts a new detection method.', Journal of Eye Movement Research, 20XX). For details please refer to this publication.

This demo includes the labelled data from the 5 Oxford participants described in the paper (due to file-size restrictions these need to be downloaded separately and copied into the folder "OxfordDataset"), as well as an example implementation of the BC method.

To use this demo run the file 'MSExtractionDemo.m', which will produce outputs for the 1000Hz data provided with this distribution. To use this code on data recorded at other sampling frequencies the variable 'RecFreq' in the file 'MS_BC_DataPreProcessing' needs to be altered to whatever recording frequency was used. Then 'MS_BC_DataPreProcessing' will ask for input on what framelength and order of smoothing should be used.


The following Matlab toolboxes are required:
robotics_system_toolbox
signal_toolbox
statistics_toolbox


For any questions please contact Anna-Katharina Hauperich at anna.hauperich@gmail.com.