Saccades Exert Control of Motion Processing for Smooth Pursuit Eye Movements
This was the first paper David published with Steve Lisberger at UCSF. Perhaps the biggest challenge in writing the paper was to communicate the task that the monkeys were trained to do.
All the MATLAB code
generatepaths.m -- will create a structure with the paths to the data for each monkey, for each experiment.
pull.m -- the main analysis code to parse trials from different types of experiments. Well-commented, this will take you from the individual trials for one experiment to a structure that is easy to analyze. Calls getlastdir.m (for directory parsing), digitaldiff.m (for differentiation) and readcxdata (mex code which will need to be compiled for your platform, for reading individual trials)
preptype*.m -- * refers to a type of experiment; takes the output of pull and runs summary-type analyses, in preparation for figure generation.
figure*.m -- code to generate each of the figures in the manuscript.
.mat files
Each figure has a corresponding .mat file, and each experiment has corresponding .mat file as well.
Data for each type of experiment is stored in the following hierarchy: trial-type/monkey/date/trialfiles.
The paper figures can be re-created from the files above, but the final version of the figures are formatted according to an XML schema compatible with an in-house plotting program, Phyplot. An archive (182 KB) of the XML files corresponding to each figure can be found here.