Skip to content

Repository files navigation

RDK_vMMM

Random dot kinematogram data analysis using a von Mises mixture model

This code belongs to the paper "Psychophysics and computational modeling of feature-continuous motion perception" by Töpfer, Barbieri et al. (2022), accepted for publication in the Journal of Vision. It consists of MATLAB scripts processing behavioral data, also available in this repository, and generating results and figures, as they appear in the paper.

Requirements

This code was written by Felix Töpfer (main folder) and Joram Soch (sub-folder tools/vMMM/).

The data analysis was developed and run in MATLAB R2019b. No further toolboxes are required.

Instructions

To re-analyze the behavioral data, proceed as follows:

  1. Clone the repository to some folder on your computer.
  2. Move to that folder and run BEHAVIOR_ANALYSIS.m.
  3. Modify line 21 (see line 22) and re-run the script.

This should generate Figures 3/4, 5A/B, 6/7 and 9-16 from the paper, depending on whether you have chosen the response method to be meth = 'trackball' (trackball, see step 2) or meth = 'bar' (rotating bar, see step 3). Note that the panels of Figure 3/4 appear in separate windows, as each panel consists of several subplots. These windows are labeled as "Scatterplot of trial-wise responses (stimulus-type: coherence-level)" where stimulus_type is either "TM" (transparent motion), "BM" (Brownian motion) or "WM" (white noise motion) and coherence-level varies between 0, 12.5, 25, 50 and 100%.

About

Random dot kinematogram (RDK) data analysis using a von Mises mixture model (vMMM)

Resources

Stars

3 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages