Quickstart Mousetracking

This repository provides (almost) everything you need to get started at mouse tracking research using python/OpenSesame.

-->  -->  -->  -->  -->  -->  -->  -->  -->  -->  -->  
Beginner's tip: Download this entire project as a
zip file using the button to the right. 
-->  -->  -->  -->  -->  -->  -->  -->  -->  -->  -->  

Data Collection

The procedure directory contains a temlate for a simple mouse tracking experiment implemented using the OpenSesame experiment builder, which you can install from here. You can also see my tutorial on mousetracking in OpenSesame here. Note that you'll need to adjust the experiment's screen resolution in OpenSesame if your monitor is different to mine.

Data Analysis

I've completed this experiment a few times myself, and copied the results into the results directory (in the data directory within it). The results directory also contains a python script, process.py, which will parse the individual .csv files in results/data/, merge them, and calculate relevant summary statistics, as well as saving standardised cursor trajectories for plotting and analysis.

To do this, you'll need to have a few things installed:

• An interpreter for the python programming language.
• The SciPy set of python scientific packages. The easiest way to install all of these on Windows is to download the anaconda scientific pyton distribution, which is a version of python which comes with almost all of the scientific tools you could need preinstalled.
• Squeak, my python package for processing mouse data. The easiest way to install this is to type pip install squeak at the command prompt.

Finally, a note on usability: I've tried to make all of this as user-friendly as possible, but this remains a solution for researchers with at least some computing know-how. To modify the example experiment provided, you'l need to be able to at least understand the python code which powers it. Python is perhaps the easiest programming language to learn, and a practical application like this is a great way to start, but it will still be more difficult than clicking on a magic button. Running the analyses also requires a little bit of knowledge about running python scripts, and even just what a script is, more generally. This is something which isn't difficult to learn though, and there are plenty of resources online (here is a good starting point).

If all of this sounds like a lot of work, I want to stress that Jon Freeman's MouseTracker is an excellent tool for mouse tracking research, and should be the first option most people look to. What I provide here is intended for researchers who want to design experiments which go outside MouseTracker's capabilities (and if MouseTracker is to be easy to use, it can't support every conceivable experimental design), or for those interested in learning a little more about the fine detail of mouse trajectory recording and analysis.

Happy tracking!