Shiny Eyes is Now Available on

Colin Quirk · 2020/01/23 · 2 minute read

My project ShinyEyes for manually coding eyetracking data is now available on! To check it out, click here. The source is available here. If you wish (or if I run out of free hours) you can always download the source and run it locally.

Most (all?) researchers use automatic algorithms to classify their data into fixations, saccades, blinks, smooth pursuit, and other bins. However, in my experience it is difficult to check and alter these classifications if you have a lot of trials. There is research into the effectiveness of these algorithms, but personally I found that the algorithms fail more often than I was personally comfortable with. For example, I found that the EyeLink algorithm could not differentiate between a saccade and the beginning of a blink.

I hope that ShinyEyes can meet two needs. The first is providing an easy way to explore your data. Once your data is in the csv format ShinyEyes requires, you can flip through your trials individually and optionally view a gif of the trial. Secondly, I hope that if you do need to alter your data due to misclassifications ShinyEyes will make it easy. If you upload your own codes you can fix any issues you come across and download the new codes.

For full usage instructions, check the readme on github. If you use ShinyEyes, let me know! I am happy to add requested functionality if it will make the tool more useful for you.