PUPILS pipeline: A flexible Matlab toolbox for eyetracking and pupillometry data processing

TL;DR

The PUPILS toolbox is a flexible Matlab tool designed to simplify processing of noisy eye tracking and pupillometry data by detecting, interpolating, and denoising raw recordings for easier analysis.

Contribution

It introduces a comprehensive Matlab toolbox that automates pupillary data processing, including event detection, interpolation, and noise reduction, tailored for researchers.

Findings

Effective removal of noise and artifacts from pupillometry data.

Automated detection of blinks and saccades.

Streamlined data export for further analysis.

Abstract

With the development of widely available commercial eye trackers, the use of eye tracking and pupillometry have become prevalent tools in cognitive research, both in industry and academia. However, dealing with pupil recordings often proves challenging, as the raw data produced by the eye tracker is subject to noise and artifacts. With the PUPILS toolbox the authors wish to simplify the processing of pupil data for researchers. The toolbox receives the raw data extracted from the eye tracker output files and provides an analysis of the relevant pupillary events (e.g., blinks and saccades), interpolates missing data and denoises the recorded traces. The final output of the toolbox is a cleaned up recording that can be easily exported for use in further analysis.