Modeling Issues with Eye Tracking Data

TL;DR

This paper compares various statistical models for binary eye-tracking data, highlighting unresolved issues and proposing novel approaches that do not rely on observed lag predictors, advancing the analysis methods in eye-tracking research.

Contribution

It introduces three innovative modeling approaches for serial correlation in eye-tracking data without using lag-1 predictors, expanding analytic options.

Findings

Multiple models reveal unresolved issues in eye-tracking data analysis

Novel approaches outperform traditional lag-based models in certain scenarios

Results suggest new directions for methodological development in eye-tracking analysis

Abstract

I describe and compare procedures for binary eye-tracking (ET) data. The basic GLM model is a logistic mixed model combined with random effects for persons and items. Additional models address error correlation in eye-tracking serial observations. In particular, three novel approaches are illustrated that address serial without the use of an observed lag-1 predictor: a first-order autoregressive model and a first-order moving average models obtained with generalized estimating equations, and a recurrent two-state survival model used with run-length encoded data. Altogether, the results of five different analyses point to unresolved issues in the analysis of eye-tracking data and new directions for analytic development. A more traditional model incorporating a lag-1 observed outcome for serial correlation is also included.