Modeling Issues with Eye Tracking Data

TL;DR

This paper investigates how binary eye-tracking data can be analyzed when serial correlation is present. It compares a traditional generalized linear mixed model with random effects to several alternative approaches, including lag-1 predictors, AR(1)-based generalized estimating equations, and run-length-encoded two-state survival models. The findings show that AR(1) and survival approaches yield effect estimates similar to GLM but with larger standard errors and distinct interpretations, while run-length encoding enables efficient survival analyses. The study highlights unresolved issues in modeling eye-tracking data, notes identifiability challenges, and points to Markov-transition frameworks as a fruitful direction for future work.

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.