Cognitive strategies revealed by clustering eye movement transitions

Šimon Kucharský, Gabriela Olivia Trutescu, Martina Zaharieva, Ingmar Visser, Paulo G. Laurence, Maartje E.J. Raijmakers

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

In cognitive tasks, solvers can adopt different strategies to process information which may lead to different response behavior. These strategies might elicit different eye movement patterns which can thus provide substantial information about the strategy a person uses. However, these strategies are usually hidden and need to be inferred from the data. After an overview of existing techniques which use eye movement data for the identification of latent cognitive strategies, we present a relatively easy to apply unsupervised method to cluster eye movement recordings to detect groups of different solution processes that are applied in solving the task. We test the method's performance using simulations and demonstrate its use on two examples of empirical data. Our analyses are in line with presence of different solving strategies in a Mastermind game, and suggest new insights to strategic patterns in solving Progressive matrices tasks.

Original languageEnglish
Pages (from-to)1-20
Number of pages20
JournalJournal of Eye Movement Research
Volume13
Issue number1
DOIs
Publication statusPublished - 26 Feb 2020

Keywords

  • Cognitive strategies
  • Eye tracking
  • Latent groups
  • Scanpaths
  • Unsupervised clustering

Cite this