In many daily life activities, eye movements provide strong clues about underlying cognitive processes. For example, patients with cognitive deficits have atypical eye movement patterns. People’s eye movement behavior before decision making reveals their preference. Thus, eye movement has become an important measure in the broad research fields on human behavior.

Recent research has reported substantial individual differences in eye movements during visual tasks. Nevertheless, most of the current analysis methods do not adequately reflect these individual differences. In addition, they focus on spatial information (fixation locations) whereas temporal information (transitions among fixation locations) is typically overlooked. In view of these issues, Chuk, Chan, and Hsiao (2014) have developed a novel eye movement data analysis method, Eye Movement analysis with Hidden Markov Models (EMHMM), which summarizes each individual’s eye movement pattern using a hidden Markov model (HMM; a type of machine learning model for time
series data), including person-specific regions of interest (ROIs) and transition probabilities among the ROIs. Individual HMMs can be clustered according to similarities to discover common patterns, and the similarity between individual patterns can be quantitatively assessed through machine learning methods. This similarity measure then can be used to examine associations between eye movement patterns and other behavioral, cognitive, or neuroscience measures. This method has now been applied to different types of visual tasks across different fields and made discoveries thus far not
revealed by other methods. New methodologies for more complex cognitive tasks have also been developed, including using switching HMMs for tasks involving cognitive state changes such as a decision-making task, and using the machine learning algorithm co-clustering for tasks involving stimuli with different feature layouts such as a visual search task.

In short, the EMHMM methodology allows us to summarize, quantitatively assess, and compare individual eye movement patterns across stimuli and tasks, and examine how they are associated with other measures. The Matlab Toolbox for
EMHMM is available at http://visal.cs.cityu.edu.hk/research/emhmm/. This minicourse aims to introduce the use of eye tracking technology in experimental research and how to conduct eye movement data analysis using the EMHMM methodology, so that students can use it in their own research.