A Study on the Extraction and Analysis of a Large Set of Eye Movement Features during Reading

📝 Abstract

This work presents a study on the extraction and analysis of a set of 101 categories of eye movement features from three types of eye movement events: fixations, saccades, and post-saccadic oscillations. The eye movements were recorded during a reading task. For the categories of features with multiple instances in a recording we extract corresponding feature subtypes by calculating descriptive statistics on the distributions of these instances. A unified framework of detailed descriptions and mathematical formulas are provided for the extraction of the feature set. The analysis of feature values is performed using a large database of eye movement recordings from a normative population of 298 subjects. We demonstrate the central tendency and overall variability of feature values over the experimental population, and more importantly, we quantify the test-retest reliability (repeatability) of each separate feature. The described methods and analysis can provide valuable tools in fields exploring the eye movements, such as in behavioral studies, attention and cognition research, medical research, biometric recognition, and human-computer interaction.

💡 Analysis

This work presents a study on the extraction and analysis of a set of 101 categories of eye movement features from three types of eye movement events: fixations, saccades, and post-saccadic oscillations. The eye movements were recorded during a reading task. For the categories of features with multiple instances in a recording we extract corresponding feature subtypes by calculating descriptive statistics on the distributions of these instances. A unified framework of detailed descriptions and mathematical formulas are provided for the extraction of the feature set. The analysis of feature values is performed using a large database of eye movement recordings from a normative population of 298 subjects. We demonstrate the central tendency and overall variability of feature values over the experimental population, and more importantly, we quantify the test-retest reliability (repeatability) of each separate feature. The described methods and analysis can provide valuable tools in fields exploring the eye movements, such as in behavioral studies, attention and cognition research, medical research, biometric recognition, and human-computer interaction.

📄 Content

1

A Study on the Extraction and Analysis of a Large Set of Eye Movement Features during Reading

Ioannis Rigasa, Lee Friedmana, Oleg Komogortseva

a Department of Computer Science, Texas State University, San Marcos, USA rigas@txstate.edu, lfriedman10@gmail.com, ok@txstate.edu

Abstract This work presents a study on the extraction and analysis of a set of 101 categories of eye movement features from three types of eye movement events: fixations, saccades, and post-saccadic oscillations. The eye movements were recorded during a reading task. For the categories of features with multiple instances in a recording we extract corresponding feature subtypes by calculating descriptive statistics on the distributions of these instances. A unified framework of detailed descriptions and mathematical formulas are provided for the extraction of the feature set. The analysis of feature values is performed using a large database of eye movement recordings from a normative population of 298 subjects. We demonstrate the central tendency and overall variability of feature values over the experimental population, and more importantly, we quantify the test-retest reliability (repeatability) of each separate feature. The described methods and analysis can provide valuable tools in fields exploring the eye movements, such as in behavioral studies, attention and cognition research, medical research, biometric recognition, and human-computer interaction.

Keywords: eye movements, feature extraction, variability, reliability, reading paradigm 2

1. Introduction The extraction of eye movement features that can model the characteristics of the oculomotor system’s structure and functionality is a vital topic in many fields of research. Human eye movements are inherently connected to the guiding mechanisms of visual attention and thus can serve as an investigation tool for cognitive and behavioral studies. The classic study of (Yarbus, 1967) showed in a systematic way that the eye movements performed during the inspection of a visual stimulus are related to the performed cognitive task. With the advances in eye-tracking technology, analysis of eye movements was adopted as a useful tool for the observation of visual behavior in various studies of cognitive psychology in fields like linguistics, spatial processing, reading, and problem solving (Just & Carpenter, 1976; Rayner, 1998). Several research studies were also conducted to explore the underlying mechanisms connecting the generation of eye movements with visual attention and perception (Collins & Doré-Mazars, 2006; Eckstein, Beutter, Pham, Shimozaki, & Stone, 2007; Schütz, Braun, & Gegenfurtner, 2011). The increasing affordability of mobile eye-trackers facilitated the inspection of natural behavior in out-of-the-lab environments (Hayhoe & Ballard, 2005; Land, 2009). The examination of eye movements can also facilitate studies focusing on the interconnections of the oculomotor behavior and individual characteristics. Eye movements have been explored in relation to individual motivation (Kaspar & König, 2011) and the ‘Big 5’ personality traits (agreeableness, conscientiousness, extraversion, neuroticism, and openness) (Rauthmann, Seubert, Sachse, & Furtner, 2012). Recently, vigor of eye movements was associated with the personal impulsiveness during decision-making tasks (Choi, Vaswani, & Shadmehr, 2014). Also, the research of personal eye movement traits has served as the basis for the field of eye movement biometrics (Rigas & Komogortsev, 2017). The properties of eye movements have been investigated in medical research in order to indicate pathophysiological neural abnormalities, and for identifying early signs of neurodegenerative diseases (MacAskill & Anderson, 2016). For example, the reading paradigm was used in the past to explore the eye movement characteristics in cases of early Alzheimer disease (Fernández et al., 2013) and Parkinson disease (Wetzel, Gitchel, & Baron, 2011). The oculomotor behavior has been also studied in 3

relation to various behavioral disorders such as ADHD (Fried et al., 2014) and autism (Klin, Jones, Schultz, Volkmar, & Cohen, 2002; Shirama, Kanai, Kato, & Kashino, 2016). Given the large span of applications of eye movement analysis, the extraction of eye movement features has been always a fragmented research area, with most of the studies focusing on small sets of relevant each time features. This motivated our current work for the presentation of a unified framework for the extraction and analysis of a very large set of eye movement features. We focus on the reading paradigm since it allows for the exploration both of physical and of cognitive properties of the oculomotor activity. The extracted features cover a wide gamut of temporal, positional, and dynamic characteristics of three basic eye movement events: fixations, saccades and post-saccadic oscillations. The contr

This content is AI-processed based on ArXiv data.