Detection of Complexity in General and Medical-language Texts Using Eye-Tracking Data

Ivchenko, Oksana; Grabar, Natalia

Document type :

Communication dans un congrès avec actes

Title :

Detection of Complexity in General and Medical-language Texts Using Eye-Tracking Data

Author(s) :

Ivchenko, Oksana [Auteur]
Savoirs, Textes, Langage (STL) - UMR 8163 [STL]
Grabar, Natalia [Auteur]

Savoirs, Textes, Langage (STL) - UMR 8163 [STL]

Conference title :

La complexité en Sciences du langage

Conference organizers(s) :

Université de Nanterre

City :

Paris

Country :

France

Start date of the conference :

2024-12-12

English keyword(s) :

Medical Texts
Types of Texts
Simplification
Reading
Eye-Tracking
Fixations

HAL domain(s) :

Sciences cognitives

English abstract : [en]

This study explores the complexity of medical texts, aiming to objectively detect difficult words and passages using eye-tracking data. We investigate how four types of texts (general topics from Wikipedia, medical topics ...
Show more >This study explores the complexity of medical texts, aiming to objectively detect difficult words and passages using eye-tracking data. We investigate how four types of texts (general topics from Wikipedia, medical topics from Wikipedia, clinical cases, and their manually simplified versions) are read and understood by participants. These simplifications were made manually at both lexical and syntactic levels, following text simplification guidelines. Using eye-tracking technology, we measure the cognitive load imposed by these texts, focusing on key eye-tracking metrics that serve as reliable indicators of reading difficulty (Ekstrand et al., 2021; Clifton et al., 2007; Singh et al., 2016). The goal is to identify which lexical and syntactic constructions in medical texts hinder comprehension and to provide insights for simplifying such texts. To date, we analyze data from 50 participants. Eye-tracking metrics are being analyzed both quantitatively and qualitatively to identify reading patterns and assess how reading behavior varies across different text types. In a preliminary analysis, we compare the number and duration of fixations across four types of texts. Fixations, characterized by brief pauses during reading, are crucial for information processing and serve as indicators of cognitive engagement with the text. Longer fixations often signal processing difficulty or heightened interest, while more frequent fixations may suggest that the text is either challenging or highly engaging for the reader. The average duration of an eye fixation on a word during reading varies depending on several factors, including the complexity of the text, the reader’s familiarity with the content, and the purpose of reading (Hyönä & Kaakinen, 2019). Generally, studies on eye movements during reading indicate that adults typically fixate on a word for about 200 to 250 milliseconds (ms) when reading in their native language under normal circumstances (Rayner & Reingold, 2015).Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

ANR Project :

Communication, Literacy, Education, Accessibility, Readability

Collections :

Savoirs, Textes, Langage (STL) - UMR 8163

Source :

Harvested from HAL

Detection of Complexity in General and ... BibTeX CSV Excel RIS

Detection of Complexity in General and ...

BibTeX

CSV

Excel

RIS