Eye movements predict students' computer-based assessment performance of physics concepts in different presentation modalities

Sheng Chang Chen, Hsiao Ching She, Ming Hua Chuang, Jiun Yu Wu, Jie Li Tsai, Tzyy Ping Jung

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Despite decades of studies on the link between eye movements and human cognitive processes, the exact nature of the link between eye movements and computer-based assessment performance still remains unknown. To bridge this gap, the present study investigates whether human eye movement dynamics can predict computer-based assessment performance (accuracy of response) in different presentation modalities (picture vs. text). Eye-tracking system was employed to collect 63 college students' eye movement behaviors while they are engaging in the computer-based physics concept questions presented as either pictures or text. Students' responses were collected immediately after the picture or text presentations in order to determine the accuracy of responses. The results demonstrated that students' eye movement behavior can successfully predict their computer-based assessment performance. Remarkably, the mean fixation duration has the greatest power to predict the likelihood of responding the correct physics concepts successfully, followed by re-reading time in proportion. Additionally, the mean saccade distance has the least and negative power to predict the likelihood of responding the physics concepts correctly in the picture presentation. Interestingly, pictorial presentations appear to convey physics concepts more quickly and efficiently than do textual presentations. This study adds empirical evidence of a prediction model between eye movement behaviors and successful cognitive performance. Moreover, it provides insight into the modality effects on students' computer-based assessment performance through the use of eye movement behavior evidence.

Original languageEnglish
Pages (from-to)61-72
Number of pages12
JournalComputers and Education
Volume74
DOIs
Publication statusPublished - 2014 May 1

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Education

Fingerprint Dive into the research topics of 'Eye movements predict students' computer-based assessment performance of physics concepts in different presentation modalities'. Together they form a unique fingerprint.

Cite this