Using eye tracking to support professional learning in vision-intensive professions: a case of aviation pilots - UTU Research Portal

A1 Refereed original research article in a scientific journal

Using eye tracking to support professional learning in vision-intensive professions: a case of aviation pilots

Authors: Hämälainen, Raija; De Wever, Bram; Sipiläinen, Katriina; Heilala, Ville; Helovuo, Arto; Lehesvuori, Sami; Järvinen, Miitta; Helske, Jouni; Kärkkäinen, Tommi

Publisher: SPRINGER

Publishing place: NEW YORK

Publication year: 2024

Journal: Education and Information Technologies

Journal name in source: EDUCATION AND INFORMATION TECHNOLOGIES

Journal acronym: EDUC INF TECHNOL

Number of pages: 31

ISSN: 1360-2357

eISSN: 1573-7608

DOI: https://doi.org/10.1007/s10639-024-12814-9

Self-archived copy’s web address: https://research.utu.fi/converis/portal/detail/Publication/456982052

Abstract

In an authentic flight simulator, the instructor is traditionally located behind the learner and is thus unable to observe the pilot's visual attention (i.e. gaze behaviour). The focus of this article is visual attention in relation to pilots' professional learning in an Airbus A320 Full Flight Simulator. For this purpose, we measured and analysed pilots' visual scanning behaviour during flight simulation-based training. Eye-tracking data were collected from the participants (N = 15 pilots in training) to objectively and non-intrusively study their visual attention behaviour. First, we derived and compared the visual scanning patterns. The descriptive statistics revealed the pilots' visual scanning paths and whether they followed the expected flight protocol. Second, we developed a procedure to automate the analysis. Specifically, a Hidden Markov model (HMM) was used to automatically capture the actual phases of pilots' visual scanning. The advantage of this technique is that it is not bound to manual assessment based on graphs or descriptive data. In addition, different scanning patterns can be revealed in authentic learning situations where gaze behaviour is not known in advance. Our results illustrate that HMM can provide a complementary approach to descriptive statistics. Implications for future research are discussed, including how artificial intelligence in education could benefit from the HMM approach.

Downloadable publication

This is an electronic reprint of the original article.
This reprint may differ from the original in pagination and typographic detail. Please cite the original version.

s10639-024-12814-9.pdf

Funding information in the publication:
The work was supported by the Academy of Finland under Grant numbers 353325 and 331817.