TY - CHAP A1 - Nivala, Markus A1 - Hauser, Florian A1 - Mottok, Jürgen A1 - Gruber, Hans T1 - Developing visual expertise in software engineering: An eye tracking study T2 - IEEE Global Engineering Education Conference (EDUCON), 10-13 April 2016, Abu Dhabi, UAE N2 - Program comprehension and the ability to find program errors are key skills of software engineering. The aim of this pilot study was to examine the visual processes of novice and advanced programmers in authentic tasks. Fifteen novices and eight advanced programmers were given eight short pieces of code. Their task was to either identify an error or give the output of the code. Eye movements and keyboard activity were recorded. On average, the novices spent more time reading the code than composing the response, whereas the more advanced programmers started composing the response sooner and spent more time on it. In general, the advanced programmers had shorter fixations and saccades. The results suggest that the advanced programmers are quicker to grasp the essence of the code and able to see more details in it. The advanced programmers had shorter fixations and saccade lengths during the second phase which might indicate the process of chunking. KW - program comprehension KW - debugging KW - software engineering KW - visual expertise KW - eye tracking Y1 - 2016 U6 - https://doi.org/10.1109/EDUCON.2016.7474614 SP - 613 EP - 620 ER - TY - JOUR A1 - Gegenfurtner, Andreas A1 - Boucheix, Jean-Michel A1 - Gruber, Hans A1 - Hauser, Florian A1 - Lehtinen, Erno A1 - Lowe, Richard K. T1 - The gaze relational index as a measure of visual expertise JF - Journal of Expertise N2 - Eye tracking is a powerful technique that helps reveal how people process visual information. This paper discusses a novel metric for indicating expertise in visual information processing. Named the Gaze Relational Index (GRI), this metric is defined as the ratio of mean fixation duration to fixation count. Data from two eye-tracking studies of professional vision and visual expertise in using 3D dynamic medical visualizations are presented as cases to illustrate the suitability and additional benefits of the GRI. Calculated values of the GRI were higher for novices than for experts, and higher in non-representative, semi-familiar / unfamiliar task conditions than in domain-representative familiar tasks. These differences in GRI suggest that, compared to novices, experts engaged in more knowledge-driven, top-down processing that was characterized by quick, exploratory visual search. We discuss future research aiming to replicate the GRI in professional domains with complex visual stimuli and to identify the moderating role of cognitive ability on GRI estimates. KW - eye tracking KW - fixations KW - information processing KW - professional vision KW - visual expertise Y1 - 2020 UR - https://www.journalofexpertise.org/articles/volume3_issue1/JoE_3_1_Gegenfurtner.pdf SN - 2573-2773 VL - 3 IS - March SP - 32 EP - 40 PB - Michigan State University ER -