Abstract :
[en] Impact of two video formats (Viso by Noldus) on the visual strategies and user experience of pre-service teachers in French-speaking Belgium for post-simulation debriefing.
SUMMARY STATEMENT
The aim of the study is to assess the impact of two video presentation formats on pre-service teachers in terms of a) objects viewed and visual behaviour recorded by an eyetracker and b) user experience.
KEYWORDS
pre-service teachers, debriefing, eye tracking, video format, Viso
PROBLEM STATEMENT
As part of the teacher training developed by Bocquillon (2020), pre-service teachers practice their practical skills during micro-teaching sessions. This approach involves them giving a lesson and then debriefing on the basis of a video extract of their performance. The choice of this extract is made by the pre-service teachers after watching their whole lesson before the debriefing.
To achieve this, the room where the micro-teaching takes place is equipped with four cameras (A, B, C, D) positioned in each corner of the room. As part of the ARC Sim'Pro research programme (Derobertmasure, Demeuse, Van Daele, & Gosselin, 2020), the recording system has been modernised, in particular by introducing the Viso software (Noldus). Viso has two video generation methods: it can either generate four full-screen video streams (View_P) or a single stream made up of four images of equal size, each image representing the stream from a different camera, known as the "mosaic format" (View_M).
To prepare their debriefing, the trainers chose to send the View_M to the pre-service teachers rather than the View_P. However, the View_M means that there are many elements on the screen, some of which move and are reduced in size, thus increasing the density of information to be processed. The ability of pre-service teachers to perceive and interpret classroom situations and to make links between the different elements observed to form a coherent whole is not yet very effective at this stage of their training (e.g. Keskin, Seidel, Stürmer, & Gegenfurtner, 2024).
QUESTION
Compared to a View_P, we want to understand how the View_M influences
a) the objects viewed
b) the visual strategies used by pre-service teachers in terms of changing trajectories between the 4 views (Figure 2)
c) the user experience of pre-service teachers
METHODOLOGY
An excerpt (1'10'') featuring a struggling student and an off-task student was viewed by 12 pre-service AESS teachers on a fixed eye tracking device (Gazepoint). The presentation was made in random order, with a 5 second interval between each viewing. The excerpt is presented in two formats: View_P and View_M.
1.Objects viewed
According to View_P or View_M, the analysis compared the points of attention of the pre-service teachers in terms of the duration of fixation of zones of interest (ZDI) on the teacher, the students, their teaching material (e.g. the mobile phone) or personal material (e.g. the water bottle) and elements of the setting (e.g. the control room window).
2.Visual strategies
Three types of indicators were used to understand how pre-service teachers navigate visually in the_M view, namely a) the duration of fixations for each ZDI and b) the frequency of changes of gaze direction from one zone to another (e.g. from A to B; from A to C, etc.).
3.User experience
Each prospective teacher completed a validated 'short user experience' scale (Schrepp et al., 2017) after viewing the extracts in View_P and View_M.
RESULTS
1. Objects viewed
The teacher remained the main focus in both views (View_P = 20.24%; View_M = 20.87%). However, the students were observed for less time in View_M (13.99%) than in View_P (26.88%), except for those interacting directly with the teacher. In addition, the off-task student and his mobile phone attracted less attention in View_M (2.83% and 0.25%) than in View_P (9.35% and 4.51%), despite being a central element of the excerpt .
2. Eye strategies
a) Duration of fixations
When viewing View_P, the pre-service teachers had a shorter average fixation duration (0.492 seconds) than in View_M (0.644 seconds). This may indicate a more dynamic visual exploration of the scene in View_P than in View_M. The results also suggest that View_M leads to a more uniform observation experience (var=1.35; standard deviation=0.872) compared to the View_P, as shown by the slightly lower scores (var=1.7; standard deviation=0.92).
b) Frequency of changes
In general, the pre-service teachers switched their gaze between views C and D (n=18 times). This can be explained by the fact that the teacher is in close-up for a long time and has a one-to-one conversation with a struggling student. Also, horizontal (n=37) and vertical (n=36) movements were preferred to oblique movements (n=7).
3. User experience
For the purposes of practical debriefing on professional teaching actions, the Vue_P is appreciated for its simplicity (2.44) and clarity (2.22), outperforming the Vue_M which, although considered somewhat original (0.5) and avant-garde (0.22), is less intuitive. The results for the View_P indicate greater consistency in pre-service teachers' evaluations than for the View_M.
DISCUSSION
Work from the cognitive theory of visual expertise (e.g. Keskin, Seidel, Stürmer, & Gegenfurtner, 2024; Lachner, Jarodzka, & Nückles, 2016) suggests that effective strategies for visual exploration of classroom scenes are particularly dynamic and student-centred. In our study, these tendencies are more present when the classroom scene is presented in the View_P than in the View_M view, which seems less conducive to dynamic visual exploration of the key elements of the excerpt and students' work. Furthermore, the View_M is less valued by pre-service teachers than the View_P view. In fact, despite its avant-garde nature, the View_M is perceived as less clear and less intuitive than the View_P. This perception is corroborated by a greater dispersion of scores for the View_M, suggesting a preference for the simplicity of the View_P.
CONCLUSIONS
The findings suggest guidelines for instructors on the importance of using clear video material, as suggested by the View_P. In addition, the study of trainers' visual exploration techniques on the same video clips is a promising avenue of research, as is the analysis of visual strategies used during debriefing using eye-tracking glasses.
