Are you with me? Measurement of Learners’ Video-Watching Attention with Eye Tracking

Srivastava, Namrata; Nawaz, Sadia; Newn, Joshua; Lodge, Jason M.; Velloso, Eduardo; Erfani, Sarah M.; Gašević, Dragan; Bailey, James E.

doi:10.1145/3448139.3448148

Cited by 21 publications

(10 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Tobii Pro X2-30 has a lower sampling frequency than the more professional Tobii Pro eye tracker (Tobii Pro Fusion, , accessed on 8 July 2021) or EyeLink 1000 (EyeLink 1000 Plus, , accessed on 8 July 2021). However, Tobii Pro X2-30 is being used in various studies for analyzing human gaze behavior [ 58 , 59 , 60 , 61 , 62 , 63 ]. Additionally, Tobii Pro SDK (Tobii Pro SDK, , accessed on 8 July 2021) and eye tracker manager software (Tobii Pro Eye Tracker Manager, , accessed on 8 July 2021) for professional eye-tracking research are provided.…”

Section: Methodsmentioning

confidence: 99%

Gaze Behavior Effect on Gaze Data Visualization at Different Abstraction Levels

Yoo

Jeong

Jang

2021

Sensors

View full text Add to dashboard Cite

Many gaze data visualization techniques intuitively show eye movement together with visual stimuli. The eye tracker records a large number of eye movements within a short period. Therefore, visualizing raw gaze data with the visual stimulus appears complicated and obscured, making it difficult to gain insight through visualization. To avoid the complication, we often employ fixation identification algorithms for more abstract visualizations. In the past, many scientists have focused on gaze data abstraction with the attention map and analyzed detail gaze movement patterns with the scanpath visualization. Abstract eye movement patterns change dramatically depending on fixation identification algorithms in the preprocessing. However, it is difficult to find out how fixation identification algorithms affect gaze movement pattern visualizations. Additionally, scientists often spend much time on adjusting parameters manually in the fixation identification algorithms. In this paper, we propose a gaze behavior-based data processing method for abstract gaze data visualization. The proposed method classifies raw gaze data using machine learning models for image classification, such as CNN, AlexNet, and LeNet. Additionally, we compare the velocity-based identification (I-VT), dispersion-based identification (I-DT), density-based fixation identification, velocity and dispersion-based (I-VDT), and machine learning based and behavior-based modelson various visualizations at each abstraction level, such as attention map, scanpath, and abstract gaze movement visualization.

show abstract

Section: Methodsmentioning

confidence: 99%

Gaze Behavior Effect on Gaze Data Visualization at Different Abstraction Levels

Yoo

Jeong

Jang

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…More analyses of the usage, perceptions and attitudes of students on an exclusively online learning environment during the pandemic (Chen et al, 2020;Coman et al, 2020;Popa et al, 2020) and attempts at capturing the students' attention level and monitoring it while they watch video or live lectures (Pham and Wang, 2015;Robal et al, 2018;Lee et al, 2021;Srivastava et al, 2021) are all very important, and the support for better guidance for students in more selectively/efficiently reviewing the video contents is a notable gap identified and addressed in this paper.…”

Section: Does the Proposed Visualization Of Weighted Playback By Peer...mentioning

confidence: 99%

Playback-centric visualizations of video usage using weighted interactions to guide where to watch in an educational context

et al. 2022

View full text Add to dashboard Cite

The steady increase in use of online educational tools and services has led to a large amount of educational video materials made available for students to watch. Finding the right video content is usually supported by the overarching learning management system and its user interface that organizes various video items by course, categories and weeks, and makes them searchable. However, once a wanted video is found, students are often left without further guidance as to what parts in that video they should focus on. In this article, an additional timeline visualization to augment the conventional playback timeline is introduced which employs a novel playback weighting strategy in which the history of different video interactions generate scores based on the context of each playback. This includes whether the playback started after jumping forward or backward in the video, whether the playback was at a faster or slower speed, and whether the playback window was in focus on the student's screen or was in the background. The resultant scores are presented on the additional timeline, making it in effect a playback-centric usage graph nuanced by how each playback was executed. Students are informed by this visualization on the playback by their peers and can selectively watch those portions which the contour of the usage visualization suggests. The visualization was implemented as a fully-fledged web application and deployed in an undergraduate course at a university for two full semesters. A total of 270 students used the system throughout both semesters watching 52 videos, guided by visualizations on what to watch. Analysis of playback logs revealed that students selectively watched portions in videos corresponding to the most important portions of the videos as assessed by the instructor who created the videos. The characteristics of this method as a way of guiding students as to where to watch as well as a complementary tool for playback analysis, are discussed. Further insights into the potential values of this visualization and its underlying playback weighting strategy are also discussed.

show abstract

“…Partly due to the inherent temporal nature of video medium that requires a viewer to "playback" in order to understand the contents, and partly due to the unedited and linear nature of lecture videos (unlike production videos featuring content-induced structure such as camera shots and scenes, chapters, and transitions), one consequence in the consumption of the video materials generated in this context is an increased burden to the students who face a large amount of unstructured lecture videos and screencasts at the time of reviewing: simply re-watching all video materials is not feasible, and yet there is no way to know what parts of the videos they should focus on. There are recent studies on tracking students' eye-gaze while watching a video lecture [4] or logging their level of attention during online classes [3], the knowledge of which could be used in suggesting the parts of videos that students should/could focus on in reviewing. This requires the capturing of the data at the time of watching/lecturing, an overhead on the front-end such as camera or other installed software to capture the eye-tracking/attention data.…”

Section: Pedagogical Backgroundmentioning

confidence: 99%

Usage-Based Summaries of Learning Videos

Lee

Liu

Scriney

et al. 2021

Technology-Enhanced Learning for a Free, Safe, and Sustainable World

View full text Add to dashboard Cite

Much of the delivery of University education is now by synchronous or asynchronous video. For students, one of the challenges is managing the sheer volume of such video material as video presentations of taught material are difficult to abbreviate and summarise because they do not have highlights which stand out. Apart from video bookmarks there are no tools available to determine which parts of video content should be replayed at revision time or just before examinations. We have developed and deployed a digital library for managing video learning material which has many dozens of hours of short-form video content from a range of taught courses for hundreds of students at undergraduate level. Through a web browser we allow students to access and play these videos and we log their anonymised playback usage. From these logs we score to each segment of each video based on the amount of playback it receives from across all students, whether the segment has been re-wound and re-played in the same student session, whether the on-screen window is the window in focus on the student's desktop/laptop, and speed of playback. We also incorporate negative scoring if a video segment is skipped or fast-forward, and overarching all this we include a decay function based on recency of playback, so the most recent days of playback contribute more to the video segment scores. For each video in the library we present a usage-based graph which allows students to see which parts of each video attract the most playback from their peers, which helps them select material at revision time. Usage of the system is fully anonymised and GDPR-compliant.

show abstract

Are you with me? Measurement of Learners’ Video-Watching Attention with Eye Tracking

Cited by 21 publications

References 22 publications

Gaze Behavior Effect on Gaze Data Visualization at Different Abstraction Levels

Gaze Behavior Effect on Gaze Data Visualization at Different Abstraction Levels

Playback-centric visualizations of video usage using weighted interactions to guide where to watch in an educational context

Usage-Based Summaries of Learning Videos

Contact Info

Product

Resources

About