Acer has today unveiled a research partnership with the University of Technology Sydney (UTS) aimed at using biometric data readily available from student devices to improve attention and engagement in class.
Acer and UTS devised the pilot as a way to understand what attracts or students’ attention and tailor class programmes to boost learning outcomes for both individuals or the whole class.
The Learner Attention Analytics Pilot Program takes data from device webcams used to track a student’s gaze, plus mouse movements and keyboard clicks and feeds it into machine learning algorithms to gauge how much attention a student is paying to their learning device.
While not everything in a uni tutorial or school classroom is based around digital devices, it still represents a significant (and growing) portion of activities, and certainly enough to give an idea of how distractions impact on learning outcomes.
Importantly, the pilot program passively collects data from devices students are already familiar with and using - rather than introducing another point of distraction.
The end goal of all of this is to give overworked teachers and lecturers an unobtrusive, easy to use tool that automatically flags a number of instances in which the class program could be tweaked, or when a student might be having external problems that impact on their studies, UTS’s executive director of Data Science professor Fang Chen said.
“Using learners’ behaviour as a fundamental indicator of attention and analysing this with AI and machine learning technologies will enable the education sector to optimise the pace and learning materials for the needs of different learners.”
While this could be to help a student going through family difficulties, Chen said an extreme example, if enough data is gathered over time, would be identifying possible correlations between students’ lunchtime activities and how they perform in class afterwards.
The content being accessed on devices will also be monitored, which will mainly focus on web traffic, but if keyboard data shows the only keys a student presses are the arrow keys or A, W, S and D, that’s another fairly strong indicator they’re not paying attention to coursework.
The program is currently being piloted with 200 data science students at UTS’ Faculty of Engineering and Information Technology, with a view to bring secondary and primary schools onboard in the near future.
Given that children will be involved in the program, Chen said the researchers are taking steps to mitigate privacy concerns while complying with school regulations and legal requirements.
Students who consent to participate in the trial will be assigned a random number, which will be the only identifier researchers will have. The schools will keep track of the numbers so teachers can identify which students are the ones who need help engaging with course material.
Additionally, Chen said the program is only focused on eye-tracking and won’t store images of participants’ entire faces.
Acer Oceania managing director Darren Simmons added that the analytics platform being developed is OS-agnostic to accommodate the popularity of different devices and operating systems in use across the education sector.
“In addition to education, it will also assist technology providers, such as Acer, to develop new computers and software applications and behaviour-aware computer technology to better facilitate the changing needs of the education sector.”
Over time, the learning analytics program could be expanded to offer insights into learner stress, frustration or hesitation.
The university said determining when and why this occurs is “an essential step towards customised teaching and learning and will be integral in improving student experience and wellbeing.