A non-verbal teaching behaviour analysis for improving pointing out gestures: The case of asynchronous video lecture analysis using deep learning

Background: Research into enhancing the effectiveness of information delivery in asynchronous video lectures remains sparse. This study analyzes the nonverbal teaching behaviours in asynchronous online videos, drawing comparisons between pre-service and in-service teachers (ITs). Objectives: This research primarily aims to juxtapose the nonverbal teaching behaviours, such as arm extensions and body orientation, utilized by pre-service teachers (PTs) and ITs within asynchronous online videos. Methods: Asynchronous video lectures from four pre-service and four ITs across four diverse subject topics were scrutinized. Leveraging deep learning technology, teachers' poses during their instruction towards a video camera were quantified, with a particular focus on arm stretch range and body orientation in relation to the subject being taught. Results: The findings revealed that PTs were deficient in effectively employing pointing gestures. Their arm stretches and body orientation towards the board were not differentiated across subjects. Conversely, ITs demonstrated subject-specific variations in their arm extension and body orientation, signalling their effective strategies for knowledge dissemination. Conclusions and Discussion: This study emphasizes the importance of assessing nonverbal teaching behaviours in the development of effective instructional training. It accentuates the need for nonverbal communication and subject-specific teaching strategy training in PTs. Future investigations could broaden their scope to include larger sample sizes and expanded subject areas to discern more comprehensive trends in nonverbal teaching behaviours.