Abstract
The gradual emergence of online education in China in recent years requires new means of real-time monitoring and timely feedback to students. This research examines the effectiveness of synchronized eye movement assessment of attention engagement through two experiments. The first experiment used 24 university students in school as participants and made them watch the same video in high and low attentional engagement states (serial subtraction task) to compare the Inter-Subject Correlation (ISC) of participants' eye movements in different conditions. The results showed that the ISC of eye movements was significantly higher for participants in a high attentional engagement state than for participants in a low attentional engagement state. The second experiment had 26 university students in school as participants, as part of which they were made to watch video materials under the condition of having eye movement modeling examples. The results showed that the ISC of eye movements was significantly lower for participants in the group with eye movement modeling examples than those without eye movement modeling examples. However, overall test scores were significantly higher in the former than the latter. The first experiment showed that the eye movement trajectories of participants with high attentional engagement were more consistent than of those with low attentional engagement. Therefore, the ISC of participants' eye movements could be used as an objective indicator to assess and predict students' attentional conditions during online education. The second experiment showed that the eye movement modeling examples interfered with the participants' attention distribution to some extent; nevertheless, they positively affected the improvement in teaching effectiveness. Overall, the studies showed that the Inter-Subject Correlation is reliable to assess attentional engagement status in domestic online education.