| With the advent of the information age,learning through electronic screens has become the main way for us to acquire knowledge,and the resulting visual fatigue problem has been widely concerned.Watching videos and reading text are two main ways for people to acquire knowledge through electronic screens.Therefore,studying the process and characteristics of visual fatigue induced by watching video and reading text will be beneficial for people to scientifically and efficiently use electronic screens for learning,and has very important theoretical and practical significance.In this study,eye movement tracking technology was used to carry out experimental research.In Experiment 1,the characteristics of eye movement indicators that induced visual fatigue during video viewing and text reading over time were studied.The detected eye movement indicators mainly included average blink duration,average gaze duration,average eye hop amplitude,number of blinks,number of fixation points,average pupil area,heart rate,and subjective visual fatigue;Combining machine learning and multiple regression modeling methods to analyze the predictive effectiveness of various eye movement indicators in the formation of visual fatigue,in an attempt to find eye movement indicators that can more accurately predict visual fatigue.Experiment 2 further investigated the effects of two key factors,video playback speed and display brightness of reading text,on visual fatigue induced by electronic screens.The experimental research results show that:(1)In both experiments,subjects’ self-reported visual fatigue levels significantly increased over time,indicating that visual fatigue induced by electronic screen learning is a common phenomenon.(2)Experiment 1 analyzed various eye movement indicators and found that during video viewing and text reading tasks,the average blink duration,number of blinks,average gaze duration,number of gaze points,and average saccade amplitude all changed continuously over time,indicating that these indicators may be common eye movement indicators for detecting visual fatigue induced by the two screen learning methods;Since the average pupil area size only changes significantly over time during video viewing tasks,it may be a unique eye movement indicator for detecting visual fatigue induced by video viewing.(3)Experiment 1 combined machine learning and multiple regression model analysis results found that average fixation duration,number of fixation points,and average pupil area were key eye movement indicators for predicting visual fatigue induced by watching video;Average fixation duration,average blink duration,and number of blinks are key eye movement indicators for predicting visual fatigue induced by reading text.This indicates that there may be differences in the mechanism of visual fatigue induced by watching video and reading text.(4)In experiment 2,it was found that the average fixation duration of watching video at 2X magnification was significantly shorter than that of watching video at0.5X and 1X magnification;In the task of reading text,the average fixation duration of reading text at 20% display brightness was significantly shorter than that at 80%display brightness.This indicates that average gaze duration may be one of the key eye movement indicators for detecting visual fatigue,and may be affected by video playback speed and screen brightness.(5)Interestingly,in Experiment 2,there was a significant interaction between display brightness and time period on the eye movement index of blink count in reading text tasks.After comparison,it was found that under 20% display brightness,the number of blinks significantly increased during the time period(10-15min),while under 40% display brightness,the number of blinks significantly increased during the time period(35-40min).This indicates that different display brightness can affect the generation of visual fatigue at different times.In summary,this study used eye tracking technology to explore the eye movement characteristics of visual fatigue induced by different screen learning methods(watching video and reading text).Combining machine learning and multiple regression modeling methods,it was found that there are differences in key eye movement indicators that predict visual fatigue induced by the two screen learning methods,and they are affected by video playback speed and text screen brightness. |
PDF Full Text Request