Changes in Ocular Fixation Characteristics Over Time during Reading

Abstract

In this report, we evaluate eye-movements during reading. There is a huge literature on this topic, but our report is not focused on the typical questions raised in this literature and our task design is very atypical for an eye-movement/reading study. While most reading studies evaluate mental processes during reading, the only mental process we evaluate is fatigue. While most reading studies use stimuli presented in the middle of the screen, our poem sections are presented in 24 lines of text that span the top to the bottom of the display. While most multi-line reading studies use a very large interline spacing, our interline spacing is more typical of text reading in newspapers, magazines and books. We report on changes in the characteristics of fixations as a function of time-on-task (TOT). We determine the start and end of reading for each subject/session and divide this reading time into 10 equal length (number of samples) periods (referred to as epochs). We looked for changes in fixation characteristics across epochs. We emphasize our results for horizontal position because these changes were monotonic and interpretable. For horizontal position signals, we found that the mean intersample distances in fixations increases, the rate of changes of fixation position decreases and the total fixation width increases as a function of epoch. We interpret our results to mean that, as reading progresses, it becomes more difficult to hold the eye still. Early in reading, the total fixation width is lower, the mean intersample distance is lower and there are frequent adjustments (i.e., changes in direction) to keep the eye well focused on the target word. As time progresses, it becomes more and more difficult to hold the eye perfectly steady, so fixations become wider and mean intersample distance increases.