December 2022
Volume 22, Issue 14
Open Access
Vision Sciences Society Annual Meeting Abstract  |   December 2022
Predicting reading speed based on eye movement features
Author Affiliations & Notes
  • Haojue Yu
    Northeastern University
  • Foroogh Shamsi
    Northeastern University
  • MiYoung Kwon
    Northeastern University
  • Footnotes
    Acknowledgements  This work was supported by NIH/NEI Grant R01 EY027857 and Research to Prevent Blindness (RPB) / Lions’Clubs International Foundation (LCIF) Low Vision Research Award.
Journal of Vision December 2022, Vol.22, 3071. doi:https://doi.org/10.1167/jov.22.14.3071
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Haojue Yu, Foroogh Shamsi, MiYoung Kwon; Predicting reading speed based on eye movement features. Journal of Vision 2022;22(14):3071. https://doi.org/10.1167/jov.22.14.3071.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

The environment we lived in sometimes poses challenging puzzles for our visual system. We all once tried to read the words where the letters seem faint; Or when we are outside on a foggy day, the signs everywhere seem so blurry and melt into the midst of the whiteness. This becomes more relevant to those with visual impairment. Thus, here we asked how degraded viewing conditions such as blur, low contrast, or dim light impact the way our eyes move from word to word during reading. To this end, we assessed eye movement patterns and reading speed under varying viewing conditions. We recruited 14 normally-sighted subjects and asked them to read a series of texts presented on the screen out loud. Text images were manipulated in the aspect of either background luminance (1.3 to 265 cd/m2), text blur (sigma=5 to no blur), or text contrast (2.6 to 100%) in four different levels respectively. For each subject, their reading time and eye movements were recorded with a high-speed eye-tracker. We found that, compared to the original text, for severely blurred text, fixation duration and the number of regressive saccades increased by 41% and 160% respectively, whereas saccade amplitude decreased by 25%; for the lowest text contrast, fixation duration and the number of regressive saccades increased by 68% and 210% respectively, whereas saccade amplitude decreased by 40%. On the other hand, no significant change in eye movements was found for the range of background luminance tested. Importantly, these three eye-movement features accounted for 84% of the variance in reading speed and allowed us to predict a person’s reading speed under varying viewing conditions. Our results show how the system optimizes eye-movement strategies when facing ambiguity and uncertainty. Our findings further offer great potential for predicting reading speed solely based on eye-movement patterns.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×