Eye Tracking on Text Reading with Visual Enhancements

TL;DR

This work investigates how visual enhancements in text—specifically highlights, icons, and word-sized graphics—alter reading behavior using two within-subject eye-tracking experiments with $N=12$ participants. Experiment 1 compares plain text, highlighting, and symbols, while Experiment 2 adds word-sized graphics to a plain baseline; both use short factual texts and fact-based questions to assess reading speed, accuracy, and gaze patterns. The results indicate that visual enhancements, especially highlights and word-sized graphics, modulate gaze and fixation patterns, with icons generally perceived as least distracting and associated with favorable subjective and performance signals, though many effects lack statistical significance due to small sample size. The authors discuss the potential of visual enhancements to aid memory and attention while acknowledging possible distraction and design challenges, and they propose hypotheses for more rigorous, larger-scale studies to quantify effects and optimize design choices for readability and information retention.

Abstract

The interplay between text and visualization is gaining importance for media where traditional text is enriched by visual elements to improve readability and emphasize facts. In two controlled eye-tracking experiments ($N=12$), we approach answers to the question: How do visualization techniques influence reading behavior? We compare plain text to that marked with highlights, icons, and word-sized data visualizations. We assess quantitative metrics~(eye movement, completion time, error rate) and subjective feedback~(personal preference and ratings). The results indicate that visualization techniques, especially in the first experiment, show promising trends for improved reading behavior. The results also show the need for further research to make reading more effective and inform suggestions for future studies.

Figures (8)

• Figure 1: Excerpts from stimuli for the experiment conditions highlight(top) and icons(bottom).
• Figure 2: Excerpts from the stimuli for the experiment condition word-sized graphics.
• Figure 3: Workflow of the experiments. $C_1i$ are selected (Latin square balanced) from the conditions baseline, highlight, and icon, and $C_2i$ from the baseline 2and word-sized-graphicsconditions.
• Figure 4: Reading time in seconds per stimulus, the time it took participants to answer the question for each stimulus, and answer accuracy in percent. Average with standard deviations (left chart, respectively). Pairwise differences between the study conditions (right chart, respectively), with 95% bootstrap confidence intervals and red lines indicating the Bonferroni corrections for 3 pairwise comparisons in the first experiment.
• Figure 5: Example of the AOIs we defined for stimuli for the experiment conditions (top) highlight, and (bottom) word-sized graphics.