Poster A68, Saturday, March 25, 5:00 – 7:00 pm, Pacific Concourse
Effects of Text Difficulty during Natural Reading: A co-registered eye tracking and fMRI study
Wonil Choi1, Matthew Lowder2, John Henderson2; 1GIST College, 2Center for Mind and Brain, University of California, Davis
We conducted a co-registered eye-tracking and fMRI experiment to examine how global text difficulty influences eye movements and neural activation during natural reading. Forty participants read forty paragraphs representing a wide range of difficulty levels (assessed using Flesch-Reading-Ease scores) in the scanner while their eye movements were recorded. Each paragraph was presented for twelve seconds. Eye tracking results showed robust relationships between text difficulty and online reading measures: As texts became more difficult, readers had longer fixation durations, shorter saccadic amplitudes, and slower reading rates (measured in words per minute). These results suggest that eye movement behaviors are sensitive to objective measures of global text difficulty. fMRI results demonstrated that there were strong negative correlations between text difficulty and activation in language-related areas such as bilateral superior temporal gyrus/sulcus, bilateral angular gyrus, right middle temporal gyrus, and right inferior frontal gyrus (pars triangularis), suggesting that deeper semantic networks are involved when readers process easy texts compared to difficult texts. In addition, activations in right superior frontal gyrus, right calcarine sulcus, left middle occipital gyrus, left lingual gyrus, and left amygdala were also negatively correlated with global text difficulty. However, global text difficulty was positively correlated with brain areas such as bilateral posterior cingulate gyrus, bilateral precuneus, and left dorsal angular gyrus, demonstrating that these regions were more active when the text was more difficult to understand. Overall, the results suggest that patterns of eye movements as well as brain activation are sensitive to changes in the global difficulty of texts.
Topic Area: LANGUAGE: Semantic