The interrelation of print exposure with patterns of oculomotor activity and text comprehension

The current study aims to investigate how individual differences in print exposure affect text processing. Print exposure represents the amount of time one devotes to reading. It is an essential measure for psycholinguistic research on language processing: it has been shown that exposure to print can affect reading comprehension (Martin-Chang & Gould, 2008; Mar & Rain, 2015), reading speed (Martin-Chang & Gould, 2008; Mano & Guerin, 2018), and some eye movement patterns during the reading (Choi et al., 2015, Slattery & Yates, 2018). We focus on two (fiction and non-fiction) text types and investigate how print exposure affects both oculomotor reading behavior and comprehension of coherent fiction and non-fiction texts in Russian. 40 native speakers of the Russian aged 18 to 28 years volunteered to take part in the study. All the participants had higher education or were university students, all had normal or corrected-to-normal vision. They were asked to read six short texts in Russian: three fiction texts were short stories by Felix Krivin, and three non-fiction texts were Wikipedia-style paragraphs taken from the materials of the Multilingual Eye-Movement Corpus (Siegelmann et al., 2022). Eye movements during reading were recorded using EyeLink 1000+ desktop mount eyetracker with a chin rest. Comprehension questions were asked after each text. To assess fiction texts comprehension we also used the keyword extraction task and text summarization task. As for individual differences in print exposure, the most effective methodology to assess them is the Author Recognition Test, or ART (Stanovich & West, 1989). This test was adapted for different languages, in our study we use yhe Russian version of ART (Chernova & Bakhturina, 2021). We modeled the eye-movement data with a mixed-effects regression, including random intercepts and random slopes by a participant and by a text in the model. We show that ART-score, which reflects individual differences in print exposure, significantly affects saccade amplitude (b=0.052, SE=0.015, t=3.416; p=0.002), and average regression count per word (b=-0.021,SE=0.010, t=-2.068; p=0.046). Print exposure affects eye-movements both in reading fiction texts (b=0.016, SE=0.006, t=2.5; p=0.018 for saccade amplitude) and non-fiction texts (b=0.019, SE=0.007, t=2.7; p=0.012 for saccade amplitude) taken separately. No significant effect of print exposure on text comprehension was found. Print exposure, i.e. previous experience in reading, significantly affects reading fluency: the more experienced a reader is, the larger saccades and the fewer fixations s/he makes. It should be noticed that ART measures print exposure for fiction, and the effect on eye movements is pronounced both for fiction and nonfiction texts. No evidence is found for the effect of print exposure on text comprehension, but interrelation between comprehension scores and regression rates can give evidence for interrelation between online and offline text processing. The authors declare no conflicts of interests.

Keywords

Authors

References