Benchmark measures of eye movements during reading in Russian children

Journal of Experimental Psychology: Learning Memory and Cognition. 2022.

Lopukhina A., Zdorova N., Loshchinina V., Ladinskaya N., Kaprielova A., Goldina S., Vedenina O., Bartseva K., Dragoy O.

In this study, we investigated eye movements during silent reading in 222 typically developing Russian children from grades 1 through 6. First, we established eye-movement benchmarks and detected two periods (between grades 1-2 and grades 3-4) when reading development was the fastest. We compared the basic eye-movement measures in children to those in adults and concluded that neither reached the adults’ level, with the most prominent difference in the probability of skipping and the number of fixations within a word. Second, we investigated the lexical effects of word length, frequency, predictability, and word class, as well as length-by-frequency and frequency-by-predictability interactions on eye movements in children across grades. The results revealed that long, low-frequency, and low-predictability words are read slower and the effect sizes of these lexical features decreased with grade. The effect of word length was larger for high-frequency words, which is opposite to German and Finnish. Also, the facilitative effect of predictability was more prominent for low-frequency words, which is similar to English and opposite to Chinese. Additionally, we found that children read nouns and adjectives faster than other word classes. Third, we tested whether individual differences in phonological processing correlated with eye-movement measures and whether the effect changed with grade. We found that children with better complex phonological skills fixated on words for a shorter period of time. The magnitude of the effect did not change much with grade. The child version of the Russian Sentence Corpus (ChiRSC) is the first systematic description of eye movements during reading in Cyrillic. The annotated ChiRSC is freely

Keywords: Russian, eye tracking, reading development, cross-sectional corpus study