Word frequency and text complexity: an eye-tracking study of young Russian readers

Although word frequency is often associated with the cognitive load on reader and widely used for automated text complexity assessment, to date, no eye-tracking data have been obtained effectiveness of this parameter prediction Russian primary school readers. Besides, optimal ways taking into account individual words assess an entire not yet precisely determined. This article aims fill these gaps. The study was conducted a sample 53 children age. As stimulus material, we 6 texts that differ in classical Flesch readability formula texts. sources data, common dictionary based material National Corpus DetCorpus - corpus literature addressed children. speed reading aloud per minute averaged over grades employed as measure complexity. best predictive results relative time were using lemma from DetCorpus. At level, highest correlation shown by coverage list 5,000 most frequent words, while both lists showed almost same values. For more detailed analysis, also calculated parameters specific forms lemmas three oculomotor activity: dwell time, fixations count, average duration fixations. word-by-word demonstrated time. confirm feasibility assessment task demonstrate calculate data.