Abstract:

Previous studies have shown that the basic characteristics of children’ s eye movements during reading undergo substantial changes in early years, becoming adult-like around the age of 11: as chronological age increases, sentence reading times and fixation durations decrease, saccade amplitudes increase, fewer fixations and regressions are made, refixation probability decreases and word skipping probability increases. Then, a basic question is what accounts for the aforementioned changes with respect to the development of children’s eye movements characteristics. There are at least two possible answers: the first is “linguistic proficiency hypothesis” – that the development of reading skill in turn causes the observed changes in eye-movement behavior; the second is “oculomotor tuning hypothesis”- that readers are able to “tune” their oculomotor control system through learning so that the eye movements themselves become more optimal during reading. The present study wished to examine which hypothesis is more convincing; specifically we examined whether there were any developmental treads in terms of the influence of reading level on children’s basic eye movement characteristics. We tested amount of children’s reading level (approximately 200 children) in each grade. Based on their test scores, we selected three groups of children in each grade: the top 15 children were encoded as high-reading level group; the medium 15 children were encoded as medium-reading level group; the last 6-20 children were encoded as low- reading level group. Then, we instructed participants to read five age-appropriate texts when their eye movements were recorded. The results showed that, high-level groups of 9-year-old children in 3rd grade made significantly longer saccade amplitude, more forward saccades, and higher reading speed than the medium, and low groups of children, and no significant differences were found between the latter two groups of children for all the eye movement measures we adopted; for 10-year-old children in the 4th grade, high-level groups of participants made significantly less forward saccades and higher reading speed than those in low-group of participants, and no significant differences were found between the high- and medium- level groups of children; for 11-year-old children in the 5th grade, there were no reliable differences across three groups of children on all the eye movements measures. Based on the two hypothesis in terms of how children’s eye movements develop during reading, we argue that both linguistic proficiency and oculomotor tuning cause the development of children’s eye movement’s characteristics.

Key words: eye movement characteristics in reading, reading level, children, development.