Feasibility of Eye Movement Recordings with the SMI Tracking Bar in 10- to 11-Year-Old Children Performing a Reading Task.

ABSTRACT

INTRODUCTION: Eye movements during reading can be impaired in amblyopia, developmental dyslexia, reduced visual acuity, or visual field defects. To detect pathology, normative values are important for comparison. In healthy children, there is sparse data on eye movements during reading. Therefore, the aim of this study was to, in a first step, explore the feasibility of applying the SMI RED eye tracker bar to record eye movements in 10- and 11-year-old children while reading a text. MATERIALS AND METHODS: Thirty-three (19 aged 10 years, 14 aged 11 years) normally sighted children attending a primary school in Switzerland participated in our study. Visual acuity, the Lang test, and the cover test were performed as a screening for ophthalmologic pathology that might influence the results. Eye movements were recorded with the SMI RED eye tracker bar while the child read aloud two texts from the International Reading Speed Test (IReST), presented on a laptop. Both texts were in German with an equal level of difficulty and were presented in a randomized order. Reading speed (words/minute), number of saccades, number of fixations, and reading errors (mistakes in the reading) were evaluated. RESULTS: Screening did not reveal pathology other than refractive errors and children had full corrected visual acuity. Eye movements could be obtained in all but six children where the reflection of the glasses worn prevented a good pupil recording with the tracker. Younger children performed more saccades per word with a mean of 1.41 (SD 0.39) at 10 years of age versus 1.10 (SD 0.21) at 11 years of age. The number of fixations per word was also higher in younger children (mean 1.63 [SD 0.37]) than in 11-year-old children (mean 1.32 [SD 0.33]). Ten-year-old children seem to analyze a text in smaller units than 11-year-olds. Thus, 10-year-old children took more time to complete the reading task than the 11-year-olds (mean 88.8 s [SD 24.1] versus 84.4 s [SD 15.1]). In addition, 10-year-old children made more reading errors compared to 11-year-olds (mean 4.47 [SD 2.95] versus 2.28 [SD 1.72]). CONCLUSION: It is feasible to record eye movements in children aged 10 - 11, albeit this is more difficult when glasses are worn. As parameters change with age, further data is needed for a representative evaluation regarding eye movements during reading in children of different age groups. The information gained may offer help in recognizing reading difficulties and monitoring of treatment effects.