Comparing silent reading performance for single sentences and paragraphs: an eye movement-based analysis
Emmanouil Ktistakis,Angeliki Gleni,Miltiadis K Tsilimbaris,Sotiris Plainis
DOI: https://doi.org/10.1080/08164622.2023.2237974
2023-09-06
Clinical and Experimental Optometry
Abstract:CLINICAL RELEVANCE: Reading performance in clinical practice is commonly evaluated by reading 'aloud' and 'as fast as possible' single sentences. Assessing comprehensive silent reading performance using passages, composed of multiple sentences, is the preferred reading mode in real-life reading conditions.BACKGROUND: The purpose of this study was to compare eye movement-based silent reading performance for standardised short sentences and paragraphs.METHODS: A group of 15 young volunteers (age range: 22-36 years) read silently and comprehensively in two sessions: (a) a paragraph with continuous text and (b) standardised short sentences. Text print size was 0.4 logMAR (1.0 M at 40 cm distance). Eye movements during reading were recorded using video oculography (EyeLink II, SR Research Ltd). Data analysis included computation of reading speed, fixation duration, the number of fixations, saccadic amplitude and percentage of regressions. Moreover, frequency distributions of fixation durations were analysed with ex-Gaussian fittings.RESULTS: Repeatability coefficient in silent reading speed was found better for the paragraph (66 wpm) than for short sentences (88 wpm). The superiority in repeatability coefficient for the corresponding eye movement parameters, i.e. fixation duration (35 vs 73 ms), regressions (10.1 vs. 22.3%) and fixations per word (0.21 vs. 0.37 fpw), was even more pronounced. In addition, a statistically significant improvement with the paragraph was found in average fixation duration (19 ± 26 ms, p = 0.02), regressions (4.2 ± 7.0%, p = 0.04) and ex-Gaussian fixation parameter, τ (82 vs. 111 ms). No statistically significant difference was found between average reading speed with the paragraph (220 ± 59 wpm) and the short sentences (206 ± 57 wpm) (p = 0.11).DISCUSSION: Due to their superior repeatability, paragraphs are preferable to short sentences when evaluating silent comprehensive reading. The concurrent recording of eye movement parameters in silent reading further improves variability and could offer an efficient measure of reading performance and a reliable biomarker of visuo-motor function.
ophthalmology