Word frequency, predictability, and return-sweep saccades: Towards the modeling of eye movements during paragraph reading.

Abstract

Models of eye movement control during reading focus on the reading of single lines of text. Within these models, word frequency and predictability are important input variables which influence fixation probabilities and durations. However, a comprehensive model of eye movement control will have to account for readers' eye movements across multi-line texts. Line-initial words are unlike those presented mid-sentence; they are routinely unavailable for parafoveal pre-processing. Therefore, it is unclear if and how word frequency and predictability influence reading times on line-initial words. To address this, we present an analysis of the Provo Corpus (Luke & Christianson, 2017) followed by a novel eyemovement experiment. We conclude that word frequency and predictability impact singlefixation and gaze durations on line-initial words. We also observed that return-sweep error (undersweep-fixations) may, among several other possibilities, allow for parafoveal processing of line-initial words prior to their direct fixation. Implications for models of eye movement control during reading are discussed.

Spelling ability influences early letter encoding during reading: Evidence from return-sweep eye movements

“…That is not to say readers do not acquire information during these fixations. Slattery and Parker (2019), Parker and Slattery (2019), and Parker et al (2020) reported that readers can extract information at the point of, and to the left of, undersweep fixations. However, these fixations appear to be terminated based on retinal feedback following the return-sweep which, if exceeding a certain threshold, triggers a corrective saccade.…”

“…As we predicted, accurate line-initial fixations significantly decreased with increasing reading ability consistent with the importance of foveal word processing during these fixations (c.f. Parker & Slattery, 2019). We also observed an effect of spelling ability on accurate line-initial fixations, whereby durations significantly decreased with increasing spelling ability.…”

“…Finally, undersweep fixations typically result from oculomotor error with longer lines yielding more corrective saccades (Heller, 1982; Hofmeister et al, 1999; Parker, Nikolova, et al, 2019). While their duration is uninfluenced by lexical properties of the fixated word (Slattery & Parker, 2019), there is now converging evidence to suggest that readers are able to extract information at the point of the undersweep fixation (Parker et al, 2020; Slattery & Parker, 2019) and to the left of undersweep fixations (i.e., the line-initial word; Parker et al, 2020; Parker & Slattery, 2019). Given the hypothesised differential involvement of lexical processing for each population of fixation, there may be differences in the extent to which lexical quality influences the duration of each fixation population.…”

What is the optimal position of low‐frequency words across line boundaries? An eye movement investigation

“…Inhoff & Rayner, 1986). This is not surprising given the moderate to strong correlations reported in corpus‐based analyses between word length and frequency (Kliegl et al, 2004; Parker & Slattery, 2019). This means that current approaches to text wrapping may result in words that are more difficult to process being positioned at the very start of the line.…”

“…This results in longer line‐initial fixations. If, as observed in Parker and Slattery's (2019) analysis of the Provo Corpus, lexical effects are stronger at the start of the line, positioning long, low‐frequency words at the very start of the line may result in longer reading times when moving between lines. Over multiple lines of text, this has the potential to result in slower passage reading.…”