Parafoveal–foveal overlap can facilitate ongoing word identification during reading: Evidence from eye movements.

Angele, Bernhard; Tran, Randy; Rayner, Keith

doi:10.1037/a0029492

Cited by 50 publications

(87 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specifically, the magnitude of the 20 preview benefit was larger when the critical distractor was identical to the target (repeated 21 condition) than when the distractor was a frequency-matched control word that was unrelated to 22 the target (control condition). This effect is reminiscent of a repetition effect observed by 23 34 Angele, Tran, and Rayner (2012). In a boundary change experiment during reading, they 1 observed that the repetition of the foveal word in the parafovea (after the boundary) resulted in 2 shorter fixation durations on the foveal word and no evidence of additional disruption on the 3 parafoveal word when it was fixated after the boundary change.…”

mentioning

confidence: 76%

Attention and eye-movement control in reading: The selective reading paradigm.

Reingold

Sheridan

Meadmore

et al. 2016

Journal of Experimental Psychology: Human Perception and Perfor

View full text Add to dashboard Cite

1We introduced a novel paradigm for investigating covert attention and eye-movement control in 2 reading. In 2 experiments, participants read sentence words (shown in blue color) while ignoring 3 interleaved distractor strings (shown in orange color). Each single-line text display contained a 4 target word and a critical distractor. Critical distractors were located just prior to the target in the 5 text and were either words or symbol strings (e.g., @#%&). Target word availability for 6 parafoveal processing (i.e., preview validity) was also manipulated. The results indicated much 7 shallower processing of distractors than targets and this pattern was more pronounced for symbol 8 than word distractors. The influences of word frequency and fixation location on first-pass 9 fixation durations on distractors were dramatically different than the well-documented pattern 10 obtained in normal reading. Robust preview benefits were demonstrated both when the critical 11 distractors were fixated and when the critical distractors were skipped. Finally, with the 12 exception of larger preview benefits that were obtained in the condition in which the target and 13 critical distractor were identical, the magnitude of the preview effect was largely unaffected by 14 the nature of the critical distractor. Implications of the present paradigm and findings to the study 15 of eye-movement control in reading are discussed. 16 3 The use of eye movements to study reading has a rich history dating back over a century 1 (see Rayner, 1998Rayner, , 2009, for reviews). Readers move their eyes primarily forward in the text to 2 encounter new words. The magnitude of the typical forward movement (saccade) for readers of 3English is about 7-9 letter spaces. High-velocity saccadic eye movements, during which vision is 4 largely suppressed (Matin, 1974), occur at an average rate of 3-4 per second and are separated by 5 periods during which the eyes remain relatively still (fixations), and perceptual information is 6 extracted. Saccades are required in order to align the high-acuity foveal region of the retina (the 7 central 2° of vision) with the part of the text that is being encoded by the reader. However, 8 during each fixation, while the fixated word (word N) is primarily encoded using foveal vision, 9 parafoveal vision is used to extract perceptual information from at least the next two words in the 10 text (word N+1 and word N+2). Specifically, for readers of English the encoding of useful 11 perceptual information is confined to the perceptual span, an asymmetric region of the text 12 around the fixation point, which is limited to about 3-4 letter spaces to the left and 14-15 letter 13 spaces to the right of fixation (McConkie & Rayner, 1975). Importantly, when the area of useful 14 orthographic information during a fixation is restricted to word N, word N+1 and word N+2, 15 reading rate is approximately normal, while reading rate decreases by about 10% when only 16Word N and Word N+1 information is available and by over 30% wh...

show abstract

mentioning

confidence: 76%

Attention and eye-movement control in reading: The selective reading paradigm.

Reingold

Sheridan

Meadmore

et al. 2016

Journal of Experimental Psychology: Human Perception and Perfor

View full text Add to dashboard Cite

show abstract

“…As discussed in depth earlier, it could be assumed that orthographic parafoveal-on-foveal effects are due to orthographic information from the parafoveal word either facilitating (e.g. Angele et al, 2013) or inhibiting (e.g. ) the processing of a spatially adjacent word.…”

Section: Discussionmentioning

confidence: 99%

“…Angele, Tran, and Rayner (2013) examined fixation durations on a target word (e.g. news) while using the boundary paradigm to manipulate the parafoveal word to be a repetition of the target (e.g.…”

Section: Crypt)mentioning

confidence: 99%

“…Rather, our manipulation involved participants always viewing two orthographically legal and distinct words (as opposed to word n+1 being identical to word n as in Angele et al, 2013) in the parafovea. Thus, if we did observe a benefit of the transposition condition relative to the substitution condition, it could only be due to processing of word n+1 and word n+2 occurring in parallel.…”

Section: The Current Studymentioning

confidence: 99%

See 1 more Smart Citation

Is orthographic information from multiple parafoveal words processed in parallel: An eye-tracking study.

Cutter

Drieghe

Liversedge

2017

Journal of Experimental Psychology: Human Perception and Perfor

View full text Add to dashboard Cite

In the current study we investigated whether orthographic information available from 1 upcoming parafoveal word influences the processing of another parafoveal word. Across 2 experiments we used the boundary paradigm (Rayner, 1975) to present participants with an identity preview of the 2 words after the boundary (e.g., hot pan), a preview in which 2 letters were transposed between these words (e.g., hop tan), or a preview in which the same 2 letters were substituted (e.g., hob fan). We hypothesized that if these 2 words were processed in parallel in the parafovea then we may observe significant preview benefits for the condition in which the letters were transposed between words relative to the condition in which the letters were substituted. However, no such effect was observed, with participants fixating the words for the same amount of time in both conditions. This was the case both when the transposition was made between the final and first letter of the 2 words (e.g., hop tan as a preview of hot pan; Experiment 1) and when the transposition maintained within word letter position (e.g., pit hop as a preview of hit pop; Experiment 2). The implications of these findings are considered in relation to serial and parallel lexical processing during reading. Public Significance StatementThe findings of the present study suggest that readers of English only process letter information from 1 word at a time. Participants did not read any faster when 2 letters had been swapped between 2 upcoming words prior to participants looking directly at them (e.g., sit pea instead of sip tea) than when these same 2 letters had been replaced with alternative letters (e.g., six sea). These findings demonstrate that the human language processing system tends to identify words 1 at a time in serial order, rather than identifying multiple words at the same time.

show abstract

“…In all of the models, we also included random intercepts for subject, and for strategy-type (except in Experiment 3, where we did not code the strategy-types), and we included random slopes (per subject) for all within-subject predictors. Because it is difficult to calculate meaningful p-values in LMM (as the degrees of freedom in each analysis are unclear), we follow the common convention of omitting pvalues for these LMM analyses, instead treating |t| > 1.96 as corresponding to a conventional α = .05 level of statistical significance (for an example, see Angele, Tran, & Rayner, 2013).…”

Section: [Figure 1 About Here]mentioning

confidence: 99%