Eye movement control in reading: word predictability has little influence on initial landing positions in words

Rayner, Keith; Binder, Katherine S.; Ashby, Jane; Pollatsek, Alexander

doi:10.1016/s0042-6989(00)00310-2

Cited by 100 publications

(120 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In our experiments we clearly show that the saccade targeting system is not overly sensitive to even a large difference in initial bigram frequency of upcoming words, when these initial bigrams are of relatively high, or extremely high, frequency. Coupled with previous findings showing no evidence for modulation of saccade length or initial fixation location because of target words' predictability (e.g., Rayner et al, 2001), or only effects of modest size for initial letter sequence and morphological properties (e.g., Hyönä, 1995;White & Liversedge, 2004;2006;Yan et al, 2014), we may conclude that the saccade targeting system is minimally influenced by the linguistic properties of the upcoming words.…”

Section: Ge Ne Ralndiscussionsupporting

confidence: 87%

The influence of a word’s number of letters, spatial extent, and initial bigram characteristics on eye movement control during reading: Evidence from Arabic.

Hermena

Liversedge

Drieghe

2017

Journal of Experimental Psychology: Learning, Memory, and Cogni

View full text Add to dashboard Cite

The authors conducted 2 eye movement experiments in which they used the typographical and linguistic properties of Arabic to disentangle the influences of words’ number of letters and spatial extent on measures of fixation duration and saccade targeting (Experiment 1), and to investigate the influence of initial bigram characteristics on saccade targeting during reading (Experiment 2). In the first experiment, through the use of a proportional font, which is more natural-looking in Arabic compared to monospaced fonts, the authors manipulated the number of letters (5 vs. 7) and the spatial extent (wide vs. narrow) of words embedded in frame sentences. The results obtained replicate and expand upon previous findings in other alphabetic languages that the number of letters influences fixation durations, whereas saccade targeting (as indicated by measures of fixation count and probability of skipping and refixation) is more influenced by the word’s spatial extent. In the second experiment, the authors compared saccade targeting measures (saccade amplitude and initial fixation location) in 6- and 7-letter words beginning with initial bigrams that were of extremely high frequency (ال the), relatively high frequency (لل to/for the), or beginning with the letters of the word stem. The results showed negligible modulation of saccade targeting by initial bigram characteristics. The results also highlighted the importance of selecting the appropriate measures of initial fixation location (spatial vs. character-based measures) during reading text rendered using proportional fonts.

show abstract

Section: Ge Ne Ralndiscussionsupporting

confidence: 87%

The influence of a word’s number of letters, spatial extent, and initial bigram characteristics on eye movement control during reading: Evidence from Arabic.

Hermena

Liversedge

Drieghe

2017

Journal of Experimental Psychology: Learning, Memory, and Cogni

View full text Add to dashboard Cite

show abstract

“…The predictions of E-Z Reader were very close those values: 37% of predictable words and 25% of unpredictable words were skipped. The difference between predictable and unpredictable words (9% in our study and 12% for E-Z Reader) is quite consistent with prior research (Rayner et al 2001;Rayner & Well, 1996). However, inconsistency was observed in the pattern of results for fixation times (Table 1).…”

supporting

confidence: 91%

“…E-Z Reader did not predict the effect of predictability on first fixation duration and single fixation. With respect to gaze duration, the effect predicted by E-Z Reader was comparable in size to the effect obtained in other studies (Rayner et al 2001;Rayner & Well 1996), though it was lower than in our study. A closer examination of the data indicated that the prediction of E-Z -rReader was inconsistent for high frequency words: Fixation duration was longer for predictable than for unpredictable words (Table 2).…”

supporting

confidence: 80%

The effects of frequency and predictability on eye fixations in reading: An evaluation of the E-Z Reader model

2003

View full text Add to dashboard Cite

We tested whether the E-Z Reader model can be generalised to French language. The simulation showed that the model can account for the frequency effect. The predictability effect is moreover accurate for words skipping but not for fixation times. We think that this model is psychologically plausible for certain aspects of reading and we used it to evaluate the performance of dyslexic readers.The E-Z Reader 7 model was applied to data that were obtained in an eye-tracking experiment. The purpose of the study was twofold. Its first aim was to test whether the model can be generalised to French language. The second objective was to test whether the model can satisfy to contextual constraint influencing reading, namely predictability. Fifteen participants were requested to read a French text that contained 134 words varying in frequency and predictability (the latter factor was determined in a previous task by asking participants to guess word n+1 when provided with the sentence up to word n). As in Reichle et al., the words were divided into five frequency classes and eye movement was recorded.. Furthermore, we ran a simulation to determine how well the model could predict the observed data (i.e., means duration and distributions of fixations). To be in agreement with E-Z Reader 7, the first and last words of each sentence were not included in the data analysis.As shown in Figure 1, we obtained a satisfactory matching between observed and predicted values for all dependent variables. E-Z Reader 7 correctly predicted the negative monotonic relationship between first fixation duration and word frequency. However, the predicted values were slightly smaller than the observed values for gaze duration. But gaze duration, which includes refixation, is considered a global indicator of word processing as it is influenced by postlexical processing, although first fixation is supposed to be more sensitive to low-level word processing of word. Contrary to the reference experiment previously used to test the model (Schilling et al., 1998), a crucial aspect of our study was that subjects read a meaningful text, which suggests that postlexical processing were more deeply probed. In the Schilling et al. study, subjects read sentences that tended to be short, stereotyped expressions that were very simple to comprehend; and even though reading was involved with this paradigm, it did not obviously draw on the kinds of complex comprehension processes that are part of full-blown reading. However, it is noteworthy that the E-Z Rreader model provides a theoretical framework for understanding the interface between vision and low-level aspects of language processing.

show abstract

“…Reaction times as well as eye movement data point to faster processing of high than of low predictability words (e.g., Ashby, Rayner, & Clifton, 2005;Calvo & Meseguer;Duffy, Henderson, & Morris, 1989;Ehrlich & Rayner, 1981;Fischler & Bloom, 1979;Kleiman, 1980;Kliegl et al, 2004Kliegl et al, , 2006Rayner, Ashby, Pollatsek, & Reichle, 2004;Rayner, Binder, Ashby, & Pollatsek, 2001;Rayner & Well, 1996;Schuberth & Eimas, 1977;Stanovich & West, 1983;West & Stanovich, 1982). The critical question, however, is when does top-down expectation of a stimulus interact with the incoming visual information during reading and, depending on whether the prediction was correct or wrong, when does it help or hurt word processing?…”

Section: Predictability: Top-down Expectationsmentioning

confidence: 99%

Stimulus onset asynchrony and the timeline of word recognition: Event-related potentials during sentence reading

et al. 2012

View full text Add to dashboard Cite

Article in press in Neuropsychologia [http://dx.doi.org/10.1016/j.neuropsychologia.2012.04.011] Please note that this unproofed preprint may be subject to minor changes. [May 5, 2012] SOA and the Timeline of Word Recognition 2 AbstractThree ERP experiments examined the effect of word presentation rate (i.e., stimulus onset asynchrony, SOA) on the time course of word frequency and predictability effects in sentence reading. In Experiments 1 and 2, sentences were presented word-by-word in the screen center at an SOA of 700 and 490 ms, respectively. While these rates are typical for psycholinguistic ERP research, natural reading happens at a considerably faster pace. Accordingly, Experiment 3 employed a near-normal SOA of 280 ms, which approximated the rate of normal reading. Main results can be summarized as follows: (1) The onset latency of early frequency effects decreases gradually with increasing presentation rates. (2) An early interaction between top-down and bottom-up processing is observed only under a nearnormal SOA. (3) N400 predictability effects occur later and are smaller at a near-normal (i.e., high) presentation rate than at the lower rates commonly used in ERP experiments. (4) ERP morphology is different at the shortest compared to longer SOAs. Together, the results point to a special role of a near-normal presentation rate for visual word recognition and therefore suggest that SOA should be taken into account in research of natural reading.

show abstract

Eye movement control in reading: word predictability has little influence on initial landing positions in words

Cited by 100 publications

References 34 publications

The influence of a word’s number of letters, spatial extent, and initial bigram characteristics on eye movement control during reading: Evidence from Arabic.

The influence of a word’s number of letters, spatial extent, and initial bigram characteristics on eye movement control during reading: Evidence from Arabic.

The effects of frequency and predictability on eye fixations in reading: An evaluation of the E-Z Reader model

Stimulus onset asynchrony and the timeline of word recognition: Event-related potentials during sentence reading

Contact Info

Product

Resources

About