![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
Abstract
Recently, there has been considerable debate about whether readers can identify multiple words in parallel or whether they are limited to a serial mode of word identification, processing one word at a time (see, e.g., Reichle, Liversedge, Pollatsek, & Rayner, 2009). Similar questions can be applied to bimorphemic compound words: Do readers identify all the constituents of a compound word in parallel, and does it matter which of the morphemes is identified first? We asked subjects to read compound words embedded in sentences while monitoring their eye movements. Using the boundary paradigm (Rayner, 1975), we manipulated the preview that subjects received of the compound word before they fixated it. In particular, the morpheme order of the preview was either normal (cowboy) or reversed (boycow). Additionally, we manipulated the preview availability for each of the morphemes separately. Preview was thus available for the first morpheme only (cowtxg), for the second morpheme only (enzboy), or for neither of the morphemes (enztxg). We report three major findings: First, there was an effect of morpheme order on gaze durations measured on the compound word, indicating that, as expected, readers obtained a greater preview benefit when the preview presented the morphemes in the correct order than when their order was reversed. Second, gaze durations on the compound word were influenced not only by preview availability for the first, but also by that for the second morpheme. Finally, and most importantly, the results show that readers are able to extract some morpheme information even from a reverse order preview. In summary, readers obtain preview benefit from both constituents of a short compound word, even when the preview does not reflect the correct morpheme order.
Acknowledgments
This research was supported by Grant HD26765 from the National Institutes of Health. Portions of the data were reported at the CUNY (City University of New York) Sentence Processing Conference in March 2010 in New York City and the 4th China International Conference on Eye Movements in May 2010 in Tianjin, China. We thank Raymond Bertram, Marc Brysbaert, Albrecht Inhoff, and Wayne Murray for their helpful comments on an earlier version of this manuscript.
Notes
1 Including those trials in which the pretarget word was skipped during first-pass reading did not lead to a change in the pattern of results.
2 This high rate of data loss was partly because there is considerable jitter at the end of a saccade, and the eyes typically have not settled into place by this time. For both experiments, we also performed a more lenient analysis, which included data when the display change was completed up to 9 ms after the saccade ended. This resulted in the elimination of only 3.4% of the data in Experiment 1. Importantly, this analysis revealed exactly the same pattern of results as the stricter method, and all of the effects that were significant in the strict analysis were significant in the more lenient method. The only exception was a significant spillover effect of preview morpheme order on log-transformed go-past durations on the posttarget word in the more lenient analysis (b = 0.06, SE = 0.032, t = 1.99).
3 Again, due to the high rate of data loss, we also performed an analysis with the more lenient criterion, which included data when the display change was completed up to 9 ms after the saccade ended and resulted in a data loss of only 9.8%. Again, the pattern of results was identical for both criteria. The only exception was the analysis of log-transformed gaze durations on the target word in Experiment 2, which, under the more lenient data exclusion rules, yielded a significant effect of Morpheme 1 preview availability even in the transposed morpheme preview order condition (b = 0.03, SE = 0.014, t = 2.02).
4 As in Experiment 1, including those trials in which the pretarget word was skipped during first-pass reading did not lead to a major change in the pattern of results.
5 For each factor, the level that represented correct morpheme order preview or correct morpheme letter preview, respectively, was coded as –1, while the level that represented transposed morpheme order preview and incorrect morpheme letter preview was coded as 1. The three-way interactions did not reach significance in any analysis except for log-transformed first-fixation durations (b = 0.08, SE = 0.04, t = 2.08). We fitted separate models for the correct preview order and the transposed preview order condition in order to explore this interaction further and found that the interaction between Morpheme 1 letter identity and Morpheme 2 letter identity preview was significant only when the preview morpheme order had been correct (b = –0.054, SE = 0.026, t = –2.08). However, since this interaction was far from significance in the analysis using raw first-fixation duration and is not visible in the raw fixation time means, it is unclear whether it can be interpreted.
6 In addition to the analyses reported here, we performed two supplementary analyses. In the first analysis, we included semantic transparency (Libben, Gibson, Yoon, & Sandra, Citation2003) as a predictor. Specifically, we tested for effects of transparency of the first (nonhead) or second (head) morpheme as well as the overall effect of transparency of the entire compound word. Since none of these predictors showed significant effects, we do not report the results from this analysis. In the second analysis, we treated the first and the second morpheme of the compound word as separate analysis regions in order to investigate whether the effects found in the main analysis were present throughout the word or only in fixations on one of the morphemes. We did not find a consistent pattern in this analysis, although there was a slight trend of preview order effects being stronger on the first morpheme and preview letter identity effects being stronger on the second morpheme.
7 These analyses included all predictors present in the full analyses with the exception of morpheme preview order and the interaction between Morpheme 1 and Morpheme 2 letter preview, as this term did not reach significance in any measure.
8 There was no significant effect of morpheme order on landing position. There was a significant interaction between morpheme order and Morpheme 1 preview: When morpheme order was correct, the Morpheme 1 preview denied condition led to landing positions further towards the beginning of the word. This effect was not present in the reverse preview order condition.