Parafoveal processing of orthographic, morphological, and semantic information during reading Arabic: A boundary paradigm investigation.

Ehab W Hermena,Eida J Juma,Maryam Aljassmi,Nicola Molinaro

doi:10.1371/journal.pone.0254745

Abstract

Evidence shows that skilled readers extract information about upcoming words in the parafovea. Using the boundary paradigm, we investigated native Arabic readers’ processing of orthographic, morphological, and semantic information available parafoveally. Target words were embedded in frame sentences, and prior to readers fixating them, one of the following previews were made available: (a) Identity preview; (b) Preview that shared the pattern morpheme with the target; (c) Preview that shared the root morpheme with the target; (d) Preview that was a synonym with the target word; (e) Preview with two of the root letters were transposed thus creating a new root, while preserving all letter identities of the target; (f) Preview with two of the root letters were transposed thus creating a pronounceable pseudo root, while also preserving all letter identities of the target; and (g) Previews that was unrelated to the target word and shared no information with it. The results showed that identity, root-preserving, and synonymous preview conditions yielded preview benefit. On the other hand, no benefit was obtained from the pattern-preserving previews, and significant disruption to processing was obtained from the previews that contained transposed root letters, particularly when this letter transposition created a new real root. The results thus reflect Arabic readers’ dependance on morphological and semantic information, and suggest that these levels of representation are accessed as early as orthographic information. Implications for theory- and model-building, and the need to accommodate early morphological and semantic processing activities in more comprehensive models are further discussed.

Highlights

Arabic is a Semitic language that is read from right to left
We aimed to investigate the following: (a) Parafoveal processing of Arabic orthography: Replicating the basic findings that identity previews result in preview benefit, relative to orthographically unrelated previews; (b) Parafoveal ortho-morphological processing: The effects of transposing root letters in the parafoveal preview, when the letter transposition creates a new root, and when it creates a pseudo root, while preserving letter identities in both cases; (c) Morphological parafoveal processing: The effects of providing readers with parafoveal previews that preserve the root morpheme, and previews that preserve the pattern morpheme; and (d) Semantic processing in the parafovea: Exploring whether readers would obtain preview benefit if the preview was a synonymous word with the target
Seven participants were excluded from the analyses given that their sentence comprehension scores fell below 80%

Summary

Introduction

Arabic is a Semitic language that is read from right to left. It features Semitic morphology where words are built from non-concatenated combination of root and pattern morphemes. Parafoveal processing in Arabic: Orthography, morphology & semantics root ‫ كتب‬/ktb/ is interrupted by the letter ‫ و‬/u/ of the pattern morpheme /‫ مـ ـ ـ و‬- / /m_ _ u _/, see e.g., [1,2,3]). The root morpheme indicates the main semantic family to which the word belongs (e.g., in the example above, the root ‫ كتب‬/ktb/ refers predominantly to writing-related meanings), whereas the pattern morpheme provides the detailed phonological and syntactic information that are necessary for complete and accurate word identification (e.g., [2]). The tightly-knitted orthographic, morphological and semantic information within words in Arabic make it a very interesting medium to investigate how this information is extracted from the parafovea during sentence reading

Objectives

Methods

Results

Conclusion