Exploring sensitivity to number agreement violations in L2 processing: L1 effect is not the only indicator for successful parsing

Alqahtani, Wael 2024. Exploring sensitivity to number agreement violations in L2 processing: L1 effect is not the only indicator for successful parsing. PhD Thesis, Cardiff University. Item availability restricted.

	PDF - Accepted Post-Print Version Restricted to Repository staff only until 11 September 2025 due to copyright restrictions. Download (5MB)
	PDF (Cardiff University Electronic Publication Form) - Supplemental Material Restricted to Repository staff only Download (144kB)

Abstract

This thesis addresses whether L2 speakers of English are sensitive to morphosyntactic features in reading for comprehension tasks. According to L2 sentence processing research, L2 speakers are less sensitive to morphosyntactic features during L2 comprehension because they are unable to form fully specified syntactic and morphosyntactic representations during processing, as suggested by the shallow structure hypothesis (SSH, Clahsen and Felser, 2006a, 2006b, 2018). However, another line of research has found that L2 speakers can form such representations, but they are more susceptible to memory interference when retrieving morphosyntactic features from memory, as proposed by the increased susceptibility to interference model (ISI, Cunnnings, 2017). In this thesis, I examine whether L2 speakers are sensitive to the number agreement violations when reading sentences with a complex subject with two noun phrases (NPs) differing in number (e.g., the key to the cabinets is/are rusty …). These NPs present an agreement attraction effect. Previous studies reported contradictory results regarding whether L2 speakers are sensitive to such violations and whether their sensitivity may be attributed to the influence of L1 on L2 processing (e.g., Jiang, 2004; Lim & Christianson, 2015). However, no study has investigated whether the richness of morphosyntactic information in the L1 can affect the processing of the number agreement violations in subject-verb dependency in a L2 with less morphological information, as the case of English. Therefore, L2 participants with different degrees of morphological information encoded in their first languages are recruited to assess whether they are sensitive to grammaticality and attraction interference effects when reading the number agreement violations in L2 English comprehension. The assumption was whether sensitivity to such violations is influenced by whether the morphosyntactic feature (i.e., subject-verb number agreement) is instantiated in bilinguals' first language. Thus, L2 speakers of English with Arabic and Spanish first languages, which are morphologically rich languages where the number agreement is available and significant, were recruited to determine whether L1 affects L2 processing. Moreover, L2 speakers of English with Mandarin Chinese first language, which lacks inflectional marking for number agreement morphology, were recruited to test whether L2 processing sensitivity is possible even if it is absent in the first language. To test whether participants are sensitive to such violations, this thesis uses a sensicality judgement paradigm with rapid serial visual presentation (RSVP) and two-stage presentation. The use of this iii paradigm is significant to identify the extent of sensitivity to the number agreement violations. L2 processing sensitivity was examined in sentences where the attractor was embedded in two sentence structures: prepositional phrases and relative clauses. The findings of L2 speakers in both sentence structures suggest that the L1 effect is not the only indicator of L2 processing sensitivity to grammaticality and attraction interference effects. The results showed that L2 speakers with a higher level of proficiency and whose first language has a similar orthographic system to English (i.e., Spanish speakers) were more sensitive to grammaticality and attraction effects than Arabic and Mandarin Chinese speakers. Moreover, L2 speakers’ sensitivity was different across the RSVP and the two-stage presentation methods which highlights the effect of task type in L2 processing research. Most importantly, because L2 speakers were sensitive to the grammaticality effect, this indicates that they formed fully specified syntactic and morphosyntactic representations during processing. This contradicts the shallow structure hypothesis. Thus, the nature of L2 processing sensitivity agrees with the increased susceptibility to interference model, which assumes that L2 speakers can form representations, but they are more susceptible to interference during memory retrieval operations. The thesis provides insightful implications of the research findings including effect of L1 directionality mode, task type and cue-weighting on L2 processing.

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	English, Communication and Philosophy
Funders:	Saudi Arabia Government
Date of First Compliant Deposit:	11 September 2024
Last Modified:	13 Sep 2024 15:53
URI:	https://orca.cardiff.ac.uk/id/eprint/172045

Actions (repository staff only)

Edit Item