Lexical access by different script adult bilinguals; Evidence from masked primed picture naming and phoneme monitoring tasks

Alharbi, Manal 2022. Lexical access by different script adult bilinguals; Evidence from masked primed picture naming and phoneme monitoring tasks. PhD Thesis, Cardiff University. Item availability restricted.

Preview	PDF - Accepted Post-Print Version Download (4MB) | Preview
	PDF (Cardiff University Electronic Theses and Dissertations Publication Form) - Supplemental Material Restricted to Repository staff only Download (2MB)

Abstract

This study investigates lexical access in different script bilinguals; namely Arabic-English adult speakers, a group that is rarely investigated. It covers cross-language activation, the manner of lexical/phonological selection, and the flow of activation in different script bilinguals, and whether proficiency level modulates cross-language activation, manner of lexical/phonological selection, or flow of activation. Currently there is a substantial and rapidly growing body of empirical evidence describing and evaluating lexical access in bilinguals. However, the majority of these studies focus on same script bilinguals (e.g., Spanish-English), with limited research addressing different script bilinguals, such as Arabic-English bilinguals. It is argued here that the findings concerning non-selective lexical access cannot be generalized to different script bilinguals, as script differences can act as language cues and strict lexical access to target language. Therefore, we conducted five different experiments to investigate the performance of highly and less proficient adult Arabic-English bilinguals using three different tasks: a masked primed picture naming task (experiments one, two, and three), a phoneme monitoring task (experiments four and five), and an animacy decision task (experiment three). The use of the masked priming paradigm to address lexical access, manner of selection and flow of activation in bilinguals is unprecedented. In the first experiment, the participants were required to name in the L2 (English), cognate and non-cognate pictures that were preceded by L1 (Arabic) masked translation primes. A significant cognate facilitation effect, and a translation facilitation effect were observed for both highly and less proficient bilinguals. These findings suggest the bilinguals experienced non-selective access (i.e., both languages were activated simultaneously), and that the manner of lexical/phonological selection was language specific (i.e., considers activated nodes in the target language only). Moreover, the findings pertaining to cognate facilitation suggest a cascaded flow of activation for the non-target language. In experiment two, non-cognate pictures were preceded by L1 semantically related masked primes, which were presented for 50 ms, 75 ms, and 100 ms. The semantic interference effect was evident when the masked primes were presented for 75 ms, and 100 ms. This suggests that the lexical selection process is language non-specific (i.e., considers activated nodes in the target and non-target language), which contradicts the findings reported in experiment one. Experiment three investigated the locus of the semantic interference effect to establish if it is at the conceptual or lexical level. Thus, it compared the effect of semantically related masked primes across two tasks; i.e., the animacy decision task, which involves conceptual processing, and the masked primed picture naming task, which involves both conceptual and lexical processing. The results demonstrated that the semantic interference effect was obtained in the masked primed picture naming task; whereas semantic facilitation effect was obtained in the animacy decision task. This suggests the locus of semantic effect is at the lexical level. In experiment four, the participants performed a phoneme monitoring task in L2, in which they had to decide whether a visually presented phoneme was part of the L2 picture name. The phonemes were either, part of the picture name in L2 in the positive condition, part of the picture name in L1 in the critical condition, or unrelated. The results revealed the participants experienced difficulties rejecting the phoneme when it was part of the L1 picture name. Thus, the Arabic-English bilinguals’ two languages appear to be activated simultaneously, and the activation of the non-target language cascades to the phonological level. Experiment five examined whether the findings of the phoneme monitoring task in experiment four would be replicated when L1 distinct phonemes (do not exist in participants’ L2 language) were used in an additional critical condition. It was found to be hard to reject L1 phonemes, even when the phonemes are L1 distinct. The findings confirm non-selective access, as well as the cascaded flow of activation during the production process by different script bilinguals. Taken together, the results reported suggest the manner of lexical access in different script bilinguals is language non-specific, and that activation flow cascades to the phonological level. In addition, the results imply the lexical/phonological selection process considers the activation of the target and non-target lexical nodes. Regarding the role of script differences and the participants’ language proficiency level, the results suggest no modulation of cross-language activation, manner of lexical/phonological selection, or flow of activation. The implications for bilingual models of lexical access are also discussed.

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	English, Communication and Philosophy
Subjects:	P Language and Literature > PE English
Date of First Compliant Deposit:	14 July 2022
Last Modified:	14 Jul 2023 01:30
URI:	https://orca.cardiff.ac.uk/id/eprint/151294

Actions (repository staff only)

Edit Item