Biliteracy in Late Bilinguals : How L2 proficiency modulates reading strategies and related cognitive abilities in learners with and without dyslexia

Abstract

This dissertation aims to shed light on how reading strategies and associated cognitive skills are modulated by biliteracy in late second language (L2) learners with and without developmental dyslexia (DD), i.e., a specific learning disorder that is characterized by persistent word decoding and recognition difficulties. Biliteracy refers to being literate in two (or more) languages, and research suggests that it modulates reading strategies and related cognitive abilities uniquely depending on several factors, including the orthographic depth of the languages being learned. Further, whether two languages are learned simultaneously or sequentially might change the extent to which cross-linguistic transfer effects emerge and modulate the reading networks of bilingual individuals. Orthographic depth refers to the consistency of grapheme-phoneme correspondences (GPCs) in (alphabetic) languages, ranging from transparent orthographies to more opaque ones. For instance, the orthography of the English language is very opaque; the same grapheme might correspond to multiple phonemes depending on both positional and lexical constraints. Italian, by contrast, has a transparent orthography with its GPCs being consistent and predictable. Consequently, children learning to read opaque languages take more time to acquire fluent and accurate reading skills, whilst those learning to read transparent languages develop accurate decoding skills quite early. Further, there is considerable evidence that decoding strategies may also develop differently depending on orthographic depth: reading transparent orthographies requires focusing on smaller visual and phonological grains (graphemes or phonemes), whilst reading opaque orthographies might boost a stronger reliance on larger grains such as morphemes, syllables, or whole words. According to the Grain Size Accommodation Hypothesis (Lallier & Carreiras, 2018), learning to read two orthographies with different orthographic depths may lead to cognitive changes that “accommodate” the demands of the orthographic systems being learned. For instance, learning to read an opaque orthography alongside a transparent one might boost one’s ability to process multiple graphemes simultaneously (visual attention span: VAS). Whilst studies on early biliterates support this hypothesis, little is known about the effects of biliteracy in late biliterates with and without DD. The main goal of this dissertation is to address this gap by investigating how acquiring an opaque L2 (English) affects reading in learners whose L1 is highly transparent (Italian). Three studies will be presented. Study 1 investigates reading-related cognitive abilities in Italian adolescents and how they are linked to L1 lexical and sublexical reading skills. Study 2 investigates whether higher L2 proficiency leads to a VAS bootstrapping whereas Study 3 investigates whether the strategies that are used to decode (pseudo)words change depending on the linguistic context in which words are presented, L2 proficiency, or the presence of DD. Taken together, the results presented in this thesis suggest that learning a L2 whose orthographic properties differ from those of the L1 may lead late biliterates to develop reading strategies that accommodate the demands of their new orthography. This is especially true for typically developing learners, whilst cross-linguistic modulations were less evident, although not absent, in learners with DD. Further, this thesis shows that alongside a phonological deficit, a VAS deficit might be a potential source of difficulty for English learners with DD. Notably, positive correlations between English proficiency and dyslexic participants’ performance in the VAS task were observed and might indicate that becoming proficient readers in a foreign language could help these learners develop “new” cognitive strategies to compensate for their reading difficulties, highlighting the need for further research in this area.