Abstract
<p>Studying how bilingual populations compute grammatical illusions sheds light on how variables such as language use and practices impact language processing. Although encompassing variation in language of testing and bilingual profile can further clarify the role of such variables, studies tend to focus on only one of bilinguals’ languages and consider bilingual groups speaking a limited set of standard language pairs. By adopting a two-way comparative perspective, we investigate how Subject-Verb agreement mismatches are processed in two non-standard language bilingual communities of Italy, namely Italian-Agrigentino (n = 24) and Italian-Pavese (n = 23) bilinguals. Despite both being characterized by the coexistence of a standard and non-standard linguistic variety, each community presents unique sociolinguistic ecologies in terms of linguistic habits. Participants did an auditory acceptability judgement task with sentences with incorrect Subject-Verb agreement as stimuli. The same sentence was presented in Italian and either Agrigentino or Pavese according to the bilingual group, aiming to run a between-language comparison and observe whether the (non) standard status of a language influences processing outcomes. Judgements were elicited on a 5-point Likert scale and reaction times were recorded. Both bilingual groups gave higher acceptability ratings to ungrammatical stimuli in Italian compared to the non-standard language, meaning that they better spotted agreement mismatches in the standard variety. Yet, Italian-Agrigentino bilinguals were faster in judging stimuli in their non-standard language compared to Italian, while Italian-Pavese bilinguals showed the opposite trend. Other between-group differences concerned both the type of sociolinguistic factor modulating processing outcomes, such as the amount of non-standard language use and switching, and the way in which these factors played a role. Our results suggest that appreciating variation in both the language of testing and bilingual phenotype is crucial when examining bilingual language processing, especially in contexts of non-standard language bilingualism.</p>