Real-time processing of formality-register and morphosyntactic (in)congruence in adolescent readers: An eye-tracking study

This eye-tracking reading study investigates the real-time processing of context formality-register (mis)matches and morphosyntactic (in)congruence by means of eye-tracking studies in two speaker groups: German monolingual adolescent speakers and adolescent heritage speakers of Russian in Germany. The present study aims to establish the similarities and differences in the way adolescents process formality-register and morphosyntactic (mis)matches, both representationally and procedurally. Our goal is to determine whether a single mechanism and closely interconnected mental representations suffice, or if separate mental representations and mechanisms are required for morphosyntactic and register processing. Before proceeding with the description of the study, we will define some key terminology that will be used in this preregistration. · The term ‘register’ refers to the variations in language use by individuals within a speech community, shaped by the situational-functional context (Pescuma et al., 2023). · A communicative situation (comm-sit) is a social activity centered around language production and perception (Wiese, 2021). In other words, the situational-functional context defines comm-sits, while the specific linguistic features used in these contexts make up different registers. · A heritage language is a language acquired in the family, usually as the L1, and importantly it is not the dominant language of the larger (national) society (Rothman, 2009; Lohndal et al., 2019). Furthermore, the heritage language is often used only in certain comm-sits: predominantly in the family and with peers, less frequently in informal public settings, and almost never in formal settings. Due to this limited exposure to formal registers, studies on heritage speakers' production skills suggest that they experience register leveling, meaning they use linguistic structures typical of informal language even in formal situations (Wiese et al., 2022; Tsehaye, 2023; Iefremenko, manuscript submitted for publication; Zuban, 2024). · A majority language refers to the language spoken by a larger society in which the speaker resides. The term ‘majority language’ is often synonymous with the dominant language, where the dominant language is the language that is chosen by default in the majority of interaction domains, especially in public and institutional ones (Matras, 2020). Usually, the majority language is an L2 for bilingual children, which they acquire in kindergarten or school; although sometimes it is also acquired in family as the second L1. This research builds upon previous studies that investigated the processing of context formality-register (mis)matches and morphosyntactic (in-)congruence during reading in German monolingual young adults (Pescuma et al., 2024; Pescuma, Haneda, Ito, Maquate, & Knoeferle, manuscript submitted for publication). The studies showed that the processing of formality-register congruence occurs gradually during sentence comprehension. Effects of formality-register congruence were observed only after the critical word was read, suggesting ongoing integration with the preceding context at that time. In contrast, the effects of subject-verb morphosyntactic congruence were observed more pronounced and earlier (at the verb region). Consistent with previous research on syntactic processing, morphosyntactic incongruences generally resulted in longer reading times. Our current study broadens its scope by incorporating additional factors, such as age, bilingualism, and proficiency. This study includes adolescents aged 16 to 19 years, as there is limited understanding of how they process subject-verb agreement compared to adults. Event-related brain potential evidence has shown differences in morphosyntactic processing between adolescents and young adults. For example, Meier (2008) compared English-speaking adolescents and young adults, and found that morphosyntactic violations elicited an N400 response, typically taken to index semantic integration, in the adolescent group only, while a Late Positive Component (LPC) was elicited in both participant groups, albeit with a significantly different time course (with the LPC beginning, peaking, and ending earlier in the adolescent group). Such differences were interpreted as indicating that the development of verb comprehension was still partially ongoing in late adolescence. Investigating this age group will thus enhance our understanding of the cognitive processes involved in the development of language processing during adolescence. It is also relatively unclear how adolescents process verb-argument mismatches compared to adults, since psycholinguistic research involving adolescents is not as extensive as research involving adults. An event-related potential (ERP) study by Courteau et al. (2023) compared morphosyntactic number agreement processing in French-speaking adolescents with and without developmental language disorder (DLD). They found that in adolescents without DLD, number agreement processing was still developing, and their ERP patterns resembled those of lower proficiency speakers rather than skilled adults. Another ERP study reported that both adolescents and adults showed similar N400 effects for semantically incongruent compared to congruent sentences. However, unlike adults, adolescents did not show an N400 response to congruent sentences describing risky events, even though both groups rated risky sentences as riskier than predictable ones (Grose-Fifer et al., 2011). These findings suggest that differences in perspective or world view between adolescents and adults may shape aspects of semantic congruence processing. Furthermore, our study extends to bilingual speakers with Russian as their heritage language and German as the majority language. This group is well represented in Germany, which makes our research question relevant in the context of cultural and linguistic differences. Additionally, the adaptation of the German study to Russian was facilitated by the typological similarity of the two languages, both of which share the SVO word order. Due to the predominant use of German in education, social interactions, and professional settings, Russian heritage speakers in Germany often develop stronger proficiency in German, while their heritage language skills gradually weaken over time. Consequently, many heritage speakers exhibit advanced skills in German and only limited proficiency in Russian, especially in reading and writing (Bremer, 2021). It is important to include bilinguals in the study for several reasons. First, the majority of the world’s population uses two or more languages in their everyday life (Grosjean, 2010) therefore monolingual speakers cannot be taken as the norm (Rothman et al., 2023). Secondly, heritage speakers represent a unique cohort when it comes to the research on register because they grow up using two languages in different social contexts; for instance, the heritage language is mainly used in informal situations, while the majority language is used across the formal and informal contexts (Rothman, 2009). Finally, to our knowledge, no study to date has investigated how heritage speakers process formality-register (mis-)matches in real-time.