Abstract
We investigated cross-language influences from the first (L1) and second (L2) languages in third (L3) language processing, to examine how order of acquisition and proficiency modulate the degree of cross-language influences, and whether these cross-language influences manifest differently in online and offline measures of L3 processing. The study focused on morpho-syntactic processing of English as an L3 among Arabic-Hebrew-English university student trilinguals (n = 44). Importantly, both L1 (Arabic) and L2 (Hebrew) of participants are typologically distant from L3 (English), which allows overcoming confounds of previous research. Performance of trilinguals was compared to that of native English monolingual controls (n = 37). To investigate the source of cross-language influences, critical stimuli were ungrammatical sentences in English, which when translated could be grammatical in L1, in L2 or in both. Thus, the L3 morpho-syntactic structures included in the study were a mismatch with L1, a mismatch with L2, a Double mismatch, with both L1 and L2, or a no mismatch condition. Participants read the English sentences while their eye-movements were recorded (online measure), and they also performed grammaticality judgments following each sentence (offline measure). Across both measures, cross-language influences were assessed by comparing the performance of the trilinguals in each of the critical interference conditions to the no-interference condition, and by comparing their performance to that of the monolingual controls. L1 interference was evident in first pass sentence reading, and marginally in offline grammaticality judgment, and L2 interference was robust across second pass reading and grammaticality judgments. These results suggest that either L1 or the L2 can be the source of cross-language influences in L3 processing, but with different time-courses. The findings highlight the difference between online and offline measures of performance: processing language in real-time reflects mainly automatic activation of morpho-syntactic structures, whereas offline judgments might also involve strategic and meta-linguistic decision making. Together, the findings show that during L3 processing, trilinguals have access to all previously acquired linguistic knowledge, and that the multilingual language system is fully interactive.
Highlights
Multilingualism can be considered a conventional feature of linguistic experience and maturity (Hammarberg, 2010)
Data from monolingual native English speakers was collected in order to gauge the processing difficulty of the various target structures, and to characterize the baseline complexity of processing each structure in the absence of any cross-language influence
The findings of the current study suggest that the entire linguistic repertoire is activated when processing L3
Summary
Multilingualism can be considered a conventional feature of linguistic experience and maturity (Hammarberg, 2010). While in second language (L2) acquisition learners rely solely on their experience with one language, in L3 acquisition two pre-existing systems of linguistic representations are available (Westergaard, 2019; Puig-Mayenco et al, 2020). Cross language influence is evident when acquisition or processing of one language is influenced by existing knowledge of other languages (Cenoz, 2001). Such influences can be facilitative, when structures of two languages are similar (positive transfer), but can lead to language interference (negative transfer), in the presence of structural differences between the languages in question (Isurin, 2005; MacWhinney, 2005). There is a sizeable body of knowledge regarding how L1 can influence L2 processing (Hopp, 2010; Prior et al, 2017) and vice versa (Dussias and Sagarra, 2007; Degani et al, 2011), but our current understanding of how linguistic knowledge in L1 and/or L2 influences L3 learning and processing is far from being complete (Angelovska and Hahn, 2012; Rothman et al, 2019; Lago et al, 2020; PuigMayenco et al, 2020)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
