MT syntactic priming effects on L2 English speakers

MT syntactic priming effects on L2 English speakers Nat´alia Resende ADAPT Centre Dublin City University School of Computing natalia.resende@ adaptcentre.ie Benjamin Cowan ADAPT Centre University College Dublin School of Information and Communication Studies benjamin.cowan@ ucd.ie Andy Way ADAPT Centre Dublin City University School of Computing andy.way@ adaptcentre.ie Abstract In this paper, we tested 20 Brazilian Por- tuguese speakers at intermediate and ad- vanced English proficiency levels to inves- tigate the influence of Google Translate’s MT system on the mental processing of English as a second language. To this end, we employed a syntactic priming experi- mental paradigm using a pretest-priming design which allowed us to compare par- ticipants’ linguistic behaviour before and after a translation task using Google Trans- late. Results show that, after performing a translation task with Google Translate, participants more frequently described im- ages in English using the syntactic alter- native previously seen in the output of Google Translate, compared to the transla- tion task with no prior influence of the MT output. Results also show that this syntac- tic priming effect is modulated by English proficiency levels. 1 Introduction Machine Translation systems (MT), especially Google Translate, have become popular in the last decades (Clifford et al., 2013). The popularity of these systems has grown not only due to tech- nical improvements, but also due to the facilita- tion of users’ access through the proliferation of mobile applications containing a number of func- tions that allow users to translate from texts, from speech or from a text image (Gupta and Dhawan, 2019; Chinnery, 2008). The rapid development of c⃝ 2020 The authors. This article is licensed under a Creative Commons 3.0 licence, no derivative works, attribution, CC- BY-ND. MT technologies is the result of massive research investment in the field over the past decades fo- cusing on language resources, new methods and techniques with the aim of improving the qual- ity of the MT output. Consequently, the progress made in MT technology has changed the way peo- ple are engaging with these systems (Gaspari and Hutchins, 2007). In the past, MT systems were used mainly for gisting purposes, but nowadays they are also be- ing used as a tool supporting writing skills, gram- mar skills and language production in a second lan- guage (L2) (Ni˜no, 2006; Garcia and Pena, 2011). However, research in the MT field focusing on end-users is limited so that, currently, little we know about what the users’ interaction with an MT system could bring to the mental processing of a second language. In this paper, we aim at inves- tigating the role MT systems play, especially, the role that the popular MT system Google Translate plays on the processing of English as a second lan- guage. Specifically, we investigate the influence of Google Translate on the processing of English as L2 by testing whether the MT output can influence the way MT users process English syntax. We ad- dress the following research questions: 1. Is the use of MT capable of affecting the …

Content truncated to 500 words