Abstract (eng)
One of the main debates in second language acquisition research revolves around the rela-tive significance of implicit and explicit learning conditions in learners’ language develop-ment. This question is pertinent especially in exploring how learners achieve high levels of implicit and/or automatized knowledge. These types of knowledge should be aimed for in language teaching and learning, given they can be employed rapidly and effortlessly in spon-taneous language use. Today, the debate needs to be revisited in light of implicit learning conditions created through extramural, i.e., out-of-class, English, which is currently sharply increasing across Europe and beyond. Yet, to my knowledge, no study to date has explored the effect of extramural language use on implicit and/or automatized knowledge (henceforth automatized-implicit knowledge). This research gap was addressed in the present project, which further enquired into the interactions of out-of-school learning contexts with tradi-tional foreign language learning in the classroom. This was done by adopting a cross-country perspective. A total of 213 learners aged 13–14 years, 110 in Austria and 103 in Sweden, were recruited together with their respective English teachers (n = 14). The two countries differ in the kind of learning environment they provide, providing more explicit and implicit instruction re-spectively. Through a learner and teacher questionnaire as well as a teacher interview, in-formation on the type of instruction (e.g., systematic vs. incidental grammar instruction) and the frequency, starting age, and weekly use of extramural English was collected. The data were explored using (Multivariate) Analyses of Variance and qualitative interview analyses. To measure the learners’ grammatical knowledge, they performed tests of automa-tized-implicit knowledge (oral narrative test, elicited imitation, and aural and written timed grammaticality judgment tests) and explicit knowledge (untimed grammaticality judgment test and metalinguistic knowledge test). A Confirmatory Factor Analysis was computed on these widely-used measures to gauge their construct validity among young, non-academic learners in the two different types of learning environments. To enquire into the interplay of instruction, extramural English, and levels of automatized-implicit and explicit knowledge, Linear Mixed Models were computed. Concerning classroom practices, questionnaire and interview data point to more systematic and explicit grammar teaching in lower secondary English instruction in Austria as com-pared to Sweden. Among Swedish learners, regular extramural English use on average start-ed earlier and occupied more time on a weekly basis. This population typically does not have wide access to dubbed foreign-language media. Given the more implicit learning environ-ment in Sweden, the participants there presumably performed the six grammar tests largely based on automatized-implicit rather than (unautomatized) explicit knowledge. This would explain why Swedish students’ performance on the six grammar tests loaded on a single factor in the Confirmatory Factor Analysis. Compared to Swedish learners, Austrian partici-pants scored significantly higher on the metalinguistic knowledge test, and, as expected, the factor analysis produced two factors, which were labelled automatized-implicit and explicit knowledge. The effect of extramural English emerged as significantly positive only in the Swedish sample, most consistently so in terms of watching audio-visuals and gaming. Con-trarily, in Austria, instruction apparently currently holds a greater potential in aiding the construction of automatized-implicit knowledge than spare time English. I therefore con-clude that extramural English use may predict automatized-implicit knowledge, notably if it starts early and is extensive and multimodal. This study appears to be the first to demon-strate the impact of out-of-class English on the development of automatized-implicit knowledge. In addition, it shows interesting disparities between more implicit vs. explicit learning environments. Such contextual differences are accounted for in a proposed theoret-ical model of Context-Dependent Skill Acquisition, an elaboration of Robert DeKeyser’s Skill Acquisition Theory (2015).