One of the main debates in second language acquisition research revolves around the relative significance of explicit and implicit learning conditions. However, it is widely acknowledged that highly automatized and implicit knowledge (henceforth automatized-implicit knowledge) should be aimed for in language teaching and learning. Automatized-implicit knowledge – rather than unautomatized explicit knowledge – allows learners to use language accurately and spontaneously (e.g. DeKeyser, 2017). Surprisingly, to my knowledge, no study has explored the differential (or complementary) effect of extramural, i.e. out-of-class, language use and instructed learning on the implicit-explicit knowledge continuum. Such research is sorely needed in light of the sharp increase of extramural English across Europe and beyond.
To fill this gap, students aged 13-14 years in Austria and Sweden (N = 213) completed tests of automatized-implicit (oral narrative test, elicited imitation, aural and written timed grammaticality judgment tests) and explicit knowledge (untimed grammaticality judgment test, metalinguistic knowledge test). These countries were selected because they seem to differ in the more explicit vs. implicit learning environment they provide. Compared to Austria, the type of instruction lower secondary learners in Sweden receive appears to be more implicit (Schurz & Coumel, 2020) and extramural English use is more extensive. The lower levels of extramural English in Austria can be explained by the fact that audio tracks of foreign-language films and series screened on TV and online platforms like Netflix are usually replaced by German ones. In contrast, in Sweden, foreign-language audio tracks in audio-visuals are typically supplemented by Swedish subtitles rather than being dubbed in Swedish. In the present study, information on the participants' weekly use of extramural English and the type of instruction they experienced was collected through mixed-method learner and teacher surveys.
The findings point to interesting disparities between the more explicit vs. implicit learning environments in learners' development of automatized-implicit knowledge. For example, in linear mixed models, the effect of extramural English emerged as significantly positive only in the Swedish sample (p < .01), and most clearly so in terms of multimodal activities such as viewing videos, series, and films in English original version. Contrarily, in Austria, learning (still) seems to take place primarily in the classroom. The apparent contextual differences in knowledge development are accounted for in a proposed theoretical model elaborating on DeKeyser's Skill Acquisition Theory (2015).
