Prior research in the domain of input enhancement has shown that exposing second language learners not only to natural input, but also to pedagogically enriched input (e.g., in so-called input floods) is often not enough to trigger incidental learning. The PhD thesis proposes that learners’ chances of incidental grammar learning from audio input crucially depend on the type and token frequency distributions of the target construction in the input texts. Based on theoretical usage-based assumptions and empirical evidence from category learning, first language acquisition, and artificial language learning, a series of two-week classroom studies investigated the effects (i) of varying degrees of type variation in the input and (ii) of the availability of a constructional prototype on incidental second language construction learning, either in first contact with a new construction or at later stages of constructional consolidation.
96 adult learners of German as a second language were thus exposed to a series of eight three-minute audio input texts, which featured a total of 150 tokens of an exemplary target construction whose type and token frequency distributions systematically varied across five test conditions and one control condition. Pretest-posttest comparisons on a range of dependent variables of receptive and productive knowledge showed (i) that both type frequencies and type-token ratios significantly impact incidental learning as a function of the learners’ prior target knowledge, and (ii) that frequency effects in the second language classroom may differ from those found in artificial language learning experiments in the laboratory. In sum, learners without prior target knowledge benefit most from moderately repetitive input featuring substantial amounts of surface similarity (where surface similarity may be contributed by reduced overall type frequencies or skewed type-token ratios, but not by a combination of both). Learners with some prior target knowledge benefit most from skewed input, which may trigger either implicit or explicit incidental learning. The findings suggest that a usage-based perspective is a promising complement to existing pedagogical Focus on Form approaches. They have important consequences for second language teaching and materials development as well as for second language acquisition theory and research methodology.
[published as: Madlener, Karin (2015). Frequency Effects in Instructed Second Language Acquisition. Berlin, Boston: De Gruyter Mouton (Applications of Cognitive Linguistics 29)]