Abstract This study examined the effect of repeated exposure on semi-incidental collocation learning, while also exploring the extent to which learner-related (prior vocabulary knowledge), and item-related factors (congruency, association strength) can modulate its effect. In an experiment using a Latin-square design, eighty-two learners of English at a university in Spain read a series of academic texts that included 18 collocations at different frequencies of exposure (1, 3, and 5). A control group did not undergo any treatment. Collocation gains were measured by a form recall and a form recognition posttest immediately after the treatment and two weeks later to explore the long-term effects of the treatment. Results from mixed-effects regression modelling showed a positive effect of repetition on learning gains, although not always statistically significant. Prior vocabulary knowledge interacted with repetition, suggesting that the facilitative effect of increased repetition decreases for L2 learners with a larger vocabulary. Congruency and association strength affected form recall only.
Viña et al. (Tue,) studied this question.