This study reports a small-scale experiment that was set up to estimate the extent to which (i) the use of formulaic sequences (standardized phrases such as collocations and idiomatic expressions) can help learners come across as proficient L2 speakers and (ii) an instructional method that emphasizes ‘noticing’of L2 formulaic sequences can help language learners add such phrases to their linguistic repertoire. Participants were 32 college students majoring in English. Over the course of 22 teaching hours they were exposed to considerable quantities of authentic listening and reading materials. During exploration of those materials, the experimental students (N = 17) were made aware of standardized word combinations, while in the control group (N = 15) the traditional grammar-lexis dichotomy was upheld. Afterwards, the participants’ oral proficiency was gauged in an interview by two blind judges. Both perceived the experimental group as more proficient than the control group. Two other blind judges counted the number of word combinations in the interviews that they considered to be formulaic sequences. Their counts correlated well with the oral proficiency ratings, which suggests that helping learners build a repertoire of formulaic sequences can be a useful contribution to improving their oral proficiency.

No abstract

No abstract

Many contemporary textbooks for English as a foreign language (EFL) and books for vocabulary study contain exercises with a focus on collocations, with verb–noun collocations (e.g. make a mistake) being particularly popular as targets for collocation learning. Common exercise formats used in textbooks and other pedagogic materials require learners to establish appropriate matches between sets of verbs and nouns. However, matching exercises almost inevitably carry a risk of erroneous connections, and despite corrective feedback these might leave undesirable traces in the learner’s memory. We report four small-scale trials (total n = 135) in which the learning gains obtained from verb–noun matching exercises are compared with the learning gains obtained from a format in which the target collocations are presented to the learners as intact wholes. Pre-test to post-test gains turned out small in all of the conditions, owing in part to the learners’ substitution of initially correct choices by distracters from the exercises. The latter, negative side-effect was attested more often in the matching exercises than in the exercises where the learners worked with collocations as intact wholes.