Incidental and intentional acquisition of multiword expressions from audio-visual input: The effects of typographically enhanced captions and repetition

This research project aims to extend the line of inquiry on pedagogical interventions intended to help second language (L2) learners make better progress of their mastery of multiword expressions (MWEs). Existing studies on these interventions revealed a propensity towards exclusivity in terms of input modality, item type and learning condition. Firstly, there are far more MWE studies in the context of unimodal input, e.g., written input. It is only recently that the potential of audio-visual input (i.e., L2 viewing) has been explored for MWE learning. Secondly, previous studies have by and large focused on certain types of MWEs, such as collocations. While there is merit in focusing on a certain type of item, such studies do not represent the materials that L2 learners are often exposed to. Further, authentic videos entail diverse MWE types, providing a stronger reason to include more than one type of target item. Thirdly, many MWE interventions are investigated exclusively under one of the learning conditions, i.e., intentional or incidental learning conditions. Hulstijn’s (2001) criterion is adopted to distinguish these two learning conditions, in that the presence of test announcement characterises the intentional learning condition. Due to this tendency towards a dichotomy of learning conditions, many factors known to facilitate MWE learning have been investigated under one of the learning conditions only.  Two such factors are repetition and typographic enhancement. While repetition is well established as beneficial for MWE acquisition, evidence for this is mainly furnished by studies on incidental learning through written input. Therefore, the aim of this research project is to assess how repetition, operationalised as repeated viewing, influences MWE acquisition under both learning conditions. Similarly, although typographic enhancement has been shown to draw learners’ attention and promote MWE uptake, this positive evidence is mostly observed in incidental learning studies. As such, whether typographically-enhanced MWEs are indeed learned better than unenhanced MWEs under intentional learning conditions is still under-researched. Importantly, whether typographic enhancement in captioned viewing leads to superior learning compared to normal captions is unknown. This is one of the aims of the research project, in which different caption conditions are created to explore their effectiveness in facilitating MWE learning. Of further interest is whether MWE learning under different caption conditions would modulate the effect of repetition. This is motivated by the assumption that typographic enhancement might eliminate the need for repetition.  To answer the research questions, two studies differentiated by the presence of test announcement were carried out. For both studies, ESL learners watched a video containing target MWEs under one of six conditions, which differed in terms of caption condition (no captions, normal captions or enhanced captions) and the number of viewing times (once or twice). MWE learning was assessed through tests that tap into form and meaning knowledge at the level of recall and recognition. Though not part of the research questions, the effects of caption condition and repetition on content comprehension were also assessed. The findings of both studies revealed trends that are consistent with literature on MWE learning and vocabulary learning in general. Firstly, both types of captions promoted better form recall knowledge compared to uncaptioned viewing. This was found to be true under both incidental and intentional learning conditions. Secondly, typographically enhanced captions led to better form recall compared to normal captions, but only under the intentional learning conditions. Under the incidental learning conditions, the effects of L2 viewing with typograhically enhanced captions on form recall appeared to be similar to viewing with normal captions. The findings also suggest that the presence of typographically enhanced captions reduced the number of viewings needed to make incidental gains in form recall knowledge. In addition, while repeated viewing under all caption conditions led to better knowledge of form under the incidental learning conditions, the effect of repetition was not found under the intentional learning conditions. This aligns well with the supposition that fewer repetitions are needed for intentional learning. Thirdly, neither repetition nor caption condition had an effect on the acquisition of MWE meanings under both learning conditions. Finally, vocabulary knowledge played a significant role in the amount of MWE learning that takes place, especially so when learners were not forewarned of MWE tests. Taken as a whole, the findings of this research project support the use of captions for L2 viewing as a way to foster MWE acquisition, at least at the level of form acquisition. The use of typographically enhanced captions, however, may have adverse effects on content comprehension. As such, the findings of this research project have meaningful implications concerning when typographically enhanced captions and repeated viewing should be used to optimise MWE learning through L2 viewing.