Incidental Acquisition of Multiword Expressions Through Repeated Viewing: Sequential Use of <scp>L1</scp> and <scp>L2</scp> Captions

Abstract Repeated viewing is reportedly a common learning and pedagogical strategy among autonomous second language (L2) learners and language teachers. This experimental study examined the extent to which sequential captioning use facilitates the acquisition of multiword expressions (MWEs) through repeated viewing under incidental learning conditions. A total of 109 Chinese‐speaking learners of English were randomly allocated to three experimental groups and one control group. The three experimental groups were exposed to an input video containing target MWEs three times, with the following sequences of captioning configurations: (a) L1 captions–L2 captions–no captions; (b) L2 captions–L2 captions–L2 captions; and (c) no captions–L2 captions–L1 captions. MWE pretests and posttests were administered to gauge participants' learning outcomes in form recognition and meaning recall. The control group watched a different video in which no target MWEs appeared and completed the same pretests and posttests. Mixed‐effects logistic regression modeling suggested an overall advantage of the sequential captioning conditions (i.e., L1 captions–L2 captions–no captions and no captions–L2 captions–L1 captions) over the consistent use of L2 captions. The benefits of sequential captioning use were discussed in relation to retrieval and trial‐and‐error cognitive learning mechanisms.