Vocalization, Text Presentation, and Learner‐Related Factors in Learning L2 Multiword Expressions in Sentences

ABSTRACT This study examined whether vocalization (speaking aloud) supports the learning of multiword expressions (MWEs) in sentences, and how text presentation, vocabulary size, and working memory (WM) capacity affect outcomes. Seventy‐one Japanese university and graduate students studied 80 English MWEs (verb–noun idioms) under four within‐participant conditions combining vocalization and text presentation. In each condition, 20 MWEs were introduced and assessed through item memory, form recognition, and meaning recognition tests; an additional 20 MWEs appeared only in the tests for comparison. Participants also completed a vocabulary size test and two digit span tasks. Results showed that silent learning was more effective than vocalization, while text presentation supported learning. Vocabulary size contributed more than WM capacity. These findings suggest that vocalization does not always enhance learning and that, when learning MWEs embedded in sentences, learning without vocalization may be more effective.