Bimodal Input in Processing and Learning of Novel Vocabulary During Reading: An Eye‐Tracking Study

Abstract This study compared the effectiveness of written versus bimodal input on the processing and learning of novel words in a text. While their eye movements were recorded, 59 university‐level second language (L2) English readers read a 9,500‐word text containing 24 unfamiliar words over 2 consecutive days in one of two conditions: reading only and reading while listening. Vocabulary development was assessed with a real‐time sentence‐reading test and form recognition, meaning recall, and meaning recognition tests both immediately after the second day's learning session and 1 week later. Both modalities accelerated the processing of unfamiliar words across exposures during reading, but bimodal input induced numerically longer gaze durations on initial encounters and was more effective for learning word forms. However, neither modality supported fluent retrieval of the learned words in neutral contexts. Whereas exposure frequency predicted the learning of word forms and meanings, amount of attention predicted only word form learning.