URL with Digital Object Identifier
This study investigates the potential benefits for incidental vocabulary acquisition of implementing a particular sequence of input-output-input activities. More specifically, EFL learners (n = 32) were asked to watch a TED Talks video, orally sum up its content in English, and then watch the video once more. A comparison group (n = 32) also watched the TED Talks video twice but were not required to sum it up in between. Immediate and delayed post-tests showed significantly better word-meaning recall in the former condition. An analysis of the oral summaries showed that it was especially words which learners attempted to use that stood a good chance of being recalled later. These findings are interpreted with reference to Swain’s (e.g., 1995) Output Hypothesis, Laufer and Hulstijn’s (2001) Involvement Load Hypothesis, and Nation and Webb’s (2011) Technique Feature Analysis. What makes the text-based output task in this experiment fundamentally different from many previous studies which have investigated the merits of text-based output activities is that it was at no point stipulated for the participants that they should use particular words from the input text. The study also illustrates the potential of TED Talks as a source of authentic audio-visual input in EFL classrooms.