Edited by Martin Bygate
[Task-Based Language Teaching 11] 2018
► pp. 43–73
Chapter 2Grammatical structures and oral fluency in immediate task repetition
Trigrams across repeated performances
In this study we examine to what extent words and grammatical structures are re-used when a speaking task is repeated with the same content (i.e., specific task repetition). We examine this re-use, which has been argued to support proceduralization and fluency development (N. de Jong & Perfetti, 2011), under both constant and increasing time pressure, and we investigate the correlation between re-use and fluency. The analyses are performed not only on individual words but also on trigrams, which are sequences of three words (e.g., the red car; here: lexical trigrams) or three parts of speech (e.g., det adj noun: POS trigrams), to capture grammatical structure. Thirty-nine adult ESL speakers completed repeated retellings of one to three picture stories. One group followed the 4/3/2 procedure (Nation, 1989), which involves three iterations with gradually increasing time pressure; for the other group the available time was constant. The extent of re-use of words and grammatical structures across task iterations was calculated using cosine similarity with tf-idf weighting (Manning, Raghavan, & Schütze, 2008), which adjusts for the frequency of words or trigrams, both within an iteration and across iterations and speakers. It was found that immediate task repetition had a strong effect on re-use at the level of individual words and trigrams, but increasing time pressure did not. The relationship between re-use and fluency was variable, but showed higher re-use for speakers struggling with fluency. We conclude that, if fluency development is to be stimulated by re-use of words and grammatical structures, it can be done with specific task repetition, whether under increasing time pressure or not.
Article outline
- Introduction
- Immediate task repetition
- What speakers do when they repeat a task
- Measuring similarity between iterations
- What to count: Similarity at the level of the word and above
- How to compute similarity: Using frequency to estimate informativeness
- How to compute similarity: Cosine similarity with tf-idf weighting
- Research questions
- Method
- Participants
- Materials
- Procedure
- Analysis
- Results
- Re-use between and within prompts
- Re-use under constant and decreasing time and by iteration
- Correlations between similarity and fluency
- Discussion and conclusion
-
Acknowledgements -
Notes -
References
https://doi.org/10.1075/tblt.11.02jon
References
Cited by
Cited by 7 other publications
This list is based on CrossRef data as of 05 january 2023. Please note that it may not be complete. Sources presented here have been supplied by the respective publishers. Any errors therein should be reported to them.