Exploring the Utility of Narrative Comprehension and Production Tasks to Identify Vocabulary Delay Using Eye-Tracker

Objectives: The purpose of this study was to identify group differences in various aspects of narrative tasks performance (comprehension, production, and eye-tracking variables) and finally to explore whether narrative tasks can be used to detect vocabulary delay. Methods: Children in the 1st and 2nd grades (TD=101, VD=22) of elementary school participated in this study. All children participated in story comprehension and production tasks, and eye-tracking variables were extracted from two reading conditions (‘listening+ reading’, and ‘reading’) in the story comprehension task. Results: There was a significant difference between the two groups in the ‘listening+reading’ condition in the story comprehension task, the ‘construction’ condition in the story production task, and the ‘total fixation duration (TFD)’ in the eye-tracking variables. We examined the ability of the story performances to discriminate vocabulary delays, and found that a total of 79.6% of the original cases based on the standardized vocabulary test could be correctly classified by the story performance variables. The ROC curve also showed that 84.6% of the children could be validly identified by a combination of narrative comprehension scores in the ‘listening+reading’ condition, narrative production scores in the ‘structure’ condition, and TFD in the ‘listening+reading’ condition of comprehension task. Conclusion: The significance of this study is that it explored the possibility of using narrative tasks as a useful tool for identifying micro-level vocabulary delays.