Thesis
Exploring the impact of semantic features on sentence prediction in school-age children
University of Iowa
Master of Arts (MA), University of Iowa
Spring 2024
DOI: 10.25820/etd.007456
Abstract
Comprehending sentences efficiently is essential for effective communication. Two components contributing to this process are semantic organization and prediction. Semantic organization is the categorization of concepts based on featural similarity, while prediction involves anticipating upcoming words based upon context. Although both semantic organization and prediction have been studied independently in adults and school-age children, their integration in real-time sentence comprehension, especially in children, remains poorly understood. Our study seeks to address this gap by adopting an integrated approach to understanding these skills in school-age children, who are actively developing rapid and efficient sentence comprehension abilities. Fifteen middle schoolers were tested using a dual-sentence N400 ERP task where the first sentence provided context (e.g., I saw a fat, pink animal rolling in the mud on the farm.) and the second sentence carried the target word (e.g., I was happy to see the pig.). There were four conditions: one was the match word that was expected to fit the semantic features (e.g., pig), one was a near violation that shared many semantic features with the match word (e.g., cow), one was a far violation that shared few semantic features with the match word (e.g., bird), and the final was a between-category violation that did not share any semantic features with the match word (e.g., refrigerator). We hypothesized that if semantic organization is not influencing prediction, the N400 effect will either have an equal amplitude across all violations or an N400 will only be present in the between-category violation. If semantic organization is influencing prediction, the N400 amplitude will increase by graded increments from the match word (i.e., lowest amplitude) to the between-category word (i.e., highest amplitude). We found a significant difference in N400 amplitudes between the match word and between-category violations, but only a marginal difference between the match and far violations and the near and between-category violations. These results partially support the view that school-age children use semantic organization to predict upcoming words in a sentence.
Details
- Title: Subtitle
- Exploring the impact of semantic features on sentence prediction in school-age children
- Creators
- Adalee M. Donahue
- Contributors
- Kristi Hendrickson (Advisor)Philip Combiths (Committee Member)Stewart McCauley (Committee Member)Elizabeth Walker (Committee Member)
- Resource Type
- Thesis
- Degree Awarded
- Master of Arts (MA), University of Iowa
- Degree in
- Speech Pathology and Audiology
- Date degree season
- Spring 2024
- DOI
- 10.25820/etd.007456
- Publisher
- University of Iowa
- Number of pages
- vii, 26 pages
- Copyright
- Copyright 2024 Adalee M. Donahue
- Language
- English
- Date submitted
- 04/19/2024
- Description illustrations
- Illustrations, graphs, charts
- Description bibliographic
- Includes bibliographical references (pages 23-26).
- Public Abstract (ETD)
Comprehending sentences efficiently is essential for effective communication. Two components that contribute to sentence comprehension are semantic organization and prediction. Semantic organization refers to the way humans categorize concepts based on how similar their features are to other concepts. Prediction is the ability to guess upcoming words in a sentence based on the context previously provided. Semantic organization and prediction have both been independently researched, but understanding how they influence each other in real time, especially in children, is poorly understood.
Our study seeks to address this gap in research by studying these two components simultaneously as participants read sentences. Fifteen middle schoolers were tested using sentence pairs. The first sentence in each pair provided context (e.g., I saw a fat, pink animal rolling in the mud on the farm.) and the second sentence carried the target word (e.g., I was happy to see the pig.). There were four conditions: one was the match word that was expected to fit the semantic features (e.g., pig), one was a near violation that shared many semantic features with the match word (e.g., cow), one was a far violation that shared few semantic features with the match word (e.g., bird), and the final was a between-category violation that did not share any semantic features with the match word (e.g., refrigerator). Data was collected through measuring brain waves and tracking the N400, an event-related brain potential associated with semantic deviation.
Our results showed a significant difference in N400 amplitudes between the match word and between-category violations, but only a marginal difference was found between the match and far violations and the near and between-category violations. These results partially support the view that school-age children use semantic organization to predict upcoming words in a sentence. Overall, this is preliminary research and evidence to support the integration of semantic organization and prediction in sentence comprehension as seen in school-age children. Research should continue to be conducted in this area to expand our knowledge and build on the results we have found.
- Academic Unit
- Communication Sciences and Disorders; Center for Social Science Innovation
- Record Identifier
- 9984647256202771
Metrics
1 File views/ downloads
22 Record Views