Boosting Non-Native Speaker Engagement: Simplifying Text with Large Language Models

Effective communication between native and non-native speakers is crucial for intercultural collaboration and successful teamwork. Non-native speakers often struggle to comprehend complex messages, which can lead to miscommunication or exclusion. We explored using Large Language Models (LLMs) to translate complex text into summarized and simplified versions to improve readability and accessibility. We tested this approach with non-native speakers in a controlled experiment, where participants were given both complex and simplified versions of the same text. Although the simplified text was rated as easier to read, there was no significant difference in comprehension accuracy compared to the original text. This suggests that simplification might lead to a loss of critical information, impacting overall comprehension. Our findings suggest that text simplification, while enhancing readability, must be balanced against the risk of reduced comprehension accuracy. This has implications for designing communication tools in diverse and multicultural environments, emphasizing the need for a careful approach to ensure effective communication without compromising essential information.