Proceedings of the Annual Meeting of the Cognitive Science Society

High-quality multiple-choice tests are required to assess learners' language abilities in language education. When creating such tests, it is important to create appropriate false-answer choices. Expertise is required to create such false-answer choices: inappropriate false-answer choices can make test questions too easy or invalid. Experts who can create such excellent tests are limited. In this study, we evaluated whether large-scale language models can create false-answer choices for language education. Specifically, we discuss the extent to which language models can generate false-answer choices in tests created by applied linguists. We show the results of our analysis of the generated false-answer choices.