Assessing the Impact of Large Language Models on Problem-Solving Skills of Undergraduates - A Systematic Literature Review

The integration of Large Language Models (LLMs) such as GPT-3 and GPT-4 in educational environments is rapidly transforming how undergraduate students approach problem solving tasks. This systematic literature review (SLR) aims to investigate the 32 studies that were published between 2021 and 2024 in order to assess the impact of LLMs on problem solving skills. The review identifies the significant improvements in undergraduates' problem-solving effectiveness, creativity, and computational thinking. It also emphasizes the need for careful implementation of these tools to avoid potential overreliance, which may impact the student's ability to develop independent problem-solving skills. The review identifies the importance of a well-balanced integration of LLMs in education to maximize their benefits while addressing the challenges they present. Recommendations for future research directions are suggested to further explore the long-term impacts of LLMs on the development of problem-solving skills and their applicability across diverse academic environments.