Cognitive Effects of Large Language Models: An Empirical Study of Memory Retention, Reasoning Strategies, and Critical Thinking among Students and Professionals at the University of Gujrat, Pakistan

This study investigates whether the regular use of large language models (LLMs), such as ChatGPT, is reshaping human cognition by altering memory retention, reasoning strategies, and critical thinking skills. Drawing from cognitive psychology and the extended mind hypothesis, the researcher conducted a mixed-methods study involving 120 participants—80 students and 40 professionals—divided into frequent and non-frequent LLM users. Experimental tasks measured short- and long-term recall, problem-solving approaches, and critical evaluation of arguments, complemented by surveys and interviews on user perceptions. Findings suggest that LLM users demonstrate improved efficiency in problem-solving but lower long-term recall, with a tendency to externalize reasoning and rely more heavily on AI-generated outputs. Implications for education, workplace training, and epistemic trust are discussed, emphasizing the need for pedagogical strategies that preserve critical thinking in AI-mediated environments.