Using AI-Generated Prequestions to Improve Memory and Text Comprehension

Prequestioning is a learning method that involves guessing the answers to questions about to-be-learned information (i.e., prequestions), followed by the opportunity to discover the correct answers. Its application during self-regulated learning requires access to practice questions, which can often be quite limited. Across four experiments, we investigated a potential solution: using artificial intelligence (AI)—specifically, the large language model-based chatbot, ChatGPT—to generate prequestions. Experiment 1 found that AI-generated prequestioning enhanced learning of an educational text relative to simply reading. Experiment 2 found that AI-generated questions and human-generated questions were equally effective at promoting learning. Experiments 3-4 revealed that prequestioning with questions generated using different AI prompts aided learning relative to studying an AI-generated outline. Across experiments, prequestioning benefited subsequent test performance on questions that were identical and non-identical to those previously encountered. Together, these results highlight the potential of generative AI to facilitate prequestioning, yielding substantial learning benefits.