Measuring understanding and predicting STEM learning outcomes from neural activity

How do students understand and remember new information? Measuring and decoding human brain activity during educational experiences offers new ways to address this fundamental question. Today, it is unclear how learners internalize new content, especially in real-life and online settings. In this talk, Dr. Meshulam will demonstrate how neural data can be used to diagnose individual participants’ understanding of specific topics within a broad-ranging, real-world STEM course (Introduction to Computer Science, at Princeton University). This approach hinges on the finding that understanding is mirrored in “neural alignment”: the degree to which an individual learner’s neural representations match those of experts, as well as those of other learners. If time permits, Dr. Meshulam also will discuss ongoing work in which we use recent advances in natural language processing (NLP) to model the learning process and track changes in neural concept representations over the span of the course. 

 

Dr. Meshulam is fascinated with human learning and with machine learning, which hold so much promise for each other. At Princeton, he develops tools for improving real-world learning using neuroimaging and computational methods. Previously, as a machine learning algorithm developer, he designed data analysis systems inspired by human cognition.