Think Like an Expert

Decoding understanding from student brains

How do students understand and remember new information? In Neural alignment predicts learning outcomes in students taking an introduction to computer science course, we measure understanding in student 🧠 and use it to predict and assess learning outcomes in a STEM course. The good news? Classmates are guide to success!

Neural alignment among students during computer science lectures predicted performance in a final exam. For example, in the hippocampus, neural alignment to the rest of the class predicted overall exam scores weeks in advance. And during the exam, neural alignment tracked understanding of specific concepts in individual learners. Concepts that evoked better alignment with the experts - and with other students - were better understood. This could help design better courses, but for students, it might be more useful to know why a concept is difficult and how to understand it better. We found that we can probe students’ “knowledge structure”, pinpoint the problem and potentially say, “you struggle with recursion because you need to revisit loops”.

Methods & tools:

• Functional MRI
• Multi-voxel pattern analysis (MVPA)
• Non-parametric statistics
• Machine learning / Data Science tools: Python3,R, NumPy, SciPy, Pandas, SciKit-Learn, Seaborn
• Neuroimaging tools: FSL, BrainIAK, NiBabel, Nilearn, Connectome workbench
• High-performance computing (HPC) (on prem)