Lab: Marina Bedny
School: Krieger School of Arts and Sciences
Department: Psychological & Brain Sciences
This image captures the brain’s hidden “algorithm sense.” By comparing brain responses to programming algorithms before (algorithms described in plain English) and after (algorithms described in Python code) students learned programming, we found that the neural patterns supporting code are rooted in pre-existing systems for logical reasoning. The matrix in the center shows which algorithms are more alike based on what they actually do, and the surrounding brain maps show where the brain represents that same pattern of similarity. Learning to program “recycles” what the brain can already do, and adapts it to a new format.