Neural recycling and plasticity in computer programming expertise

Johns Hopkins University, Baltimore MD

Investigators

Abstract

Computer programming skills are increasingly fundamental to many jobs across diverse fields, including in healthcare, science, communication, finance, and transport. Programming instruction is being incorporated into standard K-12 and post-secondary educational curricula. Programming is a key part of STEM education and a gateway to success in the STEM workforce. However, compared to other skills, like math and reading, we know little about the cognitive and neural underpinnings of programming. Many people become highly proficient coders and program professionally, as well as for pleasure. But there are wide individual differences in how quickly programming is learned and the ultimate programming ability that is achieved. The causes of these differences are not well understood. This proposal uses cutting edge neuroscience and cognitive science approaches to study the neurocognitive systems that support programming skills. We investigate which neural and cognitive systems support programming and how the human brain changes itself to make learning to program possible. This research is a first step to harnessing the adaptive capability of the human brain to optimize the training of programming skills. The project aims to directly engage students with disabilities and from minoritized groups to provide an opportunity to participate in cutting edge research on this critical topic. In this proposal the researchers test hypotheses about which neural systems support programming and how these systems change during learning. One hypothesis is that learning programming ‘languages’ like Python engages parts of the brain that evolved for processing natural languages, like English and Spanish. There is also evidence that programming engages logical reasoning systems in prefrontal and parietal cortices that support solving logic puzzles. This proposal uses cutting edge neuroimaging techniques to study the different contributions of these systems and their connectivity to programming skills. First, the researchers aim to measure brain function, anatomy, and behavior in the same students before and after they take their first programming class. This approach tests what pre-existing mechanisms are repurposed by programming education. Machine learning analyses can then be used to study detailed neural patterns in the brains of people before and after they learn to program and the locations and extent of changes quantified. Further, changes in the anatomical communication pathways between language and logical reasoning systems can also be quantified before and after learning. A second study compares brain function and behavior across people with widely different programming expertise, from people who are programming naïve to people who are programming experts and code every day as part of their jobs. Together these approaches can yield a better understanding of the neural and cognitive basis of programming and which cognitive abilities (e.g., language, reasoning, math) and neural measures predict programming ability. This research aims to serve as a foundation for education research and the design of interventions to optimize programming instruction. The study of programming also provides insight into mechanisms of plasticity in higher-order cognition. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.