Study shows reading code uses non-language cognitive skills

The human brain facilitates reading, writing, mathematical reasoning and scientific logic through the use of specific cognitive skills. The relatively recent evolution (in terms of cognitive tool development) of computer programming has led to rapid advancement, yet not much is known about the cognitive and neural systems required to support it.

A new study from Massachusetts Institute of Technology (MIT) suggests that reading computer code does not access the part of the brain associated with language processing, rather relying on the ‘multiple demand (MD) network’ which is typically called into action for complex tasks such as solving mathematical problems or completing crossword puzzles.

The study, titled ‘Comprehension of computer code relies primarily on domain-general executive brain regions’, was published in the journal eLife in December 2020. Researchers utilised magnetic resonance imaging (MRI) to analyse the response of two brain systems — the MD and the language system — to code written in Python (text-based programming language) and ScratchJr (a graphical programming language). The MD system exhibited strong responses to both, whereas the language system responded strongly to text-based problems and weakly or not at all to graphic programming language.

The researchers say this new information does not necessarily clarify whether coding should be taught as a math- or language-based skill, as it remains unclear as to whether learning to program (as opposed to reading code) relies on both language and MD systems.

You can read the full study here.

Image credit: ©stock.adobe.com/au/Siarhei