Objective
Mathematics is a fundamental tool by which we understand the universe, yet its cognitive and brain mechanisms are vastly understudied. I propose a systematic study of the human representation of mathematical concepts and their growth with education. The project will test the “language of thought” theory according to which humans share core concepts with other animals (numbers, objects, shapes, spatial maps) but the growth of math relies on a human symbolic composition system that recursively recombines those concepts into arbitrarily more complex expressions.
Work package (WP) 1 will generate maps of mathematical concepts in adults varying in education, using high-resolution single-subject 7T fMRI and MEG. The impact of education will be studied by testing concepts ranging from elementary to advanced, and by comparing adults whose education varies from high-school to professional math. Blind and high-functioning autism subjects will also be tested. WP2 will focus on 5 concepts of central importance in math (geometrical shape, pattern, set, number line, and graph). For each, we will map developmental change by acquiring behavioral and fMRI data in adults and children. The model predicts that concept acquisition can be modeled as a construction of increasingly complex mental expressions whose complexity is predicted by minimal description length (MDL). WP3 will map the brain changes during the acquisition of a math concept. Experiments will test the role of feedback, repetition, retrieval practice, sleep, conceptual diversity, and age, in facilitating behavioral and brain measures of conceptual change. Finally, WP4 will examine whether artificial neural networks can capture the above results and investigate how these networks can be enhanced to achieve human-like performance.
Overall, the results will shed light on how mathematical education changes the human brain and how conceptual change occurs, thus paving the way to real-life educational applications in schools.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- natural sciencesmathematics
- natural sciencescomputer and information sciencesartificial intelligencecomputational intelligence
You need to log in or register to use this function
Keywords
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Topic(s)
Funding Scheme
HORIZON-ERC - HORIZON ERC GrantsHost institution
75015 PARIS 15
France