Mathematics, Computation, Language and the Brain (MaCLinC) is a USP Project developed by Numec.
One of the major and pressing scientific challenges of the 21st century is to process and model the huge mass of data emerging from research laboratories. This requires the development of new structures and insights, through the development of new mathematics. The aim of the USP-Project "Mathematics, computation, language and the brain" (MaCLinC) is to achieve an in-depth theoretical understanding of key phenomena in linguistics and neuroscience.
The choice of linguistics is supported by over 20 years of experience that has made the Numec group internationally recognized in the area of mathematical modeling of language acquisition and change, plus the existence of the Tycho Brahe Parsed Corpus of Historical Portuguese.
Neuroscience is at present a rapidly developing field that is clearly in need of original mathematical modeling. The field is characterized by a wealth of phenomena involving very different scales and complicated interaction patterns. These features make stochastic modeling the essential tool for an efficient treatment. The project has access to a large neural database collected by members of the team.
It turns out that the mathematical models and tools developed by our team to study linguistic phenomena are natural candidates to also analyse neural data. In fact both linguistic and neuroscience models are based on probability measures defined through graphs. In linguistics these are context trees retrieving linguistic dependencies. In neuroscience, models involve more complex graphs describing interacting neurons or clusters of neurons. The team's results on model selection ---both for context tree models and for random fields supported by interaction graphs--- provide a natural basis for future developments in these areas.
Linguistics and neuroscience are the leading applications guiding the project's mathematical activity. Nevertheless the concepts, tools and theoretical framework developed during the research in these areas are expected to have much wider application. More generally, they should be useful to study other types of large systems with variable range interactions both in space and time. The MaCLinC project has for mission to host collaborations with groups interested in this expertise.
Núcleo de Apoio à Pesquisa em Modelagem Estocástica e Complexidade - Projeto Matemática, Computação, Linguagem e Cérebro (Numec-MaCLinC)
Rua do Matão, 1010
São Paulo, SP