Modeling Evolutionary Brain Process

Modeling Evolutionary Brain Process


​• The stochastic properties and features of high dimensional brain signals evolve over the course of learning a new skill and during memory tasks. 

We will develop new time series models that characterizes the evolution of connectivity over the course of learning. 

We will develop novel low-dimensional representations of these high-dimensional signals.

These models will take into account the within-trial and betwen-trial correlations as well as the variation across experimental units. 

The challenges lie in accounting for non-stationarity and multi-scale correlation in signals: local (within trial) and global (across trials).

Prior Work:

        *Modeling the Evolution of Dynamic Brain Processes During an Associative Learning Experiment, Journal of the American Statistical Association​ (2016).​

        * Video of evolutionary spectra and coherence of local field potentials (LFP)​​


Mark Fiecas​ (University of Minnesota)