Data-driven identification of nonlinear cognitive models

FIRST CALL
APPLICATION ID: ALL14

What we are looking for:

We are seeking innovative research proposals that explore data-driven approaches for identifying nonlinear cognitive models.

The context:

Neural manifolds are emerging as a paradigm shift for understanding the complex relationship between neural activity and cognitive processes, and have the potential to inform the development of therapeutic interventions (i.e. identifying biomarkers, targeted stimulation, personalized approaches, etc…).

The problem to address:

We aim to model the nonlinear dynamics of neural manifolds resulting from massive recordings of brain activity and its embodied representations. Recordings are already available and consist of either Neuropixels and multicellular calcium imaging from the dorsal hippocampus of behaving mice, together with face/body imaging. The project involves applying machine learning tools and dynamical systems approaches to evaluate neural representations in collaboration with experimentalists and engineers. We next plan to construct avatars that simulate cognitive processes underlying task performance.