System Identification of the Dynamics of Multisensory Integration

The central nervous system (CNS) processes sensory information obtained through various sensory structures in the body. These include signals with a wide variety of spatiotemporal features, such as different speed and propagation patterns.

In this stream of multimodal sensory information, CNS must decide how it should integrate these signals to construct a unique representation of the environment. How CNS accomplishes this process is not known. In this project, I will investigate the filtering mechanisms adopted by CNS to integrate multisensory information.

Specifically, I will study multisensory integration within the context of two unique behaviors: (1) glass knifefish combine visual and electrosensory cues to track the movements of a refuge in which it is hiding, and (2) zebrafish utilize vision and mechanosensory lateral line to sense the direction and velocity of the local current during their rheotaxis behavior.

To accomplish this, I will first build a novel experimental setup, a speed-controlled flow tunnel, which allows independently probing different sensory modalities for both the glass knifefish and the zebrafish.

I will adopt a control-theoretic approach to identify how CNS combines multisensory information under different sensory conflict scenarios.

Specifically, I will estimate the frequency response functions for the sensory weights assigned to different sensory modalities.

Moreover, we will observe how CNS dynamically changes these weights when there is a change in the saliency of the available sensory information.

Our goal is to use system identification theory to generate models that capture the dynamics of online, real-time sensory re-weighting mechanism adopted by these fish.