ANNOUNCEMENT: Ph.D. Research Proposal Exam

 

Name: K P Kasun D Pathirage

 

Committee:

Professor Pamela Abshire (Chair)

Professor Timothy Horiuchi

Professor Behtash Babadi


Date/time: 11/26/2025 at 11.30 AM

 

Location: AVW 1146

 

Title: Modulation and Mapping of Functional Connectivity in Engineered Neuronal Networks via Closed-Loop Optical/Electrical Stimulation and Recording

 

Abstract: 

Understanding how living neuronal networks process information and adapt to external perturbations is central to neuroscience. In this work, I investigate the mapping, modulation, and adaptive plasticity of engineered neuronal networks cultured within polydimethylsiloxane (PDMS) devices designed to constrain and guide neuronal projections.

This research is driven by three fundamental questions: 1) Mapping Connectivity: How can electrical and optical recordings be used to infer the functional connectivity of a neuronal network, specifically, the presence, directionality, and strength of synaptic interactions? 2) Modulating Connectivity: What stimulation paradigms and techniques can reliably alter this connectivity map, effectively modifying the network's synaptic weights? 3) Network Training: Can these stimulation protocols be used to train living neuronal networks to perform meaningful computations?

While prior studies have demonstrated network scale training by comparing pre- and post- stimulation activity patterns, such approaches often lack interpretability - similar to the limited interpretability of artificial neural networks. Therefore, changes to the connection strengths between individual neurons, or groups of neurons is poorly understood. A robust method for mapping functional connectivity would enable direct comparison between pre- and post- training connection strengths, revealing which stimulation strategies most effectively induce plasticity.

To address this, I will develop tools and algorithms to infer the network connectivity from large scale micro-electrode array recordings. The use of polydimethylsiloxane (PDMS) constrained networks provides a ground truth for validating these inference methods, as the imposed topologies limit the space of possible connections. In addition, the unique topology imparted by PDMS microstructures enables precise analysis of axonal and dendritic pathways, facilitating detailed studies of both structural and functional connectivity.

Once reliable connectivity maps are established, I will implement closed-loop stimulation protocols that leverage real-time feedback to deliver targeted electrical and optical inputs. These protocols aim to induce specific modifications in network connectivity, effectively “training” the network toward desired functional states. Post-intervention connectivity will be reassessed to quantify the magnitude and specificity of induced plasticity.

Together, this work elucidates how environmental constraints and temporally patterned stimulation shape the reorganization of living neural circuits. The findings will offer mechanistic insights into the principles of learning, memory, and adaptation in biological systems.