Abstract
How does information move through the brain, and could the physical shape of a system be just as important as its connections? In this lecture, I will explore a new way of thinking about neural networks—where signals travel as waves across curved surfaces, and learning happens by slowly reshaping the space through which those waves flow. Using networks inspired by cells arranged as 3D “digital neural organoids”, I will illustrate how activity spreads across layered, spherical networks, forming wavefronts, spirals, and rhythmic patterns. These waves are not just visual curiosities: their timing, direction, and stability determine how well information reaches key regions of the system. By modulating the positions of individual nodes, the network trains itself, focusing signals inward, synchronizing their arrival, and reducing noise—much like adjusting the shape of a lens to bring an image into view. The lecture will include animated visualizations of nested organoid surfaces changing over time, directing signals toward a central core, and sometimes swirling into persistent spatial patterns that can store information. No particularly advanced mathematics is required. Instead, I hope to build intuition around familiar ideas—waves, flow, and geometry—to show how learning and computation might emerge from space itself. This fresh perspective opens new ways to think about brain development, artificial intelligence, and the future of biologically inspired computing.