Abstract:
Converging evidence from cognitive science and neuroscience suggests that the brain encodes physical and abstract variablessuch as distance, time, and numerositywithin structured mental or cognitive maps. These maps are thought to play a critical role in learning, reasoning, and decision-making. In contrast, artificial neural networks (ANNs) lack map-like representations. While excelling in many domains, ANNs often require extensive training data and struggle with generalization and continual learning. In this talk, I will describe a computational framework that organizes knowledge through cognitive maps and present behavioral and neural data that support its predictions. Furthermore, I will demonstrate how integrating this framework into ANNs produces decision-making and neural activity patterns that resemble those observed in humans and other mammals. This work highlights the potential for bridging biological and artificial systems to advance our understanding of learning and memory.
Bio: Dr. Zoran Tiganj