Memory in Silico: Building a Neuromimetic Episodic Cognitive Model

We propose a neurologically plausible computational architecture to model human episodic memory and recall based on cortical-hippocampal structure and function. The model design is inspired by neuroscience findings and categorical neural semantic theory.  Fuzzy Adaptive Resonance Theory (ART) and temporal integration are used to form episodic representations.  Supported by the mathematical formalism of category theory, we theorize that correctly forming and manipulating episodic representations of sensory inputs, we can model elements of cognitive behavior.  We demonstrate the architecture’s function by comparison to human experiments in an episodic memory task.