Charan Ranganath

Professor of Psychology and Neuroscience at UC Davis and Director of the Dynamic Memory Lab, known for his research on how the human brain forms, stores, and retrieves episodic memories.

Background

Ranganath trained at UC Berkeley — where a formative class with Daniel Kahneman introduced him to the psychology of memory before Kahneman’s Nobel — and built his research programme around the neuroscience of episodic memory: how we remember specific events, why we forget, and what the relationship between memory and imagination reveals about the architecture of the mind. His lab uses fMRI, electrophysiological recordings, and computational modelling to study how the hippocampus and prefrontal cortex interact during the encoding and retrieval of experience.

He is also a musician who played in touring bands in graduate school (including contributing a soundtrack to a film), and approaches memory science with the sensibility of someone attuned to prediction, pattern, and surprise — the same elements that animate both jazz and neural learning. He describes himself as having many characteristics of ADHD and finds science an ideal environment for that kind of attention.

His book Why We Remember: Unlocking Memory’s Power to Hold On to What Matters (2024) translates his research for a general audience, arguing that forgetting is not a failure of the system but its most important feature.

Core positions

• Memory is not a recording but a reconstruction: every act of retrieval is an active reassembly of fragments, shaped by current context, prior knowledge, and what the brain predicts will be useful.
• The hippocampus encodes memories selectively, at moments of prediction error and surprise (event boundaries), not continuously. Routine experience is compressed; novelty is marked.
• Imagination and memory share the same neural substrate (the default mode network and hippocampus); they differ only in the directive given to the same recombinatory process.
• The act of remembering changes the memory — a retrieved memory is temporarily plastic and can incorporate misinformation introduced at recall.
• Human episodic memory solves the stability–plasticity dilemma in a way current AI systems do not: by keeping a fast, one-shot episodic store alongside a slow-learning semantic store, allowing rapid learning from single events without catastrophic forgetting.
• Déjà vu is a familiarity signal without a matching recollection: a partial structural match to a past experience generates the sense of having been here before, without the specific memory to complete it.

In the wiki

• False Memory — concept page synthesising this episode with Julia Shaw‘s applied research