Humans spend a lifetime learning, storing and refining a repertoire of motor memories appropriate for the multitude of tasks we perform, whether that's playing the piano, learning sign language or peeling a banana. However, it is unknown what principle underlies the way our continuous stream of sensorimotor experience is segmented into separate memories and how we adapt and use this growing repertoire.

In this lecture, Professor Daniel Wolpert will review his team’s work on how humans learn to make skilled movements, focusing on the role of context in activating motor memories. He will then present a principled theory of motor learning based on the key insight that memory creation, updating and expression are all controlled by a single computation–contextual inference.

Unlike dominant theories of single-context learning, Wolpert’s team’s repertoire-learning model accounts for key features of motor learning that had no unified explanation, and predicts novel phenomena, which they confirm experimentally. These results suggest that contextual inference is the key principle underlying how a diverse set of experiences is reflected in motor behaviour.

The Turing Lectures: The science of movement | The Alan Turing Institute