Nature human behaviourHumansNeural NetworksComputerTransferPsychology
Humans and neural networks show similar patterns of transfer and interference during continual learning.
Eleanor Holton, Lukas Braun, Jessica Af Thompson, Jan Grohn, Christopher Summerfield
Published: 202610.1038/s41562-025-02318-y
Abstract
In artificial neural networks, acquiring new knowledge often interferes with existing knowledge. Here, although it is commonly claimed that humans overcome this challenge, we find surprisingly similar patterns of interference across both types of lea…
Preview only. Read the full abstract at the source