PRISM-M: A Recurrent Framework for the Formation of Stable Internal Neural Models
PRISM-M is a recurrent framework that simulates how neural representations integrate into stable models, showing context-dependent shifts and predictive accuracy in simulations and fMRI data.