Hierarchical temporal receptive windows and zero-shot timescale generalization in biologically constrained scale-invariant deep networks.

ABSTRACT: Human cognition integrates information across nested timescales. While the cortex exhibits hierarchical Temporal Receptive Windows (TRWs), local circuits often display heterogeneous time constants. To reconcile this, we trained biologically constrained deep networks, based on scale-invariant hippocampal time cells, on a language classification task mimicking the hierarchical structure of language (e.g., 'letters' forming 'words'). First, using a feedforward model (SITHCon), we found that a hierarchy of TRWs emerged naturally across layers, despite the network having an identical spectrum of time constants within layers. We then distilled these inductive priors into a biologically plausible recurrent architecture, SITH-RNN. Training a sequence of architectures ranging from gen

SUBMITTER: Sarkar A 

PROVIDER: S-EPMC12803329 | biostudies-literature | 2026 Jan

REPOSITORIES: biostudies-literature