Keiko Ohmae

AI-like parallelogram geometry of word relations in the human neocortex

The paper shows that semantic relations like queen-king and woman-man form parallelogram geometry in the human neocortex, similar to language AI, suggesting a shared organizational principle.

Emergence of syntax and word prediction in an artificial neural circuit of the cerebellum

This study develops a biologically-constrained cerebellar neural network model to explore language-related functions like word prediction and syntax recognition, finding that a recurrent pathway supports both.

A recurrent circuit links antagonistic cerebellar modules during associative motor learning

This study investigates the functional organization of the cerebellum in mice during motor learning and identifies three operational principles that challenge the conventional view of cerebellar architecture.