Dual-Route Word Naming Race

When you read a word aloud, two neural pathways race each other inside milliseconds: a non-lexical route that sounds out letters piece-by-piece using spelling-to-sound rules, and a lexical-semantic route that recognizes the whole word and retrieves its meaning and pronunciation directly from memory. This simulator renders that race in 3D as two accumulator pathways converging on a shared output node, using the dual-route cascaded (DRC) model of reading aloud. Choose a stimulus word or nonword, adjust its frequency and your simulated processing speed, and watch which route reaches threshold first — and, for irregularly-spelled words, whether the winner produces the correct pronunciation or a telltale regularization error, the same error pattern documented in surface dyslexia.