6533b7d6fe1ef96bd1266e48

RESEARCH PRODUCT

READING WORDS, NUMB3R5 and $YMßOL$

subject

description

Words in alphabetic languages are processed via their constituents [1]. To recognize a printed word, we need to process the identity and position of its letters, hence distinguishing between cat and fat, or dog and god, but not between tABLe and TabLE, or and . Dehaene and colleagues [2] proposed a neuronal model with feed-forward connections only, according to which the brain decodes words through a hierarchy of local combination detectors in the occipito–infero–temporal pathway sensitive to increasingly larger fragments of words.