In his talk, Simon explains that as a stimulus, speech drives robust neural responses all along the auditory and language pathways, allowing the simultaneous investigation of multiple speech processing mechanisms at once, even using the same speech stimulus. He discusses recent results regarding the progression of neural representations of speech at different steps along the auditory pathway and beyond, from auditory cortical representations to lexical (language-based) representations, and further. The earliest representations are dominantly acoustic-stimulus driven (“bottom up”), but later cortical representations become more and more influenced by general cognitive (“top down”) factors and by language, and tend to reflect the speech as it is perceived rather than as the bottom-up acoustic input would predict.