Mapping the Neural Encoding of Speech Sounds in the Human Cortex
Originally Published 2 years ago — by Nature.com
Researchers used high-density multielectrode Neuropixels probes to record cellular activity from hundreds of individual neurons across the cortical layers in the superior temporal gyrus (STG) of the human brain while participants listened to naturally spoken sentences. They found that single neurons in the STG encode a wide range of speech features, including acoustic-phonetic features, onsets from silence, intensity, relative pitch, lexical stress, and phoneme and word sequence probability. The encoding patterns varied across cortical depth, with different neurons tuned to different speech properties. The findings provide insights into the cortical representation of speech and the organization of neuronal responses in the STG.