Back

Combinatorial binding of semantic information through the sharing of neural oscillatory signals

Noguchi, Y.

2023-10-17 neuroscience
10.1101/2023.10.16.562626 bioRxiv
Show abstract

We comprehend linguistic inputs (e.g. sentence) by retrieving semantic memory of each element (e.g. word) and integrating them. How semantic information is represented and bound as neural (electric) signals is an unsolved issue. I presently used a simple sentence composed of a noun phrase (NP) and a verb (V), comparing human electroencephalography (EEG) responses to a congruent sentence in which the NP and V were semantically related (e.g. "grass grows") with those to an incongruent sentence (e.g. "a key grows"). In the left temporo-parietal cortex, neural oscillation patterns (8 - 30 Hz) to the second stimulus (V) shifted toward those to the first stimulus (NP), thereby producing coherent (faster and more regular) neural responses to the congruent sentence. No such NP-V interaction was observed in the incongruent sentence. These results indicate that the "semantic unification" as a linguistic concept actually takes place in neural oscillatory signals of the healthy human brain.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.