Machine Learning Classifies Word Type Based on Brain Activity
New paper by the NeDComm group at CFIN just published in eNeuro.
Pairing machine learning with neuroimaging can determine whether a person heard a real or made up word based on their brain activity, according to a new study just published at eNeuro. These results lay the groundwork for investigating language processing in the brain as well as for the development of an imaging-based tool to assess language impairments in various conditions.
Many brain injuries and disorders cause language impairments that are difficult to establish with standard language tasks because the patient is unresponsive or uncooperative, creating a need for a task-free assessment method. Using magnetoencephalography (MEG), CFIN scientists Mads Jensen, Rasha Hyder and Yury Shtyrov examined the brain activity of participants while they listened to audio recordings of similar-sounding real words with different meanings and made up "pseudowords".
Using machine learning algorithms, the team were able to determine, based on the participant's brain activity in MEG, when they were hearing a real or made up word, as well as determine the meaning of the particular word and whether it was grammatically correct or incorrect. They also identified specific brain regions and neural activity frequencies responsible for processing different types of language information.
For full details, please see their paper at:
Read more about the NedComm group at CFIN: