![]() ![]() ![]() Near the other end of the spatial–temporal resolution axis, EEG provides a much higher temporal resolution allowing decoding of mental activity over very short time scales, but this comes at the cost of much poorer spatial resolution, limiting accurate decoding of activity located in sub-cortical brain regions. However, this comes at the cost of poor time resolution, which prevents decoding of mental activity over very short time scales. For example, fMRI provides a recording of activity throughout the entire brain with a very high spatial resolution, allowing a neural decoder the ability to decode mental states involving sub-cortical brain regions 11. Depending on the type of neuroimaging technique the neural decoder uses different types of mental processes may be decoded. Neural decoding models have been developed that make use of many different types of neuroimaging techniques including, but not limited to, functional magnetic resonance imaging (fMRI), electrocortiography (ECoG), electroencephalogram (EEG), and functional near infrared spectroscopy (fNIRS). These decoders were first developed to decode visual 2, 3 and semantic 4, 5, 6, 7 information from the brain, while more recent examples of neural decoders have been developed to decode a diverse set of activities, including, but not limited to, affective states 8, visual imagery during sleep 9, and story meaning 10. In recent years, neural decoders have been developed to identify numerous different types of mental activity from many neuroimaging modalities. ![]() Neural decoding models attempt to identify the current mental state of an individual from recordings of their neural activity 1. This demonstrates that our decoding model may use fMRI-informed source analysis to aid EEG based decoding and reconstruction of acoustic information from brain activity and makes a step towards building EEG-based neural decoders for other complex information domains such as other acoustic, visual, or semantic information. Using only EEG data, without participant specific fMRI-informed source analysis, we were able to identify the music a participant was listening to with a mean rank accuracy of 59.2% ( \(n~=~19\), \(p~<~0.05\)). We were able to reconstruct music, via our fMRI-informed EEG source analysis approach, with a mean rank accuracy of 71.8% ( \(n~=~18\), \(p~<~0.05\)). We further validated our decoding model by evaluating its performance on a separate dataset of EEG-only recordings. We then used fMRI-informed EEG source localisation and a bi-directional long-term short term deep learning network to first extract neural information from the EEG related to music listening and then to decode and reconstruct the individual pieces of music an individual was listening to. Specifically, we first used a joint EEG-fMRI paradigm to record brain activity while participants listened to music. ![]() In this study we explore how functional magnetic resonance imaging (fMRI) can be combined with EEG to develop an accoustic decoder. Recent studies have demonstrated neural decoders that are able to decode accoustic information from a variety of neural signal types including electrocortiography (ECoG) and the electroencephalogram (EEG). Neural decoding models can be used to decode neural representations of visual, acoustic, or semantic information. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |