Exploration of LLMs, EEG, and behavioral data to measure and support attention and sleep
This work addresses the problem of measuring and supporting attention and sleep for individuals, but it is incremental as it highlights limitations and needs for further data and knowledge.
The study explored using large language models (LLMs) to estimate attention states, sleep stages, and sleep quality from EEG and behavioral data, and generate personalized sleep improvement suggestions and guided imagery scripts. Results showed LLMs could estimate sleep quality from textual behavioral features and provide personalized suggestions, but detecting attention and sleep stages from EEG and activity data required more training and domain-specific knowledge.
We explore the application of large language models (LLMs), pre-trained models with massive textual data for detecting and improving these altered states. We investigate the use of LLMs to estimate attention states, sleep stages, and sleep quality and generate sleep improvement suggestions and adaptive guided imagery scripts based on electroencephalogram (EEG) and physical activity data (e.g. waveforms, power spectrogram images, numerical features). Our results show that LLMs can estimate sleep quality based on human textual behavioral features and provide personalized sleep improvement suggestions and guided imagery scripts; however detecting attention, sleep stages, and sleep quality based on EEG and activity data requires further training data and domain-specific knowledge.