EEG Macrostate Clustering for Slow Attentional Salience Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring salience, particularly slow attentional salience, are inadequate as they are often focused on fast-perceptual aspects and do not effectively assess slow attentional salience mechanisms, which are crucial for understanding attentional redirection and cognitive processes.
Innovation Solution
A computer-implemented method and apparatus that utilize multi-channel EEG signals to determine salient reactivity by obtaining topographic spectral power maps, clustering them to identify macrostates, and calculating the suppression difference between a default macrostate during target and reference stimuli, providing a numeric index for salient reactivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If EEG event related potentials and frequency domain measures are used to study salience, then fast-perceptual aspects of salience can be measured, but slow attentional salience mechanisms cannot be effectively assessed
Solution Approach 1:
The patent segments the continuous EEG signal into distinct macrostate sequences through clustering analysis, separating different brain network states (default mode network vs. task-positive networks). This segmentation allows independent measurement of transition dynamics, enabling precise quantification of slow attentional salience mechanisms that were previously confounded with faster perceptual responses.
Solution Approach 2:
The patent introduces a temporal dimension to salience measurement by analyzing the duration and sequence of macrostate transitions rather than relying solely on amplitude or frequency domain measures. This dimensional shift from static to dynamic measurement captures the temporal evolution of attentional networks, revealing slow attentional salience mechanisms that operate on different timescales than traditional measures.
2Loss of information
If traditional EEG methods focus on event related potentials, then novelty and expectancy effects can be captured, but top-down attentional salience mechanisms remain confounded
Solution Approach 1:
The patent extracts the default mode network macrostate sequence as a separate, independently analyzable component from the overall EEG signal. By isolating DMN transitions and comparing them against task-positive network transitions, the method extracts pure attentional salience signals while filtering out confounding factors like novelty and expectancy effects that are embedded in traditional ERP analyses.
Solution Approach 2:
The patent performs preliminary clustering of EEG data into macrostate sequences before analyzing salience-related transitions. This preliminary organization of data into distinct network states allows subsequent analysis to focus specifically on transitions between attentional configurations, eliminating the need to process raw EEG data and reducing computational complexity while preserving critical information.
3Measurement precision
If behavioral measures like eye-tracking are used to assess perceptual salience, then physical salience can be measured, but top-down attentional salience cannot be addressed
Solution Approach 1:
The patent creates a universal EEG-based macrostate transition measurement framework that can assess both bottom-up perceptual salience and top-down attentional salience within a single methodology. By analyzing transitions between default mode and task-positive networks, the same measurement system adapts to capture different types of salience mechanisms across various experimental paradigms and sensory modalities, eliminating the need for separate behavioral measures.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention provides a method for determining a salient reactivity of at least one target sensory stimulus applied to a subject, the method comprising: - obtaining physiological signals from the subject during a stimulation with the target sensory stimulus and during a stimulation with a reference sensory stimulus, the physiological signals comprising a plurality of channels, - determining, at a plurality of time points of the physiological signals, topographic spectral power maps, wherein a topographic spectral power map comprises spectral powers of frequency bands of the plurality of channels, - clustering the topographic spectral power maps to obtain a plurality of macrostates, - identifying, among the plurality of macrostates, a default macrostate as the macrostate which shows a largest suppression during stimulation, and - determining the salient reactivity based on a difference between the suppression of the default macrostate during the target sensory stimulus and a suppression of the default macrostate during the reference sensory stimulus.