Transcript-based estimators for characterizing interactions

Area of Science:

• Complex Systems Analysis
• Neuroscience
• Time Series Analysis

Background:

• The concept of transcripts, introduced in 2009, provides a framework for characterizing functional relationships between interacting time series using algebraic relations between ordinal patterns.
• Estimators for interaction strength, direction, and complexity based on transcripts have been developed but lack widespread application in real-world system studies.

Purpose of the Study:

• To revisit the transcript concept and demonstrate the utility of transcript-based estimators for investigating interactions in dynamical systems.
• To apply these methods to analyze human brain dynamics and uncover insights into spatial-temporal interactions related to vigilance states.

Main Methods:

• Utilized transcript-based estimators derived from algebraic relations between ordinal patterns of time series.
• Performed time-resolved analysis on multichannel, multiday recordings of human brain activity.
• Investigated interactions in coupled paradigmatic dynamical systems of varying complexity.

Main Results:

• Successfully applied transcript-based estimators to analyze complex interactions in dynamical systems.
• Demonstrated the potential of these methods for time-series-based investigation of real-world systems.
• Revealed novel insights into intricate spatial-temporal interactions in human brain dynamics during different vigilance states.

Conclusions:

• Transcript-based methods offer a powerful approach for characterizing functional relationships and interactions in complex systems.
• The application to human brain dynamics highlights the potential for novel discoveries in neuroscience.
• This work encourages wider adoption of transcript-based estimators in diverse scientific fields.