Stochastic brain dynamics exhibits differential regional distribution and maturation-related changes

Area of Science:

• Neuroscience
• Biophysics
• Data Science

Background:

• Functional magnetic resonance imaging (fMRI) measures brain activity via blood oxygenation level-dependent (BOLD) signals.
• BOLD signals result from complex interactions between deterministic and stochastic biological processes.
• Estimating stochastic brain dynamics is difficult due to dependence on deterministic assumptions.

Purpose of the Study:

• To develop a framework for estimating intrinsic stochastic brain dynamics in fMRI data.
• To characterize neural noise without prior assumptions on deterministic dynamics.
• To investigate the lifelong maturation of intrinsic brain noise.

Main Methods:

• Utilized Approximate Entropy to identify and characterize dynamical noise in fMRI time series.
• Developed a methodological framework to estimate unobservable fMRI dynamics.
• Applied the framework to large-scale fMRI datasets (Cam-CAN and Human Connectome Project).

Main Results:

• 10% to 60% of fMRI signal power is attributed to intrinsic stochastic brain elements.
• The proportion of stochastic components varies significantly across different age groups.
• Neural noise exhibits distinct spatial distributions across brain regions.

Conclusions:

• Intrinsic stochastic brain dynamics play a significant physiological role in brain function.
• Neural noise increases linearly during human brain maturation.
• The developed framework offers a novel approach to analyzing neural noise in fMRI data.