Task-induced topological and geometrical changes in whole-brain dynamics predict cognitive individual differences.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Cognitive Neuroscience

Background:

• Functional magnetic resonance imaging (fMRI) advances understanding of cognition's neural basis.
• The relationship between resting-state fMRI (rsfMRI) and task-based fMRI (tfMRI) and its link to cognitive function remains unclear.

Purpose of the Study:

• To test a computational model linking rsfMRI and tfMRI dynamics.
• To investigate how task contexts modulate brain dynamics and individual cognitive differences.

Main Methods:

• Developed Mesoscale Individualized NeuroDynamics with eXogenous inputs (MINDy-X) framework.
• Applied MINDy-X to resting and N-back working memory task data from the Human Connectome Project.
• Modeled and analyzed joint rsfMRI-tfMRI data to understand brain dynamics.

Main Results:

• The MINDy-X model accurately simulated and predicted both rsfMRI and tfMRI data.
• Task performance shifted brain dynamics from multistable to monostable states, altering attractor landscapes.
• Individual differences in dynamic shifts correlated with N-back task performance, error rates, and response caution.

Conclusions:

• rsfMRI and tfMRI reflect different states of a common nonlinear dynamical system.
• Cognitive function can be characterized by the geometry and topology of brain attractor landscapes.
• This framework offers new insights into brain activity patterns and individual variability in neuroscience research.