Probabilistic Cognitive State Modeling (PCSM): Decoding Latent Spatiotemporal Dynamics to Reveal Serial-Parallel...

Studying flexible, adaptive transitions between cognitive tasks and serial-parallel processing under changing task demands has been a central focus for understanding human cognition. Advances in neuroimaging analysis have improved the ability to link cognition with brain function, providing a foundation for developing methods that capture brain dynamics for quantitating emergent cognitive properties during task-based fMRI paradigms. Probabilistic Cognitive State Modeling (PCSM) combines Finite Impulse Response (FIR) modeling of BOLD activity with a Gaussian Mixture Model-Hidden Markov Model (GMM-HMM) to quantify dynamic, spatially distributed brain states. From these posteriors, PCSM derives interpretable, emergent properties including serial-parallel processing, cognitive demand, resource level, and serial bottleneck. A ground-truth simulation evaluated PCSM across systematically varied noise and transition probabilities. Results show that PCSM accurately recovers ground-truth latent states (> 98 %) and produces stable parameter estimates across conditions. Threshold analyses identify reliable boundaries between parallel, mixed, and serial processing modes and recover expected relations among demand, resource availability, and bottleneck persistence. These findings demonstrate that PCSM can reveal how dynamic brain states contribute to adaptive processing architectures, providing a framework for mapping individual-specific cognitive dynamics and examining cognitive processing, demand, and serial bottleneck in task-based fMRI.