Internal Representations Are Prioritized by Frontoparietal Theta Connectivity and Suppressed by alpha Oscillation Dynamics: Evidence from Concurrent Transcranial Magnetic Stimulation EEG and Invasive EEG

Switching state to engage and sustain attention: Dynamic synchronization of the frontoparietal network

Sustained attention (SA) is essential for maintaining focus over time, with disruptions linked to various neurological and psychiatric disorders. The oscillatory dynamics and functional connectivity in the dorsal frontoparietal network (dFPN) are crucial in SA. However, the neuronal mechanisms that control the level of SA, especially in response to heightened attentional demands, remain poorly understood. To examine the role of rhythmic synchronization in the dFPN in SA, we recorded local field potential and single unit activity in ferrets that performed the 5-Choice Serial Reaction Time Task (5-CSRTT) under both low and high attentional load. Under high attentional load, dFPN exhibited a pronounced state shift that corresponded with behavioral changes in the animal. Prior to the onset of the target stimulus, animals transitioned from a stationary state, characterized by frontal theta oscillations and dFPN theta connectivity, to an active exploration state associated with sensory processing. This shift was indexed by a suppression of inhibitory alpha oscillations and an increase in excitatory theta and gamma oscillations in parietal cortex. We further show that dFPN theta connectivity predicts performance fluctuations under high attentional load. Together, these results suggest that behavioral strategies for maintaining SA are tightly linked to neuronal state dynamics in the dFPN. Importantly, these findings identify rhythmic synchronization within the FPN as a potential neural target for novel therapeutic strategies for disrupted attention.

Linking the multiple-demand cognitive control system to human electrophysiological activity

The frontoparietal multiple-demand (MD) network serves as a core system for domain-general cognitive control, with robust activation with increased demand across diverse tasks. While fMRI studies have characterised the MD network's role in cognitive demand, linking these findings to electrophysiological activity remains a critical challenge. This article discusses the potential of oscillatory and aperiodic neural activity to bridge this gap. Although recent meta-analyses highlight mid-frontal theta power as a robust marker of task demand, its localised spatial distribution, limited cross-task generalisability, and potential confounds from aperiodic components limit its ability to fully represent the MD network. In contrast, aperiodic activity, particularly broadband power, has emerged as a strong candidate for indexing task demand due to its robust decoding performance and cross-task generalisability in response to diverse task demands, and spatial overlap with MD regions. Aperiodic activity may reflect fundamental neural properties, such as spiking rates and excitation/inhibition (E/I) balance, and is scale-free and exists across modalities, positioning it as a promising mechanism underpinning domain-general cognitive control that links to the MD network. Meanwhile, multiplexed low-frequency oscillations (e.g., delta and theta) may implement inter-regional synchronisation within the MD network, enabling large-scale coordination between MD subregions that supports cognitive control. Together, this article proposes a hypothetical framework linking the MD network to electrophysiological responses: Aperiodic broadband power, potentially reflecting population-level spiking activity, may support activation within MD regions, while multiplexed low-frequency oscillatory synchronisations may mediate inter-regional connectivity between MD regions.

Multimodal neuroimaging of hierarchical cognitive control

Cognitive control enables us to translate our knowledge into actions, allowing us to flexibly adjust our behavior, according to environmental contexts, our internal goals, and future plans. Multimodal neuroimaging and neurostimulation techniques have proven essential for advancing our understanding of how cognitive control emerges from the coordination of distributed neuronal activities in the brain. In this review, we examine the literature on multimodal studies of cognitive control. We explore how these studies provide converging evidence for a novel, multiplexed model of cognitive control, in which neural oscillations support different levels of control processing along a functionally hierarchical organization of distinct frontoparietal networks.