Real-time brain-state-coupled cortico-cortical paired associative stimulation of cognitive networks

Abstract

Brain networks coordinate distributed neuronal assemblies to support cognition. Spike-timing-dependent plasticity (STDP) and neuronal oscillations are key substrates for state-gated learning rules that shape network coupling and cognitive operations; nonetheless, how STDP mechanisms interact with neuronal oscillations is largely unexplored in humans. Cortico-cortical paired associative stimulation (ccPAS) provides a non-invasive system-level model of associative timing rules by pairing dual-site transcranial magnetic stimulation (TMS) across axonally connected regions with an inter-stimulus interval matched to pathway conduction. Here we: 1) synthesize ccPAS applications and barriers to brain-state-coupled implementation in cognitive networks; 2) provide an actionable roadmap for real-time state estimation, targeting, and dual-site parameter selection; and 3) demonstrate a novel implementation of theta-phase-locked fronto-parietal (FP) ccPAS with concurrent EEG in adult human participants. We tested whether ccPAS delivered at the positive phase of ongoing theta (POS) induces distinct changes in evoked EEG activity and FP connectivity compared to phase-uncoupled ccPAS (RAND) and phase-locked single-site prefrontal (PREF) controls. At the evoked level, POS produced a fronto-central polarity reversal of the canonical N45 component and a right parieto-temporal negativity relative to both controls. At the network level, POS induced frequency-specific reconfigurations in post-intervention connectivity beyond either control ingredient alone. Together, these changes in evoked activity and rapid network reconfiguration provide the first empirical evidence consistent with phase-gated STDP in humans--whereby oscillatory phase gates cortical excitability and modulates STDP efficacy--emerging as short-term network-level expression. Future work will assess long-term plasticity by tracking connectivity at later time points and testing for concomitant behavioral effects. SignificanceThe real-time brain state critically shapes how plasticity mechanisms are expressed in response to brain stimulation. This article provides a forward-looking synthesis of the scientific and technical challenges associated with ccPAS--an STDP induction model in the human cortex--and outlines the steps required to advance it toward real-time brain-state-coupled implementation. To our knowledge, this is the first application of brain-state-coupled ccPAS within a cognitive network. By personalizing stimulation to the individuals ongoing neural state, this approach may reduce variability, limit off-target effects, and enhance plasticity induction. Ultimately--by modulating network-level function in a brain-state-dependent manner--this technique could augment therapeutic outcomes in disorders marked by network dysfunction such as ADHD, Alzheimers disease, and major depressive disorder, potentially maximizing efficacy in patients unresponsive to existing treatments.