Modeling synaptic interactions between mammalian breathing and swallowing central pattern generators.

Breathing and swallowing are tightly coupled motor behaviors whose execution depends on the coordinated activity of two brainstem central pattern generators (R-CPG, Sw-CPG). We present a mechanistic neural network model that captures this coordination. The model was built by incrementally expanding the synaptic connectivity of R-CPG and Sw-CPG modules, each composed of neuronal population nodes whose dynamics incorporate firing rate adaptation. The rhythmogenic cores of both CPGs are implemented as half-center oscillators operating independently. In agreement with accompanying experimental recordings from an in situ perfused brainstem preparation, a simulated 10 s sensory drive to the superior laryngeal nerve (SLN) reliably elicits fictive sequential swallowing bursts accompanied by glottal closure and suppression of inspiratory activity. Short SLN stimuli (100 ms) produce isolated single swallows that reset the phase of the respiratory oscillator. Phase space analysis of the model dynamics offers additional insight into this SLN-evoked respiratory phase resetting. Consistent with prior experimental observations, simulated pontine inhibition reproduces apneusis and abolish swallowing-related glottal closure during sequential swallowing. Systematic perturbation of synaptic weights across specific network pathways identifies vulnerable connections whose weakening recapitulates clinically relevant breathing-swallowing disorders, including aspiration. The model thereby generates mechanistic predictions that may inform the design of therapeutic interventions for conditions characterized by impaired breathing-swallowing coordination.