Models of the medullary VRC include rostral inspiratory (I) neurons that excite more caudal I neurons, including premotor populations, as well as inhibitory I neurons that act via recurrent and feed-forward connections upon other excitatory and inhibitory I and expiratory (E) neuron populations (1). Our prior work supported the hypotheses that tonic E (t-E) neuron inhibition of premotor I neurons contributes to baroreceptor-evoked reductions in I drive (2), and that central chemoreceptor stimulation enhances drive, in part via reduced I-phase t-E neuron activity and the resulting disinhibition of premotor neurons (3). Aims of the present work were to identify further the functional connections of t-E neurons and test the hypothesis that carotid peripheral chemoreceptors also modulate drive by I-phase inhibition of t-E neurons. Spike trains were acquired by multi-electrode arrays along with signals from phrenic and vagus nerves (10 Hz-10 kHz) from 22 adult cats. Animals were anesthetized with isoflurane mixed with air (induction: 5% ; maintenance: 0.5-3.0%) until decerebration, neuromuscularly blocked (pancuronium bromide; initial bolus 0.1 mg kg−1 followed by 0.2 mg kg−1 hr−1, iv;) and monitored (4), vagotomized, and artificially ventilated. Arterial blood pressure, end-tidal CO2, tracheal pressure, and arterial PO2, PCO2, and pH were monitored. In 6 recordings from 5 animals, yielding data from 219 neurons (4,374 distinct cell pairs), carotid chemoreceptors were stimulated (5 trials) by close 30-s injections of 1.0 mL of a CO2 -saturated 0.9% saline solution. Responses were identified using a bootstrap-based statistical method; p-value threshold was set with a false discovery rate of 0.05 (4). Cross-correlogram features were identified using cycle-shifted surrogates (3) or a Monte-Carlo test with surrogate spike trains (5) generated with a gamma distribution shape parameter; false discovery rate < 0.05. Features in spike-triggered averages of full-wave rectified phrenic signals were identified (3) using a two-sided Wilcoxon signed-rank test (Bonferroni correction; p < 0.05). Overall, from 171 peri-columnar t-E neurons we identified 42 of 599 pairs with cross-correlogram central peaks (detectability index (di) = 13.0 ± 11.8 SD; half-width (hw) = 18.5 ms ± 25.7). Results from the stimulation experiments included assemblies with 8 correlogram troughs indicative of t-E neuron inhibition of I neurons (di = 7.9 ± 2.9; hw = 10.8 ms ± 9.6) and disinhibitory enhancement of I drive. When considered with results from simulations of a computational neuromechanical model for cough (1) and associated in vivo data, we conclude that multiple afferent systems and behaviors exert a “push-pull” control of I drive through modulation of a coordinated network of t-E neurons.