Laryngeal adductor responses to afferent stimulation play a key role in airway protection. Although vital for protection during cough and swallow, these responses also must be centrally controlled to prevent airway obstruction by laryngospasm during prolonged stimulation. Our purpose was to determine the role of N-methyl-D-aspartate (NMDA) receptors in modulating early R1 responses (at 9 ms) and/or later more prolonged R2 responses (at 36 ms) during electrical stimulation of the laryngeal afferent fibers contained in the internal branch of the superior laryngeal nerve in the cat. The percent occurrence, amplitude, and conditioning of muscle responses to single superior laryngeal nerve (SLN) stimuli presented in pairs at interstimulus intervals of 250 ms were measured in three experiments: 1) animals that had ketamine as anesthetic premedication were compared with those who did not, when both were maintained under alpha-chloralose anesthesia. 2) The effects of administering ketamine in one group of animals were compared with increasing the depth of alpha-chloralose anesthesia without NMDA receptor blockade in another group of animals. 3) The effects of dextromethorphan (without anesthetic effects) were examined in another group of animals. In the first experiment, the occurrence of R2 responses were reduced from 95% in animals without ketamine premedication to 25% in animals with ketamine premedication (P = 0.015). No differences occurred in the occurrence, amplitude, latency, or conditioning effects on R1 responses between these groups. In the second experiment, the occurrence of R2 responses was reduced from 96 to 79% after an increase in the depth of anesthesia with alpha-chloralose in contrast with reductions in R2 occurrence from 98 to 19% following the administration of ketamine to induce NMDA receptor blockade along with increased anesthesia (P = 0.025). In the third experiment, R2 occurrence was reduced from 89 to 27% (P = 0.017) with administration of dextromethorphan while R1 response occurrence and amplitude did not change. In each of these experiments, NMDA receptor blockade did not have significant effects on cardiac or respiratory rates in any of the animals. The results demonstrate that NMDA receptors play an essential role in long latency R2 laryngeal responses to laryngeal afferent stimulation. On the other hand, early R1 laryngeal adductor responses are likely to involve non-NMDA receptor activation.