Input-output relationships of central neural circuits involved in respiration in cats.

1. Inspiratory output responses, measured as integrated phrenic activity, to hypercapnia, to unilateral and bilateral carotid sinus nerve stimulation and to combinations of these stimuli were determined in paralysed, vagotomized and glomectomized cats whose end-tidal P(CO2) was kept constant by means of a servo-controlled ventilator. In addition, the effect on these responses of the mechanism that causes the respiratory after-discharge was determined.2. Above the threshold for rhythmic activity, the inspiratory response to hypercapnic stimulation of the central chemoreceptor was curvilinear, showing progressively smaller increments of output for equal increments of P(CO2) as the latter became higher.3. The combining of stimuli from right and left carotid sinus nerves failed to show an algebraically additive effect; the response was approximately 70% of that predicted from a summing of the separate stimuli given alone.4. The response to a constant carotid sinus nerve test stimulus was progressively decreased in magnitude as the pre-stimulus level of respiratory activity was increased by conditioning stimulation of the central chemoreceptors by hypercapnia, by stimulation of the opposite carotid sinus nerve or by the mechanism that generates an after-discharge.5. From a descriptive standpoint, our findings show that there is a negative or hypoadditive interaction between the peripheral and central inputs at the level of the central respiratory controller. However, we present evidence that, rather than being a specific interaction between peripheral and central inputs, the response is due to the properties of a neural component of the central pathway. This component is common to both inputs and develops progressive saturation of its neural elements as its activity increases.6. In addition, the neural mechanism which generates a respiratory after-discharge appears to saturate completely at a lower level of inspiratory activity than that at which the common pathway develops complete saturation. This finding supports the idea that this mechanism represents an independent input to the respiratory controller.7. Because the described a-linear response characteristics of the central respiratory controller are due to its inherent neuronal properties rather than to specific interactions between inputs, we suggest that studies of such ;interactions' must be interpreted with this consideration in mind.