Kathleen E. Cullen

Responses of vestibular and prepositus neurons to head movements during voluntary suppression of the vestibuloocular reflex.

Neurons in the vestibular nuclei and the prepositus nucleus exhibited several different types of changes in their firing behavior during voluntary cancellation of the horizontal VOR. The head velocity sensitivity of type I position-vestibular-pause neurons was reduced during cancellation, while type II vestibular neurons exhibit an increase in their sensitivity. The firing behavior of burst tonic neurons in the medial vestibular nucleus, the prepositus nucleus, like the cells in the abducens nucleus, was closely related to the eye movements generated when the VOR is cancelled. Other cells in the PH and MVN respond primarily to smooth pursuit eye movements. We suggest that the behavior of abducens neurons during the VOR and during VOR cancellation can be explained if they receive inputs from PVP neurons, burst tonic neurons, and smooth pursuit neurons.

Participation of Secondary Vestibular Neurons in Nonvisual Mechanisms of Vestibuloocular Reflex Cancellation

Our recent studies have suggested that squirrel monkeys can utilize a fast, nonvisual mechanism for canceling their vestibuloocular reflex (VOR) when they are fixating a visual target and their head is moving. We have investigated the neural basis of this nonvisual suppression of the VOR by recording from neurons in the vestibular nuclei in alert, trained squirrel monkeys during smooth pursuit and cancellation of the VOR (VORC). Secondary vestibular neurons were identified by their short latency (0.9-1.3 mseconds) response to electrical stimulation of the vestibular nerve. They were further classified on the basis of their spiking behavior during horizontal sinusoidal smooth pursuit eye movements and VOR suppression. Two main types of ipsilateral head movement sensitive secondary neurons were identified: (1) neurons sensitive only to horizontal head movements (pure vestibular) and (2) neurons whose firing increased during contralateral smooth pursuit and paused during saccades [positionvestibular-pause (PW) neurons]. The head velocity sensitivity of the neurons was investigated during two types of vestibular stimulation, sinusoidal head rotations and unpredictable steps in head acceleration. The head movement sensitivity of most pure vestibular neurons (n = 20) did not change when squirrel monkeys canceled their VOR during sinusoidal turntable rotation [mean response gain (0.8 sp/second)/(deg/second)]. The data from one of these units, averaged over 10 cycles of head rotation, is shown in FIGURE 1A. In contrast, the head movement sensitivity of most PVP neurons (n = 19) was significantly attenuated during cancellation of the VOR (mean attenuation in response gain 30%). The response of a typical P W neuron during sinusoidal VORC and VOR (corrected for eye position and desaccaded) was averaged over 10 cycles of head rotation and is displayed in FIGURE 1B. The modulation of this neuron decreased by more than 50% during VORC. The amount of attenuation in the head velocity sensitivity of P W neurons was poorly correlated with the eye velocity signal of the neuron during smooth pursuit, suggesting that the smooth pursuit eye velocity sensitivity of the neuron is not responsible for this attenuation head velocity sensitivity of the P W neurons. The amplitude of the response of most PVP neurons to unexpected passive perturbations of the head (400/second2 head acceleration steps) was attenuated in trials during which the monkeys were canceling their VOR, while the response of the pure vestibular neurons was the same regardless of the monkeys behavior. The short latency suppression of the VOR, which is generated by an unpredictable head acceleration step when the squirrel monkeys were initially canceling their VOR, is demonstrated in FIGURE 2A. The averaged population responses (corrected for each