FIRING BEHAVIOR OF BRAIN-STEM NEURONS DURING VOLUNTARY CANCELLATION OF THE HORIZONTAL VESTIBULOOCULAR REFLEX .1. SECONDARY VESTIBULAR NEURONS

1. The single-unit activity of vestibular neurons was recorded in aler t squirrel monkeys. The monkeys had been trained to track a small visu al target by generating smooth pursuit eye movements and to cancel the ir vestibuloocular reflex (VOR) by fixating a head stationary target. The monkeys were seated on a vestibular turntable, and their heads wer e held in the plane of the horizontal semicircular canals. The respons es of 45 type I vestibular neurons whose activity was related to ipsil ateral horizontal head movements were recorded. In 19 of 30 cells test ed, electrical stimulation (0.1-ms monophasic pulses, less-than-or-equ al-to 800 muA) of the ipsilateral vestibular nerve evoked a spike at a monosynaptic latency (0.7-1.3 ms). 2. The spiking behavior of each ce ll was recorded during several behavioral paradigms: 1) spontaneous ey e movements, 2) horizontal smooth pursuit of a target that was moved s inusoidally +/- 20-degrees-/s at 0.5 Hz, 3) horizontal VOR during 0.5- Hz sinusoidal turntable rotations +/- 40-degrees/s (VOR(s)), and 4) vo luntary cancellation of the sinusoidal VOR by fixation of a head-stati onary target during 0.5-Hz sinusoidal turntable rotation at +/- 400-de grees/s in the light (VORC(s)). 3. The response of most (34) of the un its was recorded during unpredictable 100-ms steps in head acceleratio n (400-degrees/s2) that were generated while the monkey was fixating a target light. The acceleration steps were generated either when the m onkey was stationary (VOR(t) paradigm) or when the turntable was alrea dy rotating, and the monkey was canceling its VOR (VORC(t) paradigm). Smaller eye movements were evoked when the acceleration step was gener ated during VOR cancellation. 4. Type I vestibular units were grouped into two classes on the basis of the relationship of their firing rate to eye movements. The discharge rate of 20 ''pure vestibular'' units was not clearly related to eye movements. The remaining 25 units were classified as position-vestibular-pause (PVP) neurons. PVP neurons inc reased their firing rate during contralateral eye movements and during ipsilateral turntable rotations, and paused during saccadic eye movem ents. 5. Most (17/20) pure vestibular neurons generated the same respo nse to vestibular stimuli when the monkeys canceled their VOR as they did during the VOR in both the sinusoidal and acceleration step paradi gms. 6. The head velocity sensitivity of most (19/24) PVP neurons was reduced by 20-60% during VORC(s), compared with their response during the VOR(s). The PVP neurons whose sensitivity of head movements was re duced during VORC(s) also exhibited a reduced vestibular sensitivity d uring VORC(t). The reduction in sensitivity during VORC(t) was apparen t at a short latency (<30 ms) after the initiation of the head acceler ation step. On the other hand, the earliest change in firing rate that could be observed when the target was accelerated instead of the turn table was almost-equal-to 80 ms, which suggests that visual signals we re not responsible for the reduction in head movement sensitivity. The response of PVPs during VORC(t) could be modeled by subtracting a low -pass filtered head velocity signal with a gain of 0.5 from their VOR( t) response. 7. The reduction in vestibular sensitivity of PVPs was po sitively but poorly correlated with the smooth pursuit eye velocity se nsitivity in individual units. In fact, several of the cells that show ed a reduction in vestibular sensitivity during VORC(s) and VORC(t) we re not sensitive to eye velocity during smooth pursuit. Thus it is unl ikely that their reduced vestibular sensitivity during VOR cancellatio n was solely due to eye movement-related inputs. 8. Because previous s tudies have shown that PVP neurons are secondary vestibular neurons th at project to extraocular motor neurons and presumably mediate the ves tibuloocular reflex, it is likely that the reduction in head movement sensitivity observed in these cells during VORC contributes to their a bility to voluntarily cancel their VOR. We suggest that the reduction in the head movement sensitivity of secondary vestibuloocular reflex p athways is part of a nonvisual mechanism that can be utilized to volun tarily suppress the VOR.