1. Using single-unit recording and microstimulation methods, a group of flocculus target neurons (FTNs) were identified in the superior vestibular nucleus (SVN) and were studied using visual-vestibular interaction paradigms in alert squirrel monkeys. The response properties of these FTNs were characterized and compared with those of flocculus projecting neurons (FPNs). 2. FTNs were monosynaptically inhibited by single-pulse flocculus stimulation. The mean inhibition latency was 1.0 +/- 0.57 (SD) ms (n = 40) and the mean inhibition period was 6.7 +/- 2.69 ms. FTNs were also monosynaptically activated by VIIIth nerve stimulation. The mean response latency was 1.10 +/- 0.25 ms (n = 12). This is about the same as that of the FPNs (1.14 +/- 0.16 ms, n = 17). 3. The most characteristic response property of the FTNs is their firing rate modulation during visual following eye movements induced by sinusoidal rotation of an optokinetic drum at 0.5 Hz. This modulation was mainly related to eye velocity and was therefore termed a visual following eye velocity signal. The average eye velocity gain for all FTNs is 0.79 spikes . s(-1). deg(-1). s(-1). In contrast, the responses of FPNs were not modulated under the same conditions. 4. Even though FTNs are inhibited by the flocculus, they have a relatively higher mean firing rate (124 +/- 23 spikes/s, n = 45) than FPNs (66 +/- 28 spikes/s, n = 42). The underlying mechanism may be related to commissural facilitation of FTNs and commissural inhibition of FPNs. 5. Thirty FTNS were identified as upward eye velocity FTNs because their firing rate increased for upward eye velocity during a visual following eye movement. The mean eye velocity sensitivity was 1.09 spikes . s(-1)deg(-1). s(-1). Most of these cells also modulated during vestibuloocular reflex (VOR) in the dark, with firing rate increasing for downward head velocity. During VOR suppression the firing rate either did not modulate or modulated in phase with head or drum velocity with a smaller amplitude in comparison with the response during visual following. For all cells (with 1 exception) the response during a visual following eye movement can be approximately predicted by a linear vectorial subtraction of the response during VOR suppression and the response during VOR in the dark [modulation response vector of FTNs during visual following of the optokinetic stimulus (OKR) approximate to modulation response vector of FTNs during VOR suppression - modulation response vector of FTNs during VOR in the dark]. This leads to the hypothesis that the FTN's response can be synthesized by an excitatory VIIIth nerve head velocity signal and an inhibitory flocculus (FL) gaze velocity signal. The response properties of the cells were determined by the ratio of these two signals. Cells with similar amplitudes from the two signals modulated with similar modulation amplitudes during visual following and VOR in the unequal amplitudes of these two signals modulated for all three paradigms. Cells with little or no VIIIth nerve input modulated during visual following and VOR suppression, but not during VOR in the dark. This last type of gaze velocity like FTNs is in the minority in the population. During steady fixation the eye position sensitivity was relatively low, and the ON direction could be the same or opposite to that of the eye velocity ON direction. 6. Upward eye velocity FTNs were identified to be vestibulooc ular neurons by antidromic activation and collision tests through oculomotor nucleus microstimulation. The mean antidromic response latency was 0.58 +/- 0.07 ms (n = 10). 7. Eleven cells were identified as downward eye FTNs, because their firing rate increased for downward eye velocity during visual following eye movement and for downward eye position during steady fixation. Their eye velocity sensitivity was much lower than that of the upward eye velocity FTNs. During VOR in the dark their firing rate increased with upward head velocity. During VOR suppression their firing rate could not be predicted by a linear summation of the response during the other two paradigms, which might due to the small signal-to-noise ratio and small sampling number. During steady fixation the eye position sensitivity was relatively high (-2.34 +/- 1.66 spikes/s) in comparison with upward eye velocity FTNs. These downward eye velocity FTNs were not tested with oculomotor nucleus stimulation, but considering their large eye position sensitivity we believe they are also vestibuloocular neurons. 8. Anatomically, FTNs are located posterior to the FPNs within the SVN. Downward eye FTNs were recorded much less frequently and were always located ventral to the upward eye velocity FTNs.
