Neuronal coding of linear motion in the vestibular nuclei of the alert cat

Research paper by J. Barthelemy, C. Xerri, L. Borel, M. Lacour

Indexed on: 01 Apr '88Published on: 01 Apr '88Published in: Experimental Brain Research


Extracellular activity from vestibular nuclei neurons and vertical eye movements were recorded in the alert cat during sinusoidal optokinetic stimulation in the vertical plane at frequencies varying from 0.0125 Hz to 0.75 Hz. Among a population of 96 vestibular units located in and around Deiters' nucleus, 73 neurons (76%) displayed a firing rate modulation which followed the input at the standard parameters of visual stimulation (0.05 Hz; 10.1 deg/s or 9.1 cm/s peak to peak velocity). Two different patterns of modulation were found. In 42 cells (57%) an increase in the firing rate was observed during motion of the visual scene in the downward direction, while 31 neurons (43%) showed the opposite behavior, with an enhanced firing rate during upward movement. The phase of the neuronal responses was close (± 45°) to the velocity peaks (+90°: downward and -90°: upward) of visual scene motion for 65 among the 73 neurons. Mean values of phase was-6.1 ± 19.5° (SD) and -3.2 ± 15.5° (SD) with respect to the +90° and -90° velocity peaks, respectively. In the frequency range 0.0125–0.75 Hz, the phase of the neuronal responses remained almost stable, with only a slight lag which reaches -22° at the 0.25 Hz visual stimulation. The firing rate modulation was found to be predominant at low frequencies (0.0125 Hz–0.25 Hz), with three distinct peaks of modulation occurring either at 0.025 Hz, 0.10 Hz or 0.25 Hz, depending on the recorded cells. Above 0.5 Hz, the cell modulation was very poorly developed or even absent. A gain attenuation was observed in all units, which was more important in cells showing a peak of modulation at 0.025 Hz as compared with the others (-20.7 dB vs -9.6 dB, respectively, in the 0.025 Hz–0.25 Hz decade). The gain of the optokinetic reflex (OKR) progressively decreased from mean values of 0.78 ± 0.15 to 0.05 ± 0.06 in the 0.025 Hz–0.5 Hz frequency range. A close correlation was observed between the OKR slow phase velocity and the modulation of the neuronal responses in the two cell populations with maximal modulations at 0.10 Hz or 0.25 Hz. No correlations were noticed in the third population characterized by a peak of modulation at 0.025 Hz. In all units, the phase of eye movement velocity and of neuronal responses were both related to the velocity of the visual surround motion. These correlations were also found when varying the amplitude of the visual stimulation at a fixed frequency. Saturation was observed in the unit responses at velocities above 68.5°/s. When considering both the gain attenuation in the frequency range and the correlation between firing rate modulation and OKR slow phase velocity, two rather different cell populations can be distinguished: one with neurons peaking at 0.025 Hz (strong gain attenuation; no correlation with OKR velocity) and one with neurons peaking at 0.10 Hz or 0.25 Hz (slight gain attenuation; correlation with OKR velocity). This study points to the influence of visual motion cues on vestibular nuclei unit activity in the low-frequency range. A velocity coding of visual — surround motion in the vertical plane is performed by vestibular neurons. Our results in the alert cat suggest that both retinal (retinal slip) and extraretinal (proprioceptive afferences from eye muscles, efference copy) inputs can be involved in this visually induced modulation of vestibular nuclei neurons.