| Abstract No.: | B-D2155 |
| Country: | Canada |
| | |
| Title: | INFORMATION TRANSMISSION IN THE VESTIBULAR SYSTEM: CENTRAL VERSUS AFFERENT NEURONS |
| | |
| Authors/Affiliations: | 1 Corentin Massot*; 1 Maurice Chacron; 1 Kathleen E. Cullen;
1 McGill University, Montreal, QC, Canada
|
| | |
| Content: | Objectives: The processing of sensory information in the vestibular system has been traditionally characterized by estimating response gain and phase over several cycles of sinusoidal head rotation. A limitation of this approach is that it averages response variability. However, our recent work (Sadeghi et al. 2007) has shown that some neurons in the peripheral vestibular system (ie. vestibular afferents) encode information using temporal as well as rate codes. Notably regular afferents transmit, on average, two times more information than irregular afferents, despite having significantly lower gains. Here we address whether the neurons in the central vestibular system, which receive input from these two classes of afferents, make use of the information that is coded in spike train variability of the regular afferents. Notably, focus on a population of neurons in the vestibular nuclei (termed vestibular-only (VO) neurons) which receive direct afferent input. These neurons have both descending and ascending projections and have been implicated in the generation of postural responses for motor control as well as the computation of spatial orientation.
Materials and Methods: Extracellular single recordings were made from the vestibular nucleus of the alert macaque monkey. During experimental sessions, the monkey was comfortably seated in a primate chair mounted on a vestibular turntable generating horizontal rotations. Head was kept restrained. Eye movements were recorded using the magnetic search coil technique. Turntable velocity was measured using an angular velocity sensor (Watson Inc.). Each recorded cell was carefully classified as VO cell on the basis of its lack sensitivity to eye movements (saccades, fixation and pursuit) as well as its sensitivity to whole body rotation during both the vestibular-ocular reflex (VOR) and VOR cancellation. VO cells responses were analyzed under several conditions: single frequency sinusoidal rotations and random white noise rotations within the behaviorally relevant frequency range 0–20 Hz and peak velocity at 50deg/s. Cell discharge was analyzed using both gain and information theoretic measures.
Results: The mutual information density of spike trains resulting from the application of random noise stimuli was computed for 8 VO neurons. Information was minimal at low frequencies and increased as a function of frequency. Thus, under a broadband stimulation protocol, VO cells preferentially transmit high frequency over low frequency information. Notably, this behavior is similar to that previously reported for irregular afferents, and contrasts with that of regular afferents which consistently transmit more information for low-frequency rotations. Moreover, the application of random noise stimuli confirmed the results of previous studies (which have used stimulation at single frequency) and showed that gain increases with frequency and further extended this observation to higher frequencies (up to 20 Hz). Our own analysis of responses of the same neurons to sinusoidal stimuli (0.6-16 Hz) resulted in qualitatively similar gains - consistent with the traditional linear system’s approach applied to the vestibular system.
Conclusion: Our experimental results suggest during broadband vestibular stimulation, the central integration of afferent input is tuned for higher frequency events.
|
| | |
| Back |
|
