A distributed, dynamic, parallel computational model: the role of noise in velocity storage
ABSTRACT Networks of neurons perform complex calculations using distributed, parallel computation, including dynamic "real-time" calculations required for motion control. The brain must combine sensory signals to estimate the motion of body parts using imperfect information from noisy neurons. Models and experiments suggest that the brain sometimes optimally minimizes the influence of noise, although it remains unclear when and precisely how neurons perform such optimal computations. To investigate, we created a model of velocity storage based on a relatively new technique--"particle filtering"--that is both distributed and parallel. It extends existing observer and Kalman filter models of vestibular processing by simulating the observer model many times in parallel with noise added. During simulation, the variance of the particles defining the estimator state is used to compute the particle filter gain. We applied our model to estimate one-dimensional angular velocity during yaw rotation, which yielded estimates for the velocity storage time constant, afferent noise, and perceptual noise that matched experimental data. We also found that the velocity storage time constant was Bayesian optimal by comparing the estimate of our particle filter with the estimate of the Kalman filter, which is optimal. The particle filter demonstrated a reduced velocity storage time constant when afferent noise increased, which mimics what is known about aminoglycoside ablation of semicircular canal hair cells. This model helps bridge the gap between parallel distributed neural computation and systems-level behavioral responses like the vestibuloocular response and perception.
- SourceAvailable from: Fred W Mast
[Show abstract] [Hide abstract]
- "(2) Offline use of forward models: In order to explore the relationship between sensory anticipation and higher vestibular processing, we propose to use a motion discrimination task, for which the vestibular afferent signals are relatively well understood, such as rotation about the earth's vertical axis. Detailed forward models of semi-circular canal afferent signals have been developed (see Karmali and Merfeld, 2012); the fact that this problem can be described at this level of detail would make self-motion an ideal task to build upon. In order to test our claims that mental imagery and cognitive processes, such as spatial perspective taking, are based on the offline usage of forward models, it is necessary to first determine the effect of expectation, for example in a leftward/rightward rotation discrimination task. "
ABSTRACT: Vestibular cognition has recently gained attention. Despite numerous experimental and clinical demonstrations, it is not yet clear what vestibular cognition really is. For future research in vestibular cognition, adopting a computational approach will make it easier to explore the underlying mech- anisms. Indeed, most modeling approaches in vestibular science include a top-down or a priori component. We review recent Bayesian optimal observer models, and discuss in detail the conceptual value of prior assumptions, likelihood and posterior estimates for research in vestibular cognition. We then consider forward models in vestibular processing, which are required in order to distinguish between sensory input that is induced by active self-motion, and sensory input that is due to passive self-motion. We suggest that forward models are used not only in the service of estimating sensory states but they can also be drawn upon in an offline mode (e.g., spatial perspective transformations), in which interaction with sensory input is not desired. A computational approach to vestibular cogni- tion will help to discover connections across studies, and it will provide a more coherent framework for investigating vestibular cognition.Multisensory research 04/2015; 28(5-6). DOI:10.1163/22134808-00002503
- [Show abstract] [Hide abstract]
ABSTRACT: To investigate the characteristics of eye movements produced by electrical stimulation of semicircular canal afferents, we studied the spatial and temporal features of eye movements elicited by short-term lateral canal stimulation in two squirrel monkeys with plugged lateral canals, with the head upright or statically tilted in the roll plane. The electrically induced vestibuloocular reflex (eVOR) evoked with the head upright decayed more quickly than the stimulation signal provided by the electrode, demonstrating an absence of the classic velocity storage effect that improves the dynamics of the low-frequency VOR. When stimulation was provided with the head tilted in roll, however, the eVOR decayed more rapidly than when the head was upright, and a cross-coupled vertical response developed that shifted the eye's rotational axis toward alignment with gravity. These results demonstrate that rotational information provided by electrical stimulation of canal afferents interacts with otolith inputs (or other graviceptive cues) in a qualitatively normal manner, a process that is thought to be mediated by the velocity storage network. The observed interaction between the eVOR and graviceptive cues is of critical importance for the development of a functionally useful vestibular prosthesis. Furthermore, the presence of gravity-dependent effects (dumping, spatial orientation) despite an absence of low-frequency augmentation of the eVOR has not been previously described in any experimental preparation.Journal of Neurophysiology 06/2012; 108(5):1511-20. DOI:10.1152/jn.01029.2011 · 3.04 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: To assess the contributions of the vestibular system to whole-body motion discrimination in the dark, we measured direction recognition thresholds as a function of frequency for yaw rotation, superior-inferior translation ("z-translation"), interaural translation ("y-translation"), and roll tilt for 14 normal subjects and for 3 patients following total bilateral vestibular ablation. The patients had significantly higher average threshold measurements than normal (p < 0.01) for yaw rotation (depending upon frequency, 5.4× to 15.7× greater), z-translation (8.3× to 56.8× greater), y-translation (1.7× to 4.5× greater), and roll tilt (1.3× to 3.0× greater)-establishing the predominant contributions of the vestibular system for these motions in the dark.The Journal of Neuroscience : The Official Journal of the Society for Neuroscience 09/2012; 32(39):13537-42. DOI:10.1523/JNEUROSCI.2157-12.2012 · 6.75 Impact Factor