Processing of Auditory Midbrain Interspike Intervals by Model Neurons |
| |
Authors: | Nathan R. Wilson Deana A. Bodnar Joseph F. Skovira Bruce R. Land |
| |
Affiliation: | (1) Department of Neurobiology and Behavior, Cornell University, Ithaca, New York, USA;(2) IBM Corporation, Armonk, New York, USA |
| |
Abstract: | A question central to sensory processing is how signal information is encoded and processed by single neurons. Stimulus features can be represented through rate coding (via firing rate), temporal coding (via firing synchronization to temporal periodicities), or temporal encoding (via intricate patterns of spike trains). Of the three, examples of temporal encoding are the least documented. One region in which temporal encoding is currently being explored is the auditory midbrain. Midbrain neurons in the plainfin midshipman generate different interspike interval (ISI) distributions depending on the frequencies of the concurrent vocal signals. However, these distributions differ only along certain lengths of ISIs, so that any neurons trying to distinguish the distributions would have to respond selectively to specific ISI ranges. We used this empirical observation as a realistic challenge with which to explore the plausibility of ISI-tuned neurons that could validate this form of temporal encoding. The resulting modeled cells—point neurons optimized through multidimensional searching—were successfully tuned to discriminate patterns in specific ranges of ISIs. Achieving this task, particularly with simplified neurons, strengthens the credibility of ISI coding in the brain and lends credence to its role in auditory processing. |
| |
Keywords: | temporal coding interspike intervals model neuron high-dimensional parameter optimization auditory midbrain |
本文献已被 PubMed SpringerLink 等数据库收录! |
|