Physiological models of the lateral superior olive

Area of Science:

• Computational Neuroscience
• Auditory System Modeling
• Biophysics

Background:

• Neuron models in computational biology often lack generalizability due to limited empirical data fitting.
• Auditory neurons in the lateral superior olive (LSO) are crucial for computing binaural sound localization cues (ITD/ILD).
• LSO neurons integrate excitatory and inhibitory inputs to process interaural acoustic differences.

Purpose of the Study:

• To investigate the credibility and generalizability of different computational models of LSO neurons.
• To compare functional ('shot-noise') and biophysically detailed (integrate-and-fire, Hodgkin-Huxley) LSO models.
• To assess model performance based on physiological detail, computational efficiency, and expandability.

Main Methods:

• Examined seven contemporary computational models of LSO neurons with varying complexity.
• Calibrated models to reproduce known monaural and binaural characteristics of LSO.
• Compared model outputs for simulating interaural time and level difference (ITD/ILD) coding.

Main Results:

• Despite differing complexity, tested LSO models produced largely similar results in simulating ITD/ILD coding.
• Simple, functional models offer computational efficiency and mathematical transparency.
• Complex models are essential for simulating sub-neuronal nonlinear processes.

Conclusions:

• The choice of LSO neuron model depends on the specific application requirements.
• Simple models are suitable for low-computational-cost applications.
• Intermediate models provide a balance between simplicity and biological plausibility for general use.