A flexible cross-correlation based population model of interaural time difference coding in barn owl's midbrain.

Area of Science:

• Computational Neuroscience
• Auditory System Modeling
• Bioacoustics

Background:

• The standard cross-correlation model explains barn owl auditory processing at the nucleus laminaris but fails to capture complex interaural time difference (ITD) tuning in the inferior colliculus (IC).
• Neurons in the IC exhibit sharper-than-sinusoidal ITD tuning, nonlinear frequency integration, and interaural level difference (ILD)-dependent modulation, which are not explained by the existing model.

Purpose of the Study:

• To develop a modified cross-correlation model that accounts for the diverse ITD tuning characteristics observed in the barn owl's inferior colliculus (IC).
• To investigate the roles of parameterized gain control, inhibitory filters, nonlinearities, and ILD-dependent modulation in shaping ITD selectivity.

Main Methods:

• Introduced parameterized gain control, linear filters with inhibitory surround, static nonlinearities, and ILD-dependent modulation into the cross-correlation framework.
• Utilized simulation-based inference for efficient parameter determination and generation of realistic neuronal populations.

Main Results:

• Divisive gain control generated realistic rate-level functions, including non-monotonic responses.
• Inhibitory filters and nonlinearities produced sharper-than-sinusoidal ITD tuning consistent with experimental data.
• The model successfully reproduced linear and nonlinear frequency integration and ILD-dependent shifts in best ITD and tuning strength.

Conclusions:

• The modified cross-correlation model provides a more comprehensive framework for understanding ITD processing in the barn owl's auditory midbrain.
• This flexible and analytically tractable model can be used to investigate population coding of auditory space.