Online stimulus optimization rapidly reveals multidimensional selectivity in auditory cortical neurons.

Area of Science:

• Neuroscience
• Auditory Perception
• Computational Neuroscience

Background:

• Neurons in sensory brain regions perform decomposition and integration of stimuli.
• Auditory neurons encode sound features like frequency, temporal modulation, intensity, and location.
• Higher sensory areas integrate features for a unified perception.

Purpose of the Study:

• To characterize how auditory cortical neurons decompose and integrate multiple sound facets.
• To develop an automated procedure for real-time acoustic stimulus manipulation.
• To investigate the multidimensional receptive fields of auditory neurons.

Main Methods:

• Developed an automated procedure to manipulate five acoustic properties in real time.
• Used single-unit feedback from awake mice to guide stimulus optimization.
• Cross-validated optimized stimuli against pure tone and spectrotemporal receptive fields.

Main Results:

• The online approach rapidly converged on stimuli that maximally drove neurons.
• Observed increased level invariance and frequency selectivity from midbrain to cortex.
• Found proportional increases in onset and steady-state spike rates with optimized stimuli.

Conclusions:

• The findings reveal interdependencies between sound features in neural processing.
• Demonstrated hierarchical shifts in neural selectivity and invariance.
• Highlighted the limitations of traditional approaches in capturing complex neural representations.