Inhibition facilitates direction selectivity in a noisy cortical environment

Area of Science:

• Neuroscience
• Computational Neuroscience
• Visual System Modeling

Background:

• Direction selectivity in primary visual cortex is often modeled by integrating spatially offset and temporally lagged inputs with a spike threshold.
• Understanding the impact of noise and background activity on these models is crucial for explaining neural computation.

Purpose of the Study:

• To investigate the robustness of direction selectivity models to uncorrelated input noise and background activity.
• To determine the role of inhibition in maintaining direction selectivity under noisy conditions.

Main Methods:

• Simulating a class of neuronal models with varying levels of excitatory and inhibitory input.
• Analyzing the effects of uncorrelated noise on direction selectivity and trial-by-trial variability.
• Investigating the impact of tuned inhibition, mirroring excitatory inputs with a delay.

Main Results:

• Without inhibition, noise significantly degraded direction selectivity.
• Inclusion of inhibition maintained direction selectivity even at high noise levels.
• Tuned inhibition enhanced direction selectivity by suppressing null-direction inputs and minimally affecting preferred-direction inputs, while also reducing response variability.

Conclusions:

• Inhibition is critical for robust direction selectivity in primary visual cortical neurons.
• Tuned inhibition allows neural circuits to reliably extract stimulus information from noisy feed-forward inputs on a single-trial basis.