Balancing the Robustness and Efficiency of Odor Representations during Learning

Area of Science:

• Neuroscience
• Sensory Coding
• Olfactory System

Background:

• Sensory codes require a balance between robustness (to combat noise) and efficiency (to maximize information transmission).
• Experience and learning are known to modify neural representations, but how this impacts the robustness-efficiency trade-off in sensory coding remains unclear.

Purpose of the Study:

• To investigate how experience, specifically during olfactory learning, alters the balance between robustness and efficiency in odor representations within the mouse olfactory bulb.
• To determine if the nature of the learned odorants (similar vs. dissimilar) influences the direction of these representational changes.

Main Methods:

• Longitudinal two-photon calcium imaging was used to monitor the activity of mitral cell ensembles in the mouse olfactory bulb over a one-week learning period.
• Mice were trained to discriminate between either two similar or two dissimilar odorants, or experienced them passively.

Main Results:

• Learning to discriminate dissimilar odorants led to less discrete mitral cell ensemble responses, increasing coding efficiency.
• Learning to discriminate similar odorants resulted in decorrelated representations, enhancing coding robustness.
• Similar representational adjustments were observed during passive odorant experience, suggesting implicit perceptual learning.

Conclusions:

• Experience dynamically adjusts odor representations in the olfactory bulb to optimize the balance between robustness and efficiency.
• The specific changes in neural coding are dependent on the perceptual relevance and similarity of the experienced odorants.
• These findings provide insights into the adaptive nature of sensory processing and perceptual learning.