The Statistical Structure of the Hippocampal Code for Space as a Function of Time, Context, and Value

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Hippocampal activity encodes diverse variables like location, context, time, and reward.
• Understanding the principles governing hippocampal representations is crucial for neuroscience.

Purpose of the Study:

• To identify a unified feature that explains CA1 neuronal activity across different behavioral contexts.
• To investigate the stability and origins of this feature.

Main Methods:

• Longitudinal calcium imaging in mice across multiple virtual environments.
• Varying reward contingencies and analyzing neuronal activity.
• Intracellular recordings to assess cell excitability.

Main Results:

• A unified probabilistic model of CA1 representations was developed, centered on 'field propensity'.
• Field propensity dictates the number of place fields per unit space and predicts reward-related activity.
• Propensity is stable across environments and months, broadly distributed, and influences spatial code sparsity.
• Cellular excitability correlates with field propensity, which varies significantly between adjacent neurons.

Conclusions:

• Field propensity is a fundamental property of CA1 neurons that governs their representational characteristics.
• The stability of propensity suggests underlying anatomical, transcriptional, or developmental determinants.
• This property provides a unified framework for understanding hippocampal spatial and reward coding.