Sparse-to-dense coding transformation between hippocampal areas CA3 and CA1.

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Spatial Navigation

Background:

• The hippocampus is vital for spatial memory and navigation, containing spatially selective neurons called place cells.
• Previous research indicated similar spatial coding in hippocampal subregions CA1 and CA3, despite their differing connectivity.
• This similarity raised questions about functional differentiation between CA1 and CA3, potentially due to experimental limitations like small arenas.

Purpose of the Study:

• To investigate whether experimental paradigms, specifically arena size, influence the observed spatial coding in hippocampal CA1 and CA3.
• To test the hypothesis that distinct coding emerges in larger, more naturalistic environments.
• To explore the functional implications of potential coding differences for spatial learning and memory.

Main Methods:

• Simultaneous recording of CA1 and CA3 place cells in bats navigating flight tunnels up to 200 meters long.
• Analysis of place field density (multiple vs. single fields) and place field size across environments of varying sizes (6m to 200m).
• Utilized a neural-network model to assess the functional consequences of observed coding transformations.

Main Results:

• CA1 neurons exhibited dense spatial coding with multiple place fields, while CA3 neurons showed ultrasparse coding with predominantly single place fields.
• Place field sizes were comparable between CA1 and CA3 across all tested environment sizes.
• A sparse-to-dense coding transformation from CA3 to CA1 was identified, facilitating rapid learning of new spatial maps.
• Place cells demonstrated trajectory-history modulation (retrospective coding) in large environments, influencing navigation over extended distances.

Conclusions:

• The study reveals a significant CA3-to-CA1 neural coding transformation in large, naturalistic environments, contrasting with previous findings in small arenas.
• This transformation reformats spatial information into a compressed neural code, enhancing the efficiency of spatial map learning.
• The findings highlight the importance of environment size in understanding hippocampal spatial coding and its role in memory.