How a 200-Metre Tunnel Exposed the Secrets of Hippocampus Spatial Memory

Inside a 200-metre flight tunnel, a bat swoops through the air, its brain silently calculating its position. For decades, neuroscientists tracking these internal maps in small laboratory cages saw only redundancy. Two neighbouring hippocampal subregions, CA1 and CA3, appeared to mirror each other perfectly, leaving a frustrating mystery as to why the brain would duplicate such complex work.

The Scale of Hippocampus Spatial Memory

To understand how mammals navigate the real world, researchers had to break out of the cramped confines of standard laboratory boxes. Our understanding of hippocampus spatial memory, historically observed within small laboratory enclosures, was limited by the scale of our experiments, which masked the true elegance of the brain's internal GPS.

By recording neural activity in bats flying across massive distances, researchers observed a striking divergence. The researchers monitored these cells simultaneously as the bats traversed the 200-metre track, ensuring identical environmental conditions. In the CA3 subregion, neurons fired in an "ultrasparse" manner, activating at only a single, specific location. Conversely, CA1 neurons exhibited dense coding, firing across multiple locations to build a detailed, overlapping map of the flight path.

This sparse-to-dense transformation suggests a highly efficient compression system. A neural-network model of this transition indicates that this architecture may facilitate the rapid learning of new spatial maps. Furthermore, the researchers observed that past trajectories altered neural firing for over 100 metres, suggesting that internal maps are deeply shaped by context and history rather than just current coordinates.