Landmark vector cells in the absence of visual input.

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Spatial navigation relies on hippocampal landmark vector cells (LVCs) representing an animal's position relative to landmarks.
• LVCs fire at specific allocentric angles and distances to landmarks.
• The necessity of visual input for LVC function remains unclear.

Purpose of the Study:

• To investigate whether visual inputs are essential for the formation and maintenance of hippocampal LVCs.
• To explore the computational mechanisms underlying LVCs and their relationship to place cells.

Main Methods:

• Rats explored a novel platform with landmarks in complete darkness.
• Hippocampal CA1 neuronal activity was recorded.
• A computational model based on McNaughton's vector encoding was updated to explain observed neuronal phenotypes.

Main Results:

• LVCs were present and functional in CA1 even in complete darkness.
• LVC vector fields accurately tracked landmark movements.
• Spatial selectivity of LVCs persisted even after landmark removal.
• An updated computational model explained LVC and place cell phenotypes through entorhinal input variations.

Conclusions:

• Visual input is not required for the formation or maintenance of LVCs.
• LVCs and place cells may share a common computational basis.
• Differences in egocentric inputs could explain distinct spatial firing patterns observed in these neurons.