Learning to use landmarks for navigation amplifies their representation in retrosplenial cortex.

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Visual landmarks are crucial for spatial navigation, influencing spatially tuned neurons like place cells.
• Understanding how visual cues acquire spatial meaning in the brain is essential for deciphering navigation mechanisms.

Purpose of the Study:

• To investigate the neural mechanisms by which visual landmarks gain spatial meaning in the mouse retrosplenial cortex (RSC).
• To characterize the dynamic changes in RSC neuronal activity during spatial learning and its dependence on behavioral context.

Main Methods:

• Chronic two-photon imaging of neuronal ensembles in mouse RSC during a spatial learning task.
• Manipulating behavioral context by decoupling treadmill motion from visual feedback.
• Computational modeling of neuronal firing dynamics, including burst firing and somatodendritic interactions.

Main Results:

• A significant increase in landmark-referenced activity was observed in RSC neurons during spatial learning, which remained stable over days.
• Neuronal responses were systematically altered by changes in behavioral context and the coherence between visual scene flow and self-motion.
• Modeling suggested that burst firing could mediate context- and coherence-dependent integration of landmark information.

Conclusions:

• Visual encoding in the RSC shifts towards landmark-referenced and context-dependent representations as spatial meaning is acquired during learning.
• These findings shed light on the neural basis of landmark-guided navigation and context-dependent spatial coding.