Uncovering 2-D toroidal representations in grid cell ensemble activity during 1-D behavior.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Minimal experimental paradigms, like head-fixed wheel running, offer advantages but limit observable behaviors, hindering functional cell-type classification.
• Discovering complex neural representations, such as grid cells, has historically relied on observing behavior in open environments.

Purpose of the Study:

• To demonstrate that analyzing neural population activity, rather than single neurons, reduces the experimental complexity needed to study internal representations.
• To investigate if grid cell modules exhibit similar stable state spaces in minimal experimental setups as observed during free exploration.

Main Methods:

• Shifted focus from single-neuron activity to population-level analysis.
• Identified grid cell modules and analyzed their state space dynamics during head-fixed wheel running.
• Correlated neural trajectories with behavioral data from virtual reality and path integration tasks.

Main Results:

• Grid cell population activity covers a stable toroidal state space during wheel running, comparable to open-field foraging.
• Neural trajectories on these state spaces correspond to single-trial runs and path integration.
• The alignment of grid cell representations rapidly adapts to changing experimental conditions.

Conclusions:

• Analyzing neural population activity enables the discovery and study of complex internal representations in simplified experimental settings.
• This approach provides a methodology to investigate neural representations with reduced experimental complexity.
• The findings suggest that minimal experimental setups can reveal fundamental properties of neural coding.