Active locomotion predictively rescues head direction attractor dynamics in head-fixed mice

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Head direction (HD) cells in the anterodorsal thalamic nuclei form the brain's internal compass.
• These cells are modeled as ring attractors using visual and inertial input for azimuth coding.
• Head-fixed animal preparations are common but may alter neural coding.

Purpose of the Study:

• To investigate how head-fixed experimental conditions affect the encoding of head direction by HD cells.
• To understand the impact of vestibular conflict and movement constraints on HD network dynamics.
• To revise computational models of the HD network.

Main Methods:

• Electrophysiological recordings in head-fixed mice (real and virtual reality).
• Analysis of unit and population activity of HD cells.
• Computational modeling of HD network dynamics with perturbed lateral connectivity.

Main Results:

• Complete head fixation with vestibular conflict disrupts HD unit and population encoding.
• Constraining head-on-body movements impairs HD population activity.
• Attractor dynamics are altered during immobility but recover before locomotion onset, suggesting reliance on efference copy or predictive signals.

Conclusions:

• The HD network functions as a context- and state-dependent predictive estimator.
• Network activity is stabilized by anticipated self-motion signals.
• Classic ring-attractor models require revision to incorporate context-dependent dynamics and prospective motor signals for stability in constrained conditions.