3D pose estimation enables virtual head fixation in freely moving rats

Area of Science:

• Neuroscience
• Biophysics
• Behavioral Science

Background:

• Quantifying spontaneous movements is crucial for understanding their impact on neuronal activity.
• Existing methods for behavior tracking can be labor-intensive and require extensive animal training.

Purpose of the Study:

• To develop a versatile, marker-free framework for precise 3D motion capture of freely moving animals.
• To investigate the relationship between specific body movements and neuronal activity in the motor cortex.
• To enable the analysis of neural coding independent of overall body posture.

Main Methods:

• A novel framework was developed to capture 3D motion of freely definable body points without markers.
• The motion tracking system was integrated with neural recordings in freely moving rats.
• A computational model was used to integrate behavioral variables and derive a virtual head fixation, removing body movement influence.

Main Results:

• The framework achieved high precision and reliability in marker-free 3D motion tracking.
• Multiplexing of information was observed in motor cortex neurons of freely moving rats.
• A significant fraction of motor cortex neurons were found to be tuned to front paw movements, a tuning previously masked by body posture.
• The virtual head fixation strategy successfully isolated the influence of specific body movements.

Conclusions:

• The developed framework offers an efficient and minimally invasive method for quantifying complex behaviors in neuroscience research.
• This approach reveals novel insights into neural representations of movement, particularly the tuning of motor cortex neurons to paw movements.
• The technology has the potential to accelerate the analysis of large behavioral and neural datasets, reducing experimental burden.