1/8 - “A gravity-based three-dimensional compass in the mouse brain” is out rdcu.be/b3BWK. Rodent head direction cells form a 'neuronal compass' in the horizontal plane, but can they encode 3D orientation when the head moves in 3D? To find out, we put mice in a rotator.
2/8 - Yes, the rodent head direction system encodes 3D orientation: 55% cells in antero-dorsal thalamus (ADN) and retrosplenial cortex (RSC) are tuned in the horizontal plane (azimuth; 1D) and 71% are tuned in vertical planes (tilt; 2D). 42% are conjunctive and encode both. 
3/8 - By manipulating visual/gravity sensory inputs, we found that tilt tuning is anchored to gravity, likely sensed by the vestibular organs in the inner ear. Gravity is sensed as an egocentric 3D vector; and receptive field of tilt-tuned cells were 3D Gaussians in this space.
4/8 - When the head is tilted, the brain measures azimuth in a “tilted” compass formed by aligning the earth-horizontal compass with head horizontal. Note that 'earth-horizontal' here is defined by gravity, but orientation in this compass is primarily sensed by vision.
5/8 - To summarize the structure of 3D responses, we propose a 3D model where tilt and azimuth tuning interact multiplicatively.
6/8 - Our rotator allowed measuring full 3D tuning curves for the first time! Our model is compatible with previous experiments in bats (it generalizes the toroidal model) and macaques, indicating that a 3D gravity-anchored head direction compass may be ubiquitous in mammals.
7/8 - I thank all the students (@AmadaAbrego et al.) and my long-time PI and collaborator Dora. I continue studying 3D spatial orientation in my new Marmoset lab in Frankfurt; stay tuned!
8/8 - And I say goodbye to my beloved 3D macaque rotator with which I made my first experiment ever (back in Zurich in 2008). It is heading for retirement after ~30 years of service (and a re-conversion into a mouse rotator).
