It would be easier than standing on and rolling a ball, which here in 3D some people and dogs have learned to do. Here's an astounding video of a dog descending a garden stair whilst atop a ball. https://www.youtube.com/shorts/3qX6Sok9_b0

For a 4D circus act, make an even bigger torus and have several people stand on and roll it about footwise. Then earn your bread by walking on your hands, jumping or backflip across the torus, having some people walking forward and others backward, and so forth. That would take real coordination.

---

Another difference is that in 3D a torus has a major and a minor radius. The length of these cannot be exchanged, as that would cause the torus to intersect itself. In 4D either radius may be larger than the other. The relatively larger the vertical diameter compared with the horizontal one, the more difficult such walking. In very small horizontal diameters it would be like standing on top of a pole that can roll. Can't have stilts like that though -- the tori are too wide.

----

All in all, lately I've been realizing how loose the connection between two perpendicular 2D planes in 4D. I mean really perpendicular, with both significant angles with magnitude of pi/2. When one imposes Cartesian coordinates in 4D the WX plane is free to orient itself however it pleases with respect to the YZ plane, and vice versa. What is surprising to me is that this carries over to the physical world.

I'm beginning to see how Clifford proved that rigid objects could rotate in two planes independently. A rotation is rigid if it does not cause angles and/or lengths between points to change. Another way to say this is that polygons do not change under rotations (except of course for their location). Well, rotations in the WX plane have no effect whatsoever on Y and Z components because the angle between any pair of WX and YZ vectors is always pi/2, no matter what those vectors are doing. As long as the lengths of the vectors doesn't change, the length of the yz-wx vector doesn't change either. No change in angles, no change in lengths, means that the rotation is rigid.

I have the feeling that this is closely related to e

_{1}e

_{2}+e

_{3}e

_{4}of geometric algebra not being a blade, but I can't say how.