I've been playing around with Tyler Morgan-Wall's Rayshader package, which allows users to turn their ggplots into 3D visualizations with just a few lines of code.
Although the use of 3D in visualization is contested, constructing 3D ggplots in R was exciting/enlightening/productive for three reasons:
1) Visualizations of regional differences (in my case, in kindergarten vaccination rates) may benefit the reader via redundant encoding, especially when the first method of visualization (color shading) is also susceptible to misinterpretation. While color is likely more effective at presenting relative differences than the depth of each county, neither is as good as something like a bar chart. There is a chance that presenting differences with both methods--color and depth--could help deliver the visualization's message to the viewer. (If anyone has literature on this, could you send it my way?)
2) This is good practice for a) creating choropleth maps and b) playing around with GIF/mp3 creation in R, something I'm very unfamiliar with!
3) Rotating 3D visualizations are beautiful.
Here's the final output. If you're interested in seeing the code that created it, you can find it here.