https://www.cell.com/trends/ecology-evolution/fulltext/S0169-5347(18)30117-4 The authors do not seem to use the term bushy to describe the non-linear evolution of Homo sapiens, but other authors have used it. They use 'African multiregionalism' instead. And also they exclude panmyxia (free for all gene flow as an assumption) in more local populations. The concept is a geographically very widespread series of populations that occasionally interbreed, then remain in isolation, a kind of off again/on again situation. So at any one time in the deep past, there are Homo sapiens populations with very different traits. Example from other sources: Homo naledi (250kya)and Homo florisiensis (30kya) are extreme examples of isolation of species of Homo concurrent with H. sapiens. http://humanorigins.si.edu/evidence/human-fossils/species/homo-floresiensis https://www.newscientist.com/articl...only-250000-years-old-heres-why-that-matters/ The Cell article shows some nice graphics of traits in skulls moving from one population to another. One takeaway concept is that the nice orderly linear progression model of human evolution we learned in elementary school apparently does not apply well. Lots of branches and dead ends is a better match.