We have a new paper in eLife (link) in which we show that specialization may evolve more readily than previously thought.
The single most important benefit of multicellularity is the fact that organisms can evolve specialized cell types (i.e., being a brain or skin cell), but we know little about how specialization first evolved. It has long been assumed that specialization will evolve only if there is an accelerating, or convex, return on investment; for example, if twice the investment in a given task produces four times the yield. We have been studying the role of group structure in the evolution of specialization, using a minimal individual-based model. Surprisingly, we found that for a broad class of sparsely connected structures, specialization evolves even with saturating, or concave, returns on investment. Sparsely connected groups are able to connect many complementary specialists, which increases the benefit of specialization, without also connecting many like-specialists, which decreases the value of specialization. Our results remove a significant barrier to the evolution of specialization – the existence of accelerating returns on investment – and thus suggest that the evolution of specialization is even more favorable than previously thought