The Genetics of Convergent Evolution of Vocal Learning

Convergent evolution occurs when two unrelated species evolve similar results. The most famous example of convergent evolution is the eye of the octopus. The octopus, an invertebrate, has a “camera” eye, just like vertebrates do. A “camera eye” has a light-excluding orb like the box of a camera. It has, a lens to  focus light, like the lens of a camera. The iris  controls the amount of light entering like the iris of a camera controls the aperture. The retina on the far side of the orb from the lens and iris reacts to the light, like film or an image sensor. Shrimp, insects, and other creatures have different eyes that work in different ways. There are many ways to evolve an eye, but two widely separated groups reached similar results.

In “Convergent transcriptional specializations in the brains of humans and song learning birds1,” Pfenning et. al. took the full genetic sequences of several hominids, parrots, songbirds, hummingbirds and non-vocal learning birds and compared them using powerful and complex data analysis on computers. The researchers demonstrate that the area of the brain devoted to learning vocalizations evolved separately in primates and in birds. Among the birds (hummingbirds, songbirds, and parrots), the area for vocal learning evolved two or three separate times. That the same ability—the ability to learn and replicate vocalizations—evolved separately several times was to be expected. The ability to learn new patterns of vocalizations has many evolutionary advantages. But what the researchers found was the 50 or so genes used to control the development in this area were similar across the three groups of birds and primates. The last common ancestor of birds and humans was 310 million years ago. That’s a long time. Dinosaurs evolved about 230 million years ago, so the last common ancestor predates the dinosaurs by about 80 million years.

How did the different groups reach the same biochemical result?

Imagine you purchase a lot in a city. What you can build is determined by the size of the land, the geology, the surroundings, and any applicable zoning regulations. If the lot already has a foundation on it, your choices are more limited. You can destroy the foundation (expensive and time-consuming) or you can build based on the restrictions caused by the foundation. If the first floor has also been built, what you can build is much more limited unless you remove the additional work and start over.

In the same way, the structure of the brain of the last common ancestor of birds and primates imposed limits on how the brain could evolve.

Remember that evolution does not progress from simpler to more complicated or even from less fit to more fit. Evolution is random; it’s stupid, it doesn’t think and it doesn’t plan. The genes of the next generation are determined by chance: chance of mutations, chance in mating, and chance of survival. Even biologists have difficulty understanding this. We think of the first cells and then think of humans and mistakenly conclude evolution had a goal of creating intelligent creatures like us, but that isn’t a correct view of evolution. I find myself having to fight the incorrect urge to speak as if evolution had a goal. Anthropomorphizing evolution is an easy mistake to make, but very, very wrong.

If a bunch of organisms with a brain happen to evolve into creatures that can learn vocalizations, what’s the most likely way it can happen? Is it more probable that structures already in place will be used by chance, or is it more likely that those structures will be lost and then a new structure will happen to evolve? Now, actually calculating the odds on those possibilities is impossible, but you don’t have to run a Monte Carlo simulation to realize that more times than not, the existing brain structures will be used.

In the case of the vocal learning areas of the brains, apparently using certain genes is the most probable way to do it.

The finding that convergent neural circuits for vocal learning are accompanied by convergent molecular changes of multiple genes in species separated by millions of years from a common ancestor indicates that brain circuits for complex traits may have limited ways in which they could have evolved from that ancestor.2

Is it the only way to produce vocal learning? No one knows, but I doubt it. There could be some creature out there that evolved vocal learning a different way. If different gene patterns for vocal learning exist, one of the four groups examined (mammals, parrots, hummingbirds and songbirds) might have used it instead.  That the four groups all used a similar genes to perform the same function shows that alternate methods are likely rare.

Going back to the octopus, some species are capable of learning to mimic other creatures or their surroundings visually. Is this “display learning” a similar form of data analysis to vocal learning or is it hardwired mimicry? Do both systems use similar genes? The last common ancestor between vertebrates and invertebrates date back at least to the Cambrian. so that’s a very long time ago and in a totally different system in the brain. If the genes for display learning and vocal learning have similar properties (let alone similar genetic sequences), it would tell us there are severe restrictions on how brains learn.

  1. Pfenning, A. R. et al. “Convergent transcriptional specializations in the brains of humans and song learning birds.” Science 346 1256846 (2014) PDF downloaded 12 Dec. 2014. []
  2. Ibid []

Written by Rob Carr


Read previous post:
What’s the Matter with an Aquarium

Note to the adult helper: Please look over the following material. You might want to cut some things out. Remember,...