Key to rapid evolution is mass extinction, researchers suggest

Key to rapid evolution is mass extinction, researchers suggest
Practical applications of the research could include the development of robots that can better overcome obstacles and human-like game agents.

Mass extinction events might not be all that bad if we look at the findings of a new study that suggests that key to rapid evolution is hidden in mass extinction events.

Researchers have suggested through a virtual mass extinction model based on real-life disasters that species may evolve more quickly and efficiently if mass extinction events wipes them off in great numbers. Researchers have based their findings on a study they carried out on robots supporting the idea that mass extinction events speed up evolution by bringing in new creative adaptations.

A team of computer science researchers at The University of Texas at Austin carried out a study, published in journal PLOS One, wherein they describe how simulations of mass extinctions promote novel features and abilities in surviving lineages.

The key here is focused destruction, researchers say and when such a destruction is carried out, it leads to surprising outcomes said Risto Miikkulainen, a professor of computer science at UT Austin. Miikkulainen added that in order to develop tools that will let you get better, there is a need for things that may seem objectively worse.

The idea of mass extinction as a trigger for accelerated evolution has been hypothesized by evolutionary biologists from some time now. According to this school of thought, mass extinction events promote lineages that are the most evolvable, meaning ones that can quickly create useful new features and abilities.

Mass extinction paves way for accelerated evolution, study says
At the start of the simulation, a biped robot controlled by a computationally evolved brain stands upright on a 16 meter by 16 meter surface. The simulation proceeds until the robot falls or until 15 seconds have elapsed. Credit: Joel Lehman

In their study with robots, Miikkulainen and Lehman found that this hypothesis does hold true. In their study, the team studied how mass destruction could aid in computational evolution. Their study is based on computer algorithms that take inspiration from evolution to train simulated robot brains, called neural networks, to improve at a task from one generation to the next.

The team connected neural networks to simulated robotic legs with the goal of evolving a robot that could walk smoothly and stably. To mimic real evolution, the team introduced random mutations through the computational evolution process. Further, they created many different niches so that a wide range of novel features and abilities would come about.

They let the robots go through hundreds of generations and let the robots gain a wide range of behaviors to fill these niches, many of which were not directly useful for walking. Then the researchers randomly killed off the robots in 90 per cent of the niches, mimicking a mass extinction.

Researchers carried out several such cycles of evolution and extinction and found that lineages that survived were the most evolvable and, therefore, had the greatest potential to produce new behaviors. Further, the researchers also observed that these lineages that survived had at their disposal better solutions to the task of walking compared to those which didn’t undergo mass extinction.

Practical applications of the research could include the development of robots that can better overcome obstacles (such as robots searching for survivors in earthquake rubble, exploring Mars or navigating a minefield) and human-like game agents.