Visualizing evolution in action: Dynamic fitness landscapes

Fitness landscapes were invented by Sewall Wright in 1932. They map fitness, or reproductive success, of individual organisms as a function of genotype or phenotype. Organisms with higher fitness have a higher chance of reproducing, and populations therefore tend to evolve towards higher ground in the fitness landscape. Even though only two traits can be visualized this way, we can actually observe evolution in action. Building on the idea of fitness landscapes, Bjørn Østman and I decided to create some animations of simulated evolving populations to illustrate concepts of evolution that are typically difficult to comprehend.

Here we demonstrate the effect of a dynamically changing environment on an evolving population. If the environment changes slowly enough, the population can adapt and “keep up” with environmental change. But if the environment changes too quickly, evolution breaks down and the population can no longer adapt to the environment.

Warning: The GIFs on this page are large and may take some time to load.



Here’s the full video that Bjørn and I submitted to the ALife 2014 Science Visualization Competition.

Dr. Randy Olson is a Senior Data Scientist at the University of Pennsylvania, where he develops state-of-the-art machine learning algorithms with a focus on biomedical applications.

Posted in data visualization, research Tagged with: , , , ,