When we think about the process of evolution, we often imagine the slow changing of birds from dinosaurs, or the domestication of crops. Evolution by natural selection is the driving principle behind all of biology, but it can be tricky to wrap your head around. Random mutation, heritability, fitness, environment, populations, allele frequencies and so on are often murky topics especially for the general public.
Fortunately not all evolutionary processes require hundreds of years to observe. Microbes (like bacteria and virus) often evolve on the order of years, months, even days. With such fast evolving systems, we can observe evolution in real-time. However what would be really awesome would be to witness evolution take place on the order of seconds or minutes. Short of inventing a time-machine, we have only one system to observe evolution happening on such short time scales, digital systems. Using computational systems, we can look at evolution occurring on populations of digital organisms. There are a plethora of such digital evolution models/games/toys freely available on the web. And I wanted to point out some of my favorites.
Evolving Images (Web App)
Mona Lisa after 25 minutesThis web application by Jacob Seidien takes a target image and uses a genetic algorithm to reconstruct the image with a small number of polygons. Organisms in the population consist of the location of 40 colored polygons. Each generation, the organisms with the highest fitness (those whose polygon placement most closely matches the desired picture) are replicated into the next generation with the potential for mutation (changes to the polygon location, shape and/or color). You can watch as the population adapts to more closely match the target image.
Primordial Life (Link to Download)
This alife (artificial life) program allows for organisms consisting of radially-symmetric line segments having different properties. Green lines generate energy (emulating photosynthesis), red lines can ‘eat’ green lines and thus steal energy, dark blue lines protect against attack by red lines, light blue generate thrust for motility, and white lines allow for the transmission of genetic material. Starting from a random population, you can watch a population evolve. Often you will see the sensitive parts of an organisms (green lines) deep within, protected by blue lines and red lines extended outwards to consume other organisms. With sexual recombination turned on, you can watch the evolution of sexual dimorphism (differentiation between males and females) evolve as females (organisms lacking white lines) growing large, while males (containing white lines) shrinking to be tiny, highly motile genetic carriers. This is very similar to the differences between female and male anglerfish.
Bitozoa 2 (Link to Download)
This programs explores the behavioral evolution of large populations of herbivores and carnivores. Organisms consist of a simple body plan which is controlled by an evolvable neural network. The initial, random population exhibits no directed movement toward food sources or away from predators. But with five minutes of evolution, the populations can demonstrate turning toward things they can consume. After another five minutes, you can observe the herbivores performing evasive maneuvers to avoid predators and predators doggedly pursuing herbivores.
Gene Pool/Swimbots (Link to Download)
In this program, the organisms consist of “swimmers” who need to navigate toward food and mates to propagate. These are highly colorful populations that can evolve a great variety of different swimming strategies. By adjusting how organisms choose their mates (what each organisms finds sexually attractive), you can watch the evolution of speciation and sexual dimorphism. You can also perform breeding experiments by cloning, feeding, hybridizing, and culling unwanted organisms from the population.
All these programs make great screen-savers as the populations therein are continually evolving new, visually-interesting traits. I’ve always been fascinated by evolution, and hopefully some of the digital evolution programs can help spread that joy to others.
This post was contributed by Josh Nahum, a graduate student studying evolutionary biology in the lab of Dr. Ben Kerr.
