Primer Home /
Evolution / Evolving beyond biology
Evolving beyond biology
Topic: Evolution
by Ben, 2020 Cohort
Evolution is the change in heritable characteristics of agents over time. Conceived in biology, evolution typically referred to the changing characteristics of natural populations over generations; however, evolution occurs in many other places, and a more general conception yields greater insight into its wider application. To understand the evolutionary process, I’ll use two simplified examples; the biological evolution of a species of finch and the evolution of strategies in a game of chess. I’ll then reintroduce some complexity by talking about environmental dynamics.
A system with evolution must meet three criteria. Firstly, there must be variation- individual species or agents change their characteristics. For the finch, this happens through genetic mutation and sexual recombination, which varies the characteristics of offspring. In the chess game, variation occurs as players choose different
Secondly, there must be selection- characteristics which produce greater fitness in the environment confer a survival advantage, and characteristics which confer worse fitness must be selected against. For the finch, this happens naturally over generations- birds more adapted to their environment are likely to survive and reproduce, while those with unhelpful characteristics will die quickly. In the chess game, weaker strategies will lose more frequently than stronger strategies, and stronger ones will then be used more frequently in future
Finally, there must be heredity- characteristics must be able to be passed down over time. In the finch, genes encode characteristics, which are conferred to offspring. In chess, players remember their successes and failures, and the attached strategies; over time, they learn which strategies are most likely to succeed in any given
‘Variation- selection- heredity’. These simple constitutive elements of evolution appear in many places, in different forms. The finch/chess examples are specific instances of a more general distinction between random evolution and *intentional *evolution. In random evolution, agents do not choose what to vary or what to pass on, like in the case of genes. In *intentional *evolution agents have choice in how they vary characteristics and what is hereditary, like the player choosing to try new strategies and what they choose to try again later. This distinction makes many fields clearly amenable to evolutionary analysis, including politics- where agents might vary policies, be selected by elections, and pass on successful policy platforms to future party members- and machine learning, where neural network algorithms are formed by varying inputs and comparing to a target function
The true complexity of evolution becomes clear in considering how agents and organisms influence the environment to which they adapt. For the finches, as they evolve to adapt to their environment their interactions with predators and food sources will change. If they evolve beaks that are particularly good at cracking snail shells, they might accidentally extinguish the snail population on their island and render the trait useless- with beaks specialised for snails, they can’t eat alternative food sources and start to die out. Complexity stems from the feedback between individual adaptation, the effects of adaptations on the environment, and the continued adaption to a changing
Explore this topic further#
Return to
Evolution in the Primer
Disclaimer#
This content has been contributed by a student as part of a learning activity.
If there are inaccuracies, or opportunities for significant improvement on this topic, feedback is welcome on how to improve the resource.
You can improve articles on this topic as a student in "Unravelling Complexity", or by including the amendments in an email to: Chris.Browne@anu.edu.au
Evolution is the change in heritable characteristics of agents over time. Conceived in biology, evolution typically referred to the changing characteristics of natural populations over generations; however, evolution occurs in many other places, and a more general conception yields greater insight into its wider application. To understand the evolutionary process, I’ll use two simplified examples; the biological evolution of a species of finch and the evolution of strategies in a game of chess. I’ll then reintroduce some complexity by talking about environmental dynamics.
A system with evolution must meet three criteria. Firstly, there must be variation- individual species or agents change their characteristics. For the finch, this happens through genetic mutation and sexual recombination, which varies the characteristics of offspring. In the chess game, variation occurs as players choose different
Secondly, there must be selection- characteristics which produce greater fitness in the environment confer a survival advantage, and characteristics which confer worse fitness must be selected against. For the finch, this happens naturally over generations- birds more adapted to their environment are likely to survive and reproduce, while those with unhelpful characteristics will die quickly. In the chess game, weaker strategies will lose more frequently than stronger strategies, and stronger ones will then be used more frequently in future
Finally, there must be heredity- characteristics must be able to be passed down over time. In the finch, genes encode characteristics, which are conferred to offspring. In chess, players remember their successes and failures, and the attached strategies; over time, they learn which strategies are most likely to succeed in any given
‘Variation- selection- heredity’. These simple constitutive elements of evolution appear in many places, in different forms. The finch/chess examples are specific instances of a more general distinction between random evolution and *intentional *evolution. In random evolution, agents do not choose what to vary or what to pass on, like in the case of genes. In *intentional *evolution agents have choice in how they vary characteristics and what is hereditary, like the player choosing to try new strategies and what they choose to try again later. This distinction makes many fields clearly amenable to evolutionary analysis, including politics- where agents might vary policies, be selected by elections, and pass on successful policy platforms to future party members- and machine learning, where neural network algorithms are formed by varying inputs and comparing to a target function
The true complexity of evolution becomes clear in considering how agents and organisms influence the environment to which they adapt. For the finches, as they evolve to adapt to their environment their interactions with predators and food sources will change. If they evolve beaks that are particularly good at cracking snail shells, they might accidentally extinguish the snail population on their island and render the trait useless- with beaks specialised for snails, they can’t eat alternative food sources and start to die out. Complexity stems from the feedback between individual adaptation, the effects of adaptations on the environment, and the continued adaption to a changing