Basic Game-Theory in Evolution

I’m forever pointing this out so I thought I make a little post I can point at when necessary.

This is a tiny recurring example of what is fundamental about cultural evolution, which always involves information exchanges of ideas. (In fact information exchange and replication is fundamental to evolution full stop, biologically and even physically, just more obvious culturally in daily life.) One mystery is cause and effect, or blame and responsibility for consequences in political life, but again the mystery exists at all levels, even physics as I say. It’s mysterious because, despite the appearances of experience and hindsight, most causation involves some interaction and questions of which-way / two-way causation. Hard to credit at the level of physics, and easy to reject in a temporal precedence sense, but then even time is fundamentally weirder than everyday human experience.

The course of a river is constrained (caused) by its banks at the same time as the shaping of the river bank is scoured (caused) by the flow in the river. The net result is predictable in nature but not in detail. So many biological examples are cast as arms-races if competitive or exchanges of interests if collaborative. Again species are a matter of hindsight, not prediction.

At the cultural (social and intellectual) level ideas and responses continually bounce off each other. Even rationally intentioned at each step – it’s a game getting to best (or even least-worst) outcomes as discourse evolves imperfect knowledge and understanding at every expression. Even rationally intentioned, those outcomes may not converge predictably. Add in irony (and metaphor, after Hofstadter) and it’s anybody’s guess – it’s a creative game.

The example here is the irony problem:

It is now standard practice in game shows to read out news items, ads and “letters” sent in. In fact it’s been standard practice to publish individual reactions as long as media have existed, it’s just that the cycle of publish and react (and to react to and/or republish the reaction to the reaction) is so much faster and more ubiquitous these days. So is the irony in the hope or in the truth of the response or hope and intent of the reaction or … where?

It’s in all of them, it’s everywhere. I’ll wager most typos or absurd “funny” contributions are ironic from the outset. The original sender is complicit in the game where the publisher knows the contribution is ironic. The sender gets published and the publisher gets read and re-published in further media – win-win-win-win. Add in those adding their own irony when sharing their response and , you can be sure even the originator is doubling down on multiple levels of irony-but-is-it in order to make moves in the game.

I’m not even sure what “I hope it’s true” means. It’s certainly true that it is ironic, but how many levels of irony? That’s anybody’s guess – which means anybody can make the next move in the game, and there are so many immediate opportunities to make that move. The game of life.


[Post Note: Seeing the usual annual news story (yawn) about who won the world championship with which word, reminds me that Scrabble is an archetype of the process / problem. In my lifetime it’s gone from a game to test family and friends vocabulary-in-use, with the dictionary on hand as adjudication in case of disagreement, to a game of who knows the most obscure letter combinations that players know exist as words in the world’s most comprehensive dictionaries. An entirely different species of game.]

[*** Post Note ***
Very sophisticated simulation of social evolutionary game theory, even if you believe you already know the standard cases. (It’s a powerful visualisation by Nicky Case of Axelrod 1984 much referenced in Dennett and other sources.)

Many different aspects demonstrable. The significance of “trust”. Specifically “fidelity”- many rounds of (conservative) replication with only few exceptions. Also individual to individual significance, as opposed to generalised individual vs mass / class or class vs class cases. That fidelity of % miscommunication (memetic mutation) is very telling – worth playing through to the end.

Trust Simulator

(Spoiler: Our problem today isn’t just that people are losing trust, it’s that our environment [unmediated media communications anyone?] acts against the evolution of trust.)

Can be replayed with different strategies and different research aims. Sorry, lost the original Tweeted link, but I suspect must have been EES sourceThis seems like definitive demonstration of my own intuitive position. Respect and trust matter, and “social-media” is destroying it. It’s anti-social media. So, sadly now, it’s even worse. We can now not only game “the system” we can game the long-run behaviour of game-theory itself. This is REALLY SCARY.]

Also published on Medium.

One thought on “Basic Game-Theory in Evolution”

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.