Posts by Jorge Peña 🇵🇸
La vida era más corta. www.youtube.com/watch?v=aRk5...
Precisely the topic of my last preprint.
We're just stupid automatons at the mercy of natural selection. Which is awful.
Who is killing the world's journalists?
Remembering Gaza scholars. Honoring the lives and legacies of Palestinian academics whose voices have been silenced.
🇵🇸
rememberinggazascholars.org
Doing genocide and running torture-rape camps is apparently fine but upsetting US Christian Zionists is a “moral wake-up call”. Absolute degeneracy.
"But he’s moved on from the AI hot girl influencer niche. He says he needs to shift to focusing on his studies."
www.wired.com/story/ai-gen...
Thanks! I will check it out.
📣 EvolDir is now managed by @eseb.bsky.social!
We are delighted to be taking the reins and express our gratitude to both Brian Golding who began this service to the community in the mid-1980s and to @rdmpage.bsky.social who ran this account until now 👏
You can now find evoldir here: evoldir.net
Ευχαριστώ!
Forgot to add the hashtag #CESRabat, and tag the conference @ces2026.bsky.social.
Also: by "paper" I mean "working paper" or "preprint"; "benevolence"/"malevolence" is what behavioral economists call "altruism"/"spite" (different from what biologists call "altruism"/"spite").
Branching is specific to adaptive dynamics (AD), but I'd expect stable polymorphisms when modelling everything with discrete preferences from the beginning. Obtaining branching in an AD model is often a good predictor of having stable mixed equilibria in a related discrete-trait model.
Our contribution is to show that there can be branching of a pro-social preference trait in an n-player public goods game with a CES production function. Branching at the level of strategies (effort levels) is impossible under convex costs (which is the usual econ assumption).
Thanks, Josh. There are at 2 other papers applying adaptive dynamics to preference traits: Lehmann & Alger (2023) link.springer.com/article/10.1..., Lehmann & @charlesmullon.bsky.social (2025) www.biorxiv.org/content/10.1.... We need to rewrite our introduction to better reflect this previous work.
Don't be shy to take on a little two-week side project. These five months will be the most precious three years of your academic journey.
I'll be presenting this work at the Cultural Evolution Society (CES) 2026 Conference in Rabat, Morocco (airess.fgses-um6p.ma/ces2026) on May 13. If you attend the conference and are interested in this, let's talk!
In Doebeli, Hauert & Killingback (2004) (www.science.org/doi/abs/10.1...) evolutionary branching under *strategy evolution* required (implausible?) concave effort cost functions. Our model is one of *preference evolution*, and it works with more realistic convex costs.
Under strong enough substitutability, the malevolent singular point λ* is also an evolutionary branching point. A monomorphic population is first dragged to λ* by directional selection, then split by disruptive selection into two diverging social preference clusters.
We model (cultural) evolution using adaptive dynamics. Strategic complementarity (at the utility level) decides the sign of the convergence stable singular strategy: strategic complements → λ*>0 (benevolence); substitutes → λ*<0 (malevolence).
Our model is an n-player public-goods game where evolution acts on a social-preference parameter λ. λ>0 is benevolent (weight partners' payoffs positively), λ<0 is malevolent, λ=0 is Homo economicus. Agents play Nash in efforts given their preferences. Fitness = material payoff at that equilibrium.
🧵 New paper w/ Guillaume Cheikbossian: "Evolutionary branching of social preferences in a public good provision game" (ideas.repec.org/p/tse/wpaper...). Can behavioural heterogeneity in cooperation be an adaptive outcome? We think the answer is yes.
😂
The world has to be fucked up for the head of the Catholic Church to be one of the few world leaders with a vertebral column right now. That is the world in which we live right now.
www.reuters.com/business/med...
"United States of America" is a ridiculous name for a country. Imagine if Switzerland were called the "United Cantons of Europe" or if your kid were called the "United Cells of Humanity", and their nicknames were, respectively, "Europe" and "Human".
Blockade, blockade.
youtube.com/shorts/HLsP5...
Poilievre better hope he never comes to Canada.
It's called evolutionary game theory.