Evolution Explains Kindness, Even When It Kills Us | Paul Bloom

The Prisoner's Dilemma models cooperation: both cooperate (3,000 each), one defects (5,000, 0), both defect (1,000 each). In one-shot games, defection is optimal. In repeated games, Tit-for-tat (start cooperating, then copy partner's previous behavior) is an ESS and can lead to cooperation evolution. This requires T > R > P > S in the payoff structure.

Reciprocity explains altruism toward non-relatives through expected future returns. The prisoner's dilemma models cooperation: two suspects must decide whether to cooperate or defect. Mutual cooperation yields moderate rewards (1 year each), mutual defection yields moderate punishment (3 years each), but one defector gains while the cooperator suffers (5 years). An evolutionary stable strategy (ESS) cannot be invaded by alternatives. Tit-for-tat (TFT) is a successful ESS: start by cooperating, then replicate the other's previous move. TFT is forgiving but punitive, allowing reconciliation while punishing defection. By-product mutualism occurs when survival depends entirely on cooperation with others, with no temptation to cheat. Group selection, while initially considered incompatible with Darwin's theory, occurs in some cases like ants where colonies function as superorganisms.

The Prisoner's Dilemma is a classic game theory model where two players must decide whether to cooperate or defect. If both cooperate, they both benefit moderately. If one cooperates while the other defects, the defector gains everything while the cooperator loses. If both defect, both lose moderately. This model helps explain why cooperation is difficult to maintain in populations, as defectors can exploit cooperators and gain advantages without reciprocating.

Evolutionary Game Theory extends classical game theory by modeling populations of agents rather than rational individuals. Agents interact, accumulate fitness from payoffs, and reproduce proportionally to success. An Evolutionary Stable Strategy (ESS) resists invasion by alternative strategies—once established, mutants cannot outcompete it. In the Prisoner's Dilemma, defection is an ESS while cooperation is not. The Stag Hunt has two stable equilibria (cooperation and individual action), while the Snowdrift has unstable equilibria requiring mixture analysis. This framework reveals why purely rational models fail to explain observed cooperation levels, setting up the need for additional mechanisms that transform game structures.

The concept of ESS explains why cooperation is difficult to achieve in biological systems. Even when all individuals would benefit from cooperating, ESS predicts that individuals will defect because defectors gain higher individual payoffs. The Prisoner's Dilemma illustrates this: in a one-shot game, defection is always the ESS because it provides the highest individual payoff regardless of the other's choice. However, in repeated interactions, conditional cooperation strategies can become ESS. Tit-for-Tat is a conditional cooperation strategy where an individual cooperates on the first round and then copies the opponent's previous move. This strategy is highly successful because it is nice (cooperates first), retaliatory (punishes defection), forgiving (returns to cooperation after opponent cooperates), and clear (opponents can easily understand the strategy). Tit-for-Tat can invade populations of defectors and resist invasion by other strategies, making it an ESS in many repeated interaction scenarios.

In 1962, W.D. Hamilton argued that group selection explains many animal behaviors, suggesting that certain behaviors might be favored because they benefit the entire group in the long term rather than just the individual. However, biologists John Maynard Smith and George Williams later demonstrated that individual selection can counteract or overwhelm group selection. They showed that selfish behaviors benefiting individuals would spread and increase in frequency, causing group selection arguments to fail. This debate established that individual fitness almost always prevails over group fitness in evolutionary explanations.

Human nature contains both selfish mechanisms (competition within groups) and groupish mechanisms (cooperation between groups). While many evolutionary psychologists emphasize individual selection, group selection (multi-level selection) is increasingly supported given evidence of intense inter-group competition throughout human history, rapid genetic evolution, and the effectiveness of group-benefiting adaptations. The rejection of group selection in the 1960s-70s occurred before understanding major transitions in evolutionary history, war's long-term prevalence, and the speed of genetic change. Group selection explains adaptations for transcending self-interest in group contexts.

Group selection is an evolutionary theory proposing that natural selection can operate at the level of groups rather than just individuals or genes. While the dominant view in evolutionary biology has been that selection favors traits that benefit individual survival and reproduction, group selection suggests that traits benefiting the group as a whole can evolve even if they disadvantage individuals within the group. For example, altruistic behaviors that help group members survive might spread through evolution even if they reduce the reproductive success of individuals who perform them. Critics argue that group selection is unnecessary because altruism can be explained by kin selection (helping relatives who share your genes) or reciprocal altruism (helping others with the expectation of future reciprocation). Proponents like E.O. Wilson argue that group selection may explain phenomena that other theories cannot, such as extreme altruism in social insects or large-scale human cooperation.

The group selection controversy was resolved through empirical and theoretical evidence. Christopher Wills demonstrated that animals do not practice family planning or control reproduction for species benefit. John Maynard Smith showed theoretically that 'cheaters' would always invade altruistic populations, making group selection unsustainable. The key insight is that only humans can police cheaters and maintain cooperation through social structures. This represents a complete reversal from the group selection view that humans had lost this ability.

Effective altruism is the philosophy of using resources to benefit others as effectively as possible. Key principles include: (1) Not all charities are equally effective—research shows differences of up to 100 times between organizations; (2) Choosing where to give matters more than how much to give; (3) All lives are equally valuable regardless of location, so we should help those in need wherever they are; (4) Global inequality is extreme—earning $60,000 in a developed country places you in the top 1% globally; (5) Effective giving can achieve greater impact than traditional career paths, as demonstrated by analyses showing doctors save only 7 lives on average while effective donations could save more lives globally.

Effective altruism is a global movement encouraging people to rethink what they can do to help others and improve the world. The core principle is that doing good is not enough—resources are limited, so we must spend them wisely on things that are truly effective. The same amount of money has vastly different purchasing power depending on where it is spent. People in developed countries like Italy, who earn over €46,000 annually, are actually in the top 1% of the global population. This means that 5 euros might be a difference between having an aperitif or not for someone in Italy, but could mean the difference between life and death for someone living in extreme poverty. Even someone earning €1,300 per month in Italy who donates 10% to highly effective causes would still be 8.6 times richer than the global average. With those donations, they could pay for 1,384 deworming treatments or 270 malaria prevention nets annually, each saving one life. Over a 40-year working life, this could save approximately 40 lives.

Effective altruism holds that we should be impartial between people in our own community and people in distant communities. Just as we would be prepared to incur costs to save a child in a pond regardless of our relationship to them, we should be prepared to help people who are far away from us. Everyone has finite resources, and even if we give away almost everything we have, it will still be finite. Effective altruists suggest that we should give away whatever we can afford while maintaining our own reasonable comfort, and consider whether we can do more good by giving the whole amount to the most cost-effective interventions.

Effective altruism is a philosophy that applies engineering, mathematical, and economic principles to solve for maximum human happiness. Practitioners believe they can use equations to determine the optimal allocation of resources to create the greatest good. Some effective altruists advocate for mosquito net donations while others support more controversial approaches like creating cryptocurrencies to distribute universal basic income.

Effective altruism is a global philosophical and social movement that uses evidence and reason to determine the most effective ways to do good in the world, advocating that individuals should donate to charities that produce the greatest positive impact per dollar, rather than giving based on emotional attachments; this approach combines moral obligations to help those in need with systematic analysis of which causes can achieve the most good, using frameworks like the ITN (Importance, Tractability, Neglectedness) model to identify high-impact opportunities such as malaria prevention, famine relief, and pandemic preparedness, while also extending to career choices and lifestyle decisions aimed at maximizing one's contribution to reducing global suffering.