The individual organism—traits are favored if they increase an individual's survival and reproductive success.
🧠 This is the basis of the “selfish gene” idea: genes succeed by helping individuals pass them on.
Yes—selection can act at the gene level, favoring “selfish” genes that replicate themselves, even at the cost of the organism.
🧬 Example: transposable elements
Yes—group selection can occur if traits that benefit the group’s survival outweigh costs to individuals.
🌍 Rare, but possible when groups with more cooperators outcompete others.
Genes that promote their own replication, even if they harm the fitness of the organism carrying them.
🧠 They succeed at the gene level, not necessarily the individual level.
The paternal sex ratio (psr) chromosome in the wasp Nasonia vitripennis:
Transmitted through sperm
Destroys all paternal chromosomes in the zygote—except itself
By converting fertilized (would-be female) zygotes into haploid males, which can then pass on the psr element again.
⚠️ This benefits the gene, but disrupts normal sex ratios and can harm the population
It increases the fitness of the group, but it does not evolve “for the good” of the group or species.
🧠 Selection still favors traits that succeed evolutionarily, not morally.
It increases the fitness of the group, but it does not evolve “for the good” of the group or species.
🧠 Selection still favors traits that succeed evolutionarily, not morally.
They cannibalize eggs and pupae, even though it may reduce group survival.
❗This shows that individual selection can favor harmful behaviors if they increase individual success.
It’s often overridden by individual selection—traits that harm the group can still evolve if they benefit the individual.
🧠 Selection works on fitness, not cooperation unless cooperation pays off.
Beetles were reared under three conditions:
Selected for high reproduction
Selected for low reproduction
Control group
Population sizes declined in all groups
Highest population size: high fecundity group
Lowest population size: low fecundity group
Highest cannibalism: control group → good for individuals, bad for group
In the control group, individual selection for cannibalism harmed group size
In the high fecundity group, group selection reduced cannibalism, reinforcing group-level benefits
🧠 Group selection can align with or counteract individual selection depending on conditions.
Altruism is behavior that appears to reduce the individual’s fitness while increasing the fitness of others.
🧠 “Self-sacrifice” in nature that still needs an evolutionary explanation.
Proposed that group selection favors altruistic individuals:
Groups with more altruists would be more stable
These groups would persist longer and reproduce (bud off) more often
Group selection is vulnerable to selfish mutants or immigrants:
A single selfish individual can outcompete altruists within a group
This would prevent altruistic traits from persisting long-term
Wynne-Edwards emphasized group-level benefits
Williams argued that individual-level selection is almost always stronger and will override group benefits
🧠 Unless altruism benefits the individual indirectly, it’s unlikely to evolve by group selection alone.
Human population could grow faster than resources, leading to famine and competition for survival—what he called the "struggle for existence."
Darwin used the idea of struggle for existence as the basis for natural selection—only individuals best suited to compete would survive and reproduce.
Extreme cooperation, like altruism in social insects (e.g., sterile worker ants), seemed to contradict natural selection, which predicts selfish behavior that maximizes individual fitness.
Because altruism means reducing one’s own fitness to increase someone else’s, which should not be favored by natural selection—unless there’s an indirect benefit.
🧠 This puzzle led to theories like kin selection and inclusive fitness.
If it increases the fitness of other individuals who also carry that allele, even if it reduces the donor’s own fitness.
When the recipient is closely related to the donor—because relatives are more likely to share the same allele.
The fitness benefit (B) to the recipient × relatedness (r) must be greater than the cost (C) to the donor:
📏 Hamilton’s Rule → rB > C
🧠 Helps explain how “selfish genes” can lead to unselfish behavior.
The total fitness effect of an allele, including:
The individual’s own reproductive success
Plus the success of others (especially relatives) who carry the same allele
An individual’s inclusive fitness = its own fitness + the fitness it helps produce in relatives
🧠 Genes can “win” by helping copies of themselves in others survive and reproduce.
An altruistic trait will evolve if:
📏 r × B > C
r = relatedness to recipient
B = fitness benefit to recipient
C = fitness cost to donor
✅ If this condition is met, the altruistic allele can spread by natural selection.
It’s the fraction of genes shared by descent between a donor and recipient.
🧬 It reflects the probability that a gene in one individual is also present in a relative due to shared ancestry.
Identical twins: r = 1
Parent-offspring: r = 0.5
Full siblings: r = 0.5
Half-siblings: r = 0.25
Cousins: r = 0.125
Because altruistic traits are more likely to evolve when helping relatives, since it increases the inclusive fitness of the shared genes.
🧠 Helping kin = helping copies of your own genes survive.
The number of extra offspring (or equivalent fitness gain) the recipient gains due to the altruistic act.
The more distantly related the recipient is, the greater the benefit must be to outweigh the cost.
🧠 Because r × B > C must still be true.
Helping a sibling (r = 0.5) gain 4 offspring: 4 × 0.5 = 2 > 1 → favors altruism
Helping a cousin (r = 0.125) gain 4 offspring: 4 × 0.125 = 0.5 < 1 → not favored
✅ Closer kin require smaller benefits for altruism to be selected.
Young Hamiltonian bee-eaters often help older relatives raise offspring instead of breeding themselves during their first few years.
🧠 This behavior delays their own reproduction but increases the success of close kin.
The decision is based on relatedness:
More likely to help full siblings than cousins
Helping close kin provides greater inclusive fitness benefits, making it worth the cost
✅ Altruism is more likely when the benefit × relatedness outweighs the cost.
Females stay in their natal colony → high relatedness
Males disperse → lower relatedness in new groups
🧠 So females are more likely to live with close kin.
Females do most of the alarm calling, despite the high personal risk, because their warning protects close relatives, increasing their inclusive fitness.
✅ Kin-based altruism: help others survive when they likely carry your genes.
Over 19 years, they observed five adoptions of orphaned red squirrel kits by unrelated adult females in the Yukon and measured whether these adoptions fit Hamilton’s Rule.
r = relatedness of adopter to orphan
B = survival benefit to orphan
C = cost to adopter’s own offspring (estimated from known litter survival data)
✅ In all adoption cases, rB > C
❌ In non-adoption cases, rB < C
Adoption decisions matched Hamilton’s Rule:
Squirrels were more likely to adopt when orphans were more closely related
No adoption occurred when the cost outweighed the genetic benefit
🧠 Even rare altruistic acts like adoption can follow kin selection logic.
The most extreme form of altruism, where individuals live in social groups and:
Only a few individuals reproduce
Most colony members are sterile and help raise the offspring of others and maintain the colony
Insects: termites, all ants, some bees and wasps
Mammals: naked mole rats are a rare eusocial mammal
Because individuals give up their own reproduction entirely to support the colony—maximizing inclusive fitness by helping close kin (often siblings) reproduce.
🧠 A lifetime of helping, not mating.
Likely due to their haplodiploid sex determination system, which increases relatedness between sisters and makes helping more genetically rewarding than reproducing.
Females are diploid: get genes from both parents
Males are haploid: develop from unfertilized eggs, and pass on 100% of their genome to daughters
🧬 This means sisters share ~75% of their genes:
50% from mom × 0.5 chance shared = 0.25
100% from dad × 1 = 0.5
👉 Total relatedness: r = 0.75
Because sterile female workers are more related to their sisters (r = 0.75) than they would be to their own offspring (r = 0.5), so helping the queen produce sisters can increase their inclusive fitness more than reproducing themselves.
🧠 Helping mom make sisters = passing on more of your genes.
To sisters: r = 0.75
To brothers: r = 0.25
🧠 Workers share more genes with sisters, so helping raise sisters boosts their inclusive fitness more.
The queen is equally related to sons and daughters (r = 0.5), so she favors a 1:1 male-to-female ratio of reproductive offspring.
Workers want a female-biased sex ratio to maximize their genetic payoff, since they are 3× more related to sisters than to brothers.
➡️ This creates queen–worker conflict over how many males vs. females are produced in the colony.
🧠 Same genes, different interests!
If the queen mates with multiple males, the average relatedness between workers drops (from r = 0.75 to closer to r = 0.5).
🧠 Workers now share fewer genes on average, especially if they have different fathers.
With lower relatedness among sisters, workers are less biased toward producing females—so the conflict with the queen is reduced.
➡️ Sex ratios shift closer to the queen's preferred 1:1 ratio.
✅ More mates for the queen = less incentive for workers to favor sisters.
Workers biased the sex ratio toward females by:
Allowing female eggs to survive
Withholding care from male larvae
➡️ Result: fewer reproductive males than the queen originally laid
Workers did not alter the sex ratio—they allowed equal care for male and female larvae.
➡️ Result: No bias in reproduction toward females.
In singly-mated colonies, high sister relatedness (r = 0.75) leads workers to favor more sisters
In multiply-mated colonies, lower relatedness (r ≈ 0.5) reduces worker incentive to bias sex ratios
🧠 The observed change in sex ratio from egg-laying to adult stage reflects worker control and genetic interest.
A form of cooperation where individuals help others with the expectation of future help in return.
🧠 It can evolve if the benefit of help received > cost of help given and interactions are repeated.
When individuals interact repeatedly
When they can remember past interactions
When cheaters can be identified and punished
✅ Trust + memory = cooperation
Bats regurgitate blood meals to feed hungry roost-mates
They do this mostly with bats they’ve shared with or received from before
Bats that don’t reciprocate are eventually excluded from the group
🧠 Give today, get tomorrow—unless you cheat.
No—food sharing occurs even between unrelated individuals.
🧠 Bats share irrespective of kinship, though many roost-mates are kin, so kin selection may have contributed to the behavior’s evolution.
They starved 20 bats of known pedigree and observed which bats shared food with them, to measure the influence of relatedness and reciprocal history.
The strongest predictor was having previously received food from that individual—not relatedness.
✅ Suggests reciprocal altruism is the main driver, with kin selection playing a supporting role.
Protection from predators or harsh environments (e.g., huddling penguins)
Improved resource access through group hunting (e.g., sailfish hunting in packs)
Future opportunities (e.g., taking over territory or mating roles)
In long-tailed manakins, two males perform coordinated dances:
Only the alpha mates
The junior male inherits the display ground when the alpha dies
🧠 Short-term cost for long-term gain.
Because kin selection may still be at play.
✅ Even when cooperation looks self-serving, relatives may benefit, so inclusive fitness could still explain the behavior.
Individuals may cooperate overall, but still compete over key resources like:
Food
Mates
Access to breeding
🧠 Cooperation doesn’t eliminate competition—just changes its context.
It depends on a cost-benefit analysis:
Fighting may bring rewards (e.g., dominance, mates)
But it also comes with potential injury or death, especially for weaker individuals
🧠 Individuals may back down if the cost of losing is too high.
Using evolutionary game theory, such as the Hawk-Dove model, which compares:
Aggressive (“Hawk”) strategies
Peaceful (“Dove”) strategies
✅ Helps predict which behaviors are evolutionarily stable based on costs and benefits of conflict.
An ESS is a strategy that, if adopted by most of the population, cannot be invaded by an alternative (mutant) strategy because it yields the highest average fitness.
🧠 It’s not about winning every fight—it’s about long-term success.
Pure strategy: individual always uses the same behavior (e.g., always fight or always flee)
Mixed strategy: individual varies behavior based on the situation or frequency of other strategies in the population
✅ Real animals often use mixed strategies.
A simple model of conflict:
Hawk: always fights aggressively, risking injury
Dove: avoids conflict, backs down when challenged
The payoff depends on benefit of winning (V) and cost of fighting (C)
🧠 When C > V, a mix of hawks and doves can be stable.
Encounter Outcome
Dove vs Dove No fight → they share the resource → each gets V/2
Hawk vs Dove Hawk wins all → Hawk gets V, Dove gets 0
Hawk vs Hawk Fight → each has 50% chance to win → average payoff = (V − C)/2
When the cost of fighting (C) is greater than the value of the resource (V), a mixture of Hawks and Doves in the population becomes stable.
✅ Too many Hawks = too costly → Doves do better
Too many Doves = Hawks exploit them → balance evolves
Player Opponent Payoff
Hawk Hawk (V − C)/2
Hawk Dove V
Dove Hawk 0
Dove Dove V/2
V = fitness benefit (value of the resource)
C = fitness cost of fighting
Fighting becomes too costly, so:
Pure Hawk strategy is not stable (too much injury)
Pure Dove strategy is not stable (gets exploited by Hawks)
✅ Result: a mixed ESS evolves with both Hawks and Doves in the population
To find it: set average payoff of Hawk = average payoff of Dove:
Hawk: p(Hawk) × (V − C)/2 + p(Dove) × V
Dove: p(Hawk) × 0 + p(Dove) × V/2
Solving yields:
📊 Equilibrium frequency of Hawks = V/C
🧠 More valuable the resource or lower the cost, more Hawks in the population.
Because if a Hawk mutant appears, it wins every encounter against Doves and gains full benefit (V), while Doves only share (V/2) among themselves.
🧠 Hawk invades easily → Dove is not an ESS.
Because Hawk vs. Hawk results in costly fights, with average payoff of (V − C)/2—which is lower than V/2, the Dove-Dove outcome.
🧠 Too many Hawks = everyone suffers → not stable.
Because Hawk vs. Hawk results in costly fights, with average payoff of (V − C)/2—which is lower than V/2, the Dove-Dove outcome.
🧠 Too many Hawks = everyone suffers → not stable.
A mixed ESS where both Hawks and Doves exist.
Rare Hawks do well because they beat Doves
But if Hawks become too common, their fitness drops due to costly fights
📊 Stable proportion of Hawks = V / C
✅ Natural selection favors a balance between aggression and avoidance.
Yes—if V > C, then (V − C)/2 > 0, so Hawks do well even when fighting each other.
✅ In that case, an all-Hawk population can be stable.
Fighting is too costly, so pure Hawk isn’t stable
But pure Dove also isn’t stable—Hawks can still invade
➡️ So neither pure strategy works
A mixed strategy where individuals play Hawk with probability V/C
📊 On average, ~80% Hawk and 20% Dove if V = 0.8C
🧠 As found by Maynard Smith (1982)—this balances costs and benefits of aggression in the population.
It assumes individuals can’t assess whether their opponent will fight or flee—they just choose a strategy blindly.
🧠 In reality, animals often evaluate their opponent before acting.
A strategy where individuals assess the opponent’s strength or condition before deciding to fight or retreat.
✅ Often leads to less costly and more efficient outcomes than fixed strategies.
Because it allows individuals to avoid unnecessary conflict:
Weak individuals back down
Strong individuals escalate when odds are good
📈 Leads to greater average fitness than always fighting or choosing randomly.
🧠 Think before you fight = win more, lose less.
🟧 Orange males: Aggressive, large territories, many females (dominant)
🔵 Blue males: Less aggressive, small territory, guards one female
🟨 Yellow males: Sneaky, mimic females, steal matings from orange males
🧠 Each morph has a distinct reproductive strategy.
Orange > Blue: Overpowers them and steals mates
Blue > Yellow: Blue guards mates well
Yellow > Orange: Sneaks in while orange is distracted
🧠 No single morph wins long-term—success depends on who’s common or rare.
Negative frequency-dependent selection:
Each morph does best when it’s rare
Fitness drops when it becomes common, so frequencies cycle over time
✅ This keeps all three strategies in the population.