Kin Selection and Social Behavior

1
Kin Selection and Social Behavior

• Chapter 11

2
Types of social interactions among members of the
same species (Table 11.1)

• The actor in any social interaction affects the
  recipient of the action as well as himself. The
  costs and benefits of interactions are measured
  in units of surviving offspring (fitness).

Actor benefits Actor is harmed
Recipient benefits Cooperative Altruistic
Recipient is harmed Selfish Spiteful
3
Kin selection and altruistic behavior

• For Darwin, the apparent existence of altruism
  presented a special difficulty, which at first
  appeared to me insuperable, and actually fatal to
  my whole theory.
• However, he also suggested a solution selection
  might favor traits that decreased the fitness of
  the actor if they increased the reproductive
  success of close relatives
• This form of selection, which takes into account
  the fitness benefits to relatives is kin selection

4
Hamiltons Rule

• William Hamilton (1964) developed a genetic model
  showing that an altruistic allele could increase
  in frequency if the following condition is
  satisfied
• Br - C gt 0
• Where B is the benefit to the recipient and C is
  the cost to the actor, both being measured in
  units of surviving offspring, and r is the
  coefficient of relatedness (or relationship)
  between actor and recipient

5
Inclusive fitness

• Hamilton introduced the concept of inclusive
  fitness, which includes direct fitness indirect
  fitness
• Direct fitness is personal reproduction
• Indirect fitness is the additional reproduction
  of relatives that is made possible by an
  individuals actions

6
The coefficient of relatedness, r

• The proportion of alleles in two individuals that
  are identical by descent (ibd)
• The coefficient of relatedness of full sibs is r
  0.5
• To see why this is so, we can use the following
  example, in which we give each of the four
  alleles at a locus in the two parents a unique
  label

7
Proportion of alleles shared ibd by full sibs

• P A1A2 x A3A4
• O
• The average proportion of alleles shared ibd by
  pairs of full sibs is 0.5

Sib 1 Sib 1 Sib 1 Sib 1
A1A3 A2A3 A1A4 A2A4
Sib 2 A1A3 1.0 0.5 0.5 0
Sib 2 A2A3 0.5 1.0 0 0.5
Sib 2 A1A4 0.5 0 1.0 0.5
Sib 2 A2A4 0 0.5 0.5 1.0
8
Some coefficients of relatedness

• Parent to offspring, r 1/2
• Full sibs, r 1/2
• Half sibs, r 1/4
• First cousins, r 1/8
• Grandparent to grandchild, r 1/4
• Aunt or uncle to niece or nephew, r 1/4

9
Altruistic behavior in Beldings ground squirrels

• Breed in colonies
• Male offspring disperse far from native burrow
• Female offspring tend to remain and breed close
  by. Therefore, females in proximity tend to be
  closely related
• Squirrels give alarm calls when predators are
  spotted (different calls for mammalian predators
  vs. birds of prey)
• Is alarm calling altruistic and can it be
  understood as a result of kin selection?

10
In ground squirrels most alarm calling is done by
females (Sherman 1977) (Fig. 11.2 b)

• Based on 102 encounters with predatory mammals
• Blue line is expected frequency if each type of
  individual called in proportion to the number of
  times it was present when a predator appeared
• Mortality is 8 for calling individual vs. 4 for
  non-callers when predator is a mammal

11
Female ground squirrels are more likely to give
alarm calls when close kin are nearby (Sherman
1977) (Fig. 11.3)

• Based on 119 encounters with predatory mammals
• Blue line is expected frequency if each type of
  pairing produced calls in proportion to the
  number of times it was present when a predator
  appeared

12
Nest helping in white-fronted bee-eaters

• In white-fronted bee-eaters (and some other birds
  where breeding opportunities are extremely
  restricted), young adults often forego their own
  reproduction to help at the nests of other
  individuals.
• This is clearly altruistic. Evidence suggests
  that nest helping can be explained by kin
  selection

13
White-fronted bee-eater (Merops bullockoides)
14
In bee-eaters, helpers assist close relatives
(Emlen and Wrege 1988) (Fig.

1. Among non-breeders, those born in a clan are much
   more likely to be nest helpers than those who
   enter a clan from outside and are unrelated to
   offspring being raised in that season
2. Nest helping is disproportionately directed
   toward close relatives

15
Nest helpers increase the number of young birds
fledged (Emlen and Wrege 1991) (Fig. 11.7)

• A group size of 2 represents a pair without
  helpers. The average number of young fledged
  0.51 for pairs
• Each additional helper at the nest increases the
  number of young fledged by 0.47 on average

16
Kin-selected discrimination in cannibalistic
spadefoot toad tadpoles (Pfennig 1999) (Fig. 11.8
a, b)

• Tadpoles develop into typical morphs that eat
  mostly decaying plant matter or into carnivores
  that eat other tadpoles.
• Carnivores are more likely to eat non-sibs than
  sibs when given a choice between one of each kind

17
Kin selection can explain the presence of
discrimination between sibs and non-sibs in
cannibalistic tiger salamander larvae (Pfennig et
al. 1999) (Fig. 11.8c)
Benefit B 2
Cost C 0
Experiment consisted of placing 1 predatory morph
6 sibling typical morphs 18 non-sibling
typical morphs in each of 18 enclosures in
natural pond.
18
Coots can avoid parasitic altruism (helping
non-kin) (Lyon 2003) (Fig. 11.10 c, d)

• If selection can favor helping kin, it should
  also favor avoiding sacrifices for non-kin
• Coots that accept eggs from other birds lose 1
  offspring of their own for each parasitic
  offspring
• Coots that reject parasitic eggs have the same
  number of offspring as unparasitized birds
  (dashed line)

19
Eusociality the ultimate in reproductive altruism

• Characteristics of eusociality
• Overlap in generations between parents and
  offspring
• Cooperative brood care
• Specialized castes of non-reproductive
  individuals
• Insects (termites, hymenoptera), snapping shrimp,
  naked mole rats

20
Haplodiploidy and eusociality in hymenoptera
(bees, wasps, ants)

• Males are haploid (develop from unfertilized
  eggs) and females are diploid
• Hamilton (1972) proposed that haplodiploidy
  predisposes hymenoptera to eusociality because
  females are more closely related to one another
  (r 3/4) than they are to their own offspring (r
  1/2)
• Females may maximize inclusive fitness by being
  sterile workers and helping to produce
  reproductive sisters (rather than by being
  reproductives themselves)

21
Proportion of alleles shared ibd by sisters in a
haplodiploid species

• P A1A2 x A3
• O
• The average proportion of alleles shared ibd by
  pairs of full sibs is 0.75

Sister 1 Sister 1
A1A3 A2A3
Sister 2 A1A3 1.0 0.5
Sister 2 A2A3 0.5 1.0
22
Is haplodiploidy the explanation of eusociality
in hymenoptera?

• Probably not
• The preceding analysis assumed only 1 male
  fertilizes a queen this is not true in
  honeybees, for example
• In some species, colonies may be founded by more
  than 1 queen
• Many eusocial non-hymenoptera are diploid (e.g.,
  termites)
• Many hymenoptera are not eusocial (eusociality
  may have three independent origins associated
  with nest-building and the need to supply larvae
  with food)
• Haplodiploidy may facilitate the evolution of
  eusociality but a more important factor may be
  the need for help in rearing young

23
Sociality and nesting behavior in hymenoptera
(Hunt 1999) (Fig. 11.13

• Families that include eusocial species are
  indicated in boldface type

24
Naked mole rats

• All young in a colony produced by a single queen
  and 2 3 reproductive males
• Not haplodiploid, but colony members are highly
  inbred (average r 0.81)
• 85 of matings are between full-sibs or parents
  and offspring
• Queens use physical dominance to coerce help from
  less closely related individuals

25
Naked mole rat queens preferentially shove
nonrelatives (Reeve and Sherman 1991) (Fig. 11.16)
26
Parent offspring conflict

• Parental care is a special case of providing
  fitness benefits for close relatives
• The offspring is the fitness of the parent (this
  means that benefits and costs of parental care
  both accrue to the parent)
• In species that provide extensive parental care,
  the fitness benefit to the parent of providing
  additional care to current offspring needs to be
  weighed against the fitness cost of that
  additional care in terms of lost future offspring

27
Parent offspring conflict occurs because
parents and offspring value the costs of parental
care differently (Trivers 1985) (Fig 11.18)

• As offspring grow, the benefit/cost ratio for the
  parent declines. Benefit (B) is measured in
  terms of increased survival of offspring
  receiving parental care cost (C) is measured in
  terms of lost future offspring due to continued
  parental care. From parents point of view, it
  should stop giving parental care when B/C
  declines to 1.
• But, the offspring devalues the cost to the
  parent by 1/2 because lost future full-sibs are
  related to the offspring by r 1/2. Therefore,
  the offspring wants parental care to continue
  until the B/C ratio for the parent is 1/2 - fig.
  (a) (or 1/4 if future offspring are half-sibs -
  fig. (b)

28
Harassment in white-fronted bee-eaters can also
be analyzed in the context of parent-offspring
conflict and kin selection

• Fathers occasionally coerce sons into helping to
  raise their siblings by harassing sons who are
  trying to raise their own young
• Sons are as closely related to their full sibs as
  they are to their own offspring (r 1/2 in both
  cases)
• Furthermore the average number of offspring in
  nests without helpers is 0.51, whereas every
  helper increases the number of surviving
  nestlings by 0.47 on average
• Therefore, the direct fitness lost by a son who
  is coerced into helping his parents is balanced
  by the increase in indirect fitness that results
  from helping his parents (provided that the son
  would not have had helpers)

29
Bee-eaters recruit helpers who are younger and
closely related (Emlen and Wrege 1992) (Fig.
11.19 b)
30
Reciprocal altruism

• Reciprocation is offered to explain altruism
  between unrelated individuals
• The necessary conditions for reciprocal altruism
  to evolve are
• The fitness cost to the actor must be the
  fitness benefit to the recipient
• Non-reciprocators must be punished in some way
  (otherwise alleles that caused cheating would
  displace alleles for altruism)
• Conditions that favor the evolution of reciprocal
  altruism are
• Stable social groups (so that individuals are
  involved in repeated interactions with one
  another)
• Lots of opportunities for altruistic
  interactions during an individuals lifetime
• Good memory
• Symmetry of interactions between potential
  altruists

31
Blood-sharing in vampire bats an example of
reciprocal altruism? (Wilkinson 1984) 1

• Reciprocal altruism has been difficult to
  document
• In vampire bats (Desmodus rotundus) the basic
  social unit is 8 12 females and their dependent
  offspring that frequently roost together
• Many individuals preferentially associate with
  one another when roosting
• The altruistic act is sharing blood meals by
  regurgitation
• 33 of young and 7 of adults fail to get a blood
  meal on any given night
• Bats are likely to starve to death if they go
  three consecutive nights without a meal

32
Blood-sharing in vampire bats an example of
reciprocal altruism? (Wilkinson 1984) 2

• Bats are more likely to share blood with
  relatives and with associates. Both effects
  (relationship and association) were statistically
  significant. (These data do not include
  regurgitation by mother to child because that is
  parental care.)
• Blood sharing was not random in a group of
  captive individuals. Bats were more likely to
  receive blood from and individual that they had
  fed previously.

33
Association, relatedness and altruism in vampire
bats (Wilkinson 1984)