Behavioral Ecology Vol. 13 No. 3: 291-300
© 2002 International Society for Behavioral Ecology
The evolution of parental and alloparental effort in cooperatively breeding groups: when should helpers pay to stay?
a Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, UK b Division of Environmental and Evolutionary Biology, Institute of Biomedical & Life Sciences, University of Glasgow, UK c School of Biological Sciences, University of Wales, Bangor, Gwynedd LL57 2UW, UK
Address correspondence to H. Kokko, Division of Environmental and Evolutionary Biology, Graham Kerr Building, University of Glasgow, Glasgow G12 8QQ, UK. E-mail: h.kokko{at}bio.gla.ac.uk .
Received 6 July 2000; revised 28 November 2000; accepted 5 February 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONREFERENCES |
|---|
We used a reproductive skew framework to consider the evolution of parental and alloparental effort in cooperatively breeding groups. The model provides the first theoretical treatment of rent payment (the "pay-to-stay" hypothesis) for the evolution of helping behavior of subordinates. According to this hypothesis, not all helping behavior is kin selected, but group members help in order to be allowed to stay in the group and potentially gain breeding positions later in life. We show that reproductive concessions may be replaced by complete skew and voluntary, costly alloparental effort by subordinates once future prospects are included in fitness calculations. This suggests that incomplete skew observed in long-lived species is not due to dominant control over reproduction. Rent payment is predicted to occur when relatedness between subordinate and dominant is low, survival is high, ecological constraints are at least moderately tight, and retaining nonhelping subordinates harms the dominant's fitness. Rent may also be required from related subordinates if helping is very costly (leading to low voluntary helping effort) and ecological constraints are moderately tight. However, related subordinates do not need to have a positive net effect on the dominant's direct fitness to be accepted as group members. We also consider compensatory responses of dominant group members as a potential threat to the stability of renting behavior. If dominants trade off parental effort against their own survival, they may selfishly reduce their own parental effort as a response to increased help. As this improves their own survival, the prospects of territorial inheritance diminish for the subordinate, and subordinates should hence be less willing to accept the rent agreement. However, we show that compensatory responses by "lazy" parents prevent group formation only in borderline cases.
Key words: alloparental care, cooperative breeding, helping at the nest, reproductive skew.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONREFERENCES |
|---|
The apparent altruism of helping—parentlike behavior toward young that are not the genetic offspring of the helper—has inspired a number of alternative, but not mutually exclusive, evolutionary explanations (see Cockburn, 1998
;
Emlen and Wrege, 1989;
Wright, 1997). The most
well-known of these is kin selection
(Hamilton, 1964;
Maynard Smith, 1964), in which
patterns of helping in many cooperatively breeding species are explained via
indirect fitness benefits that helpers obtain from provisioning young to which
they are related (see Bourke and Franks,
1995; Brown, 1987;
Emlen, 1991). In addition, by
investing in the production of younger members of their social group, helpers
may obtain mutual benefits via improved future survivorship and/or
reproduction upon inheritance of a breeding position (i.e., pseudoreciprocity
or group augmentation; Brown,
1983,
1987; Connor,
1986,
1995;
Kokko and Johnstone, 1999;
Kokko et al., 2001;
Ligon, 1981;
Ragsdale, 1999; Woolfenden and
Fitzpatrick, 1978,
1984).
Among the various other hypotheses for helping is the intriguing suggestion
that it represents the payment of a "rent" to dominant group
members, and that subordinate helpers pay to stay in order to secure group
membership and its associated benefits
(Gaston, 1978;
Kazem and Wright, in press).
In cooperatively breeding birds and fish, these benefits can include access to
a communal territory, reduced susceptibility to predation, or enhanced intra-
or extra-group mating opportunities
(Balshine-Earn et al., 1998;
Dunn et al., 1995;
Gaston, 1978; Reyer,
1980,
1984). Any lack of effort on
the part of the helper can be penalized via aggression from dominant breeders
(e.g., superb fairy wrens, Malurus cyaneus;
Mulder and Langmore, 1993),
ultimately culminating in expulsion from the group. Dominant breeders should
only tolerate helpers when they are needed (e.g., pied kingfishers, Ceryle
rudis; Reyer, 1980,
1984). Therefore, below some
minimum level of helping effort, it is not worthwhile for dominants to allow
helpers in their group because of the potential reproductive and/or foraging
competition that they represent (e.g., Florida scrub-jays, Aphelocoma
coerulescans; Goldstein et al.,
1998).
An assessment of the possibility of rent agreements between helpers and
dominants should take into account the relative costs and benefits to the
helper of membership of alternative groups within the population (Vehrencamp,
1979,
1983). Renting will not be
evolutionarily stable if the subordinate benefits more by leaving the group
than by providing the help required. In this respect there are obvious
parallels with reproductive skew theory
(Reeve, 1998;
Johnstone, 2000). Here we use
this modeling framework to develop the first formal treatment of the evolution
of paying rent. The pay-to-stay hypothesis, or "renting" as we
call it for simplicity, is intrinsically linked to the decision of staying in
a group. Most models of cooperative breeding have simply linked the decision
to stay and the decision to help together by assuming that a retained
subordinate automatically boosts the productivity of the group (e.g.,
Motro, 1993;
Pen and Weissing, 2000;
Reeve, 1998; but see also
Johnstone and Cant, 1999;
Kokko and Johnstone, 1999).
However, it is clear that the benefits that a staying subordinate brings to
the group will depend on its behavior, particularly on its eagerness to help.
Thus, one of our goals in this study was to make the distinction between the
staying and helping decisions within a modeling framework—a task whose
importance empiricists have acknowledged for a long time (see
Brown, 1987; Emlen,
1991,
1997).
Much of reproductive skew theory has focused on concessions, where
dominants allow subordinates to have a share in reproduction. We include the
possibility of concessions in our model, but our main focus is on coercive
solutions (see also Crespi and Ragsdale,
2000), where subordinates do not reproduce and are instead
required to help the dominant. We show that nonconcessive solutions can
prevail, especially in long-lived species, in which indirect fitness and/or
future fitness expectations can provided a reason for nonreproductive
subordinates to remain as helpers (Ekman
et al., 1999; Kokko and
Johnstone, 1999; Pen and
Weissing, 2000; Queller et
al., 2000; Ragsdale,
1999; Stacey and Ligon,
1991).
We also consider a mechanism that might potentially hinder the evolution of
paying rent. Parental effort often trades off with subsequent survival of the
parent (Trivers, 1972; see
also Clutton-Brock, 1991).
Therefore, a dominant may respond to the presence of a helping subordinate by
decreasing its own parental effort
(Hatchwell, 1999;
Hatchwell and Russell, 1996;
Houston and Davies, 1985;
Legge, 2000;
Wright and Cuthill, 1989;
Wright and Dingemanse, 1999).
If this improves the dominant's survival, then the prospects of territorial
inheritance may diminish for the subordinate that provides help.
The model
We evaluated the fitness of group members in the setting of Kokko and
Johnstone (1999), in which
individuals may be either alone or in a group comprising a dominant and a
subordinate. If the dominant dies, the subordinate inherits its territory.
This queuing for dominance establishes an incentive to stay that is often
enough to make the subordinate willing to remain without any direct immediate
fitness benefits such as reproductive concessions
(Kokko and Johnstone, 1999;
Ragsdale, 1999). For the sake
of completeness, we retain the possibility of concessions in the model, but we
show that at equilibrium, concessions can often equal zero in stable groups. A
list of symbols and their explanations is provided in
Table 1.
|
To consider the possibility of rent payment, we extend the model by Kokko
and Johnstone (1999) by
assuming that the dominant and the subordinate can decide independently on the
effort, hD and hS,
that they put into raising offspring. Additionally, we assume that the
presence of a nonhelping subordinate changes the productivity of the group by
hO. Typically, nonhelping subordinates would decrease
group productivity (hO < 0); this would occur as they
consume resources of the territory (Brown,
1987). However, positive values of hO are
possible—for example, if a subordinate aids in predator detection, even
if it does not provide active altruistic help
(Clutton-Brock et al., 1999;
Connor, 1986,
1995;
Hamilton, 1971; Wright et al.,
in preparation). We contrast the passive effect of the subordinate,
hO, with active helping by the subordinate,
hS. The latter always increases the productivity
of the group (hS > 0). Finally, the dominant
may also adjust its own parental effort, and therefore group productivity also
depends on the effort, hD, of the dominant. Thus,
group productivity, k, equals hD +
hO + hS. Reproduction is
shared among group members so that the subordinate produces kp
offspring, and the dominant produces k(1 - p) offspring. The
productivity of a lone individual depends only on its own parental effort:
kL = hL. Between two
breeding attempts, a lone individual is joined by a subordinate with a
probability a, as in Kokko and Johnstone
(1999).
The effort to raise offspring is costly for the individual, and survival,
s, between breeding attempts therefore decreases with increasing
effort, hL, hD, or
hS. We use a function which allows for varying
costs of helping effort,
 |
This function implies that survival has its maximum value
Smax when no effort is put into raising offspring
(h = 0) and drops to zero at h = 1. We assume that
alloparental and parental effort are equally costly. The parameter 
scales the cost of giving small amounts of help. With large , small or
moderate effort levels are relatively cheap, and the cost of raising
off-spring increases sharply only at high levels of effort, h. With
small , helping is costly even at small help levels, h (see also
Kokko et al., 2001). Because
large implies that helping is cheap, we refer to as the ease of
helping.
Kokko and Johnstone (1999)
derived the direct lifetime fitness, wD, wS,
and wL of dominants, subordinates, and lone individuals,
respectively. In each equation, the fitness of an individual equals the sum of
current reproduction [e.g., k(1 - p) for the dominant] and
the fitness of other states, scaled by the probability of ending up in these
states after the current breeding attempt. For example, a dominant becomes a
lone individual if it survives, the subordinate dies, and no new subordinate
arrives. To examine the evolution of helping, we take equation 2 of Kokko and
Johnstone (1999) and
substitute hD + h0 +
hS for the productivity of the group and
hL for the productivity of the lone individual. This
yields
 | (1) |
).
We first examine the levels of parental effort that evolve voluntarily
(that is, without any coercion by the dominant). We thus ask the question, how
much would a subordinate help, if it had decided to stay in the group and if
its acceptance in the group was independent of its help level? Similarly, we
seek the dominant's optimal effort, given that it has a subordinate that
decides independently on its effort. Equation 1 yields solutions for
wD, wS, and wL for each
combination of effort levels hD, hS, and
hL (solutions derived in the same way as in
Kokko and Johnstone, 1999). We
seek the best response (cf. Houston and
Davies, 1985) of a dominant by maximizing its inclusive fitness,
WD = wD +
rDwS:
 | (2a) |
 | (2b) |
 | (2c) |
In principle, the evolutionarily stable strategy is found by seeking the
values of parental effort hD hS and
hL for which it does not pay for any individual to alter
its effort. Mathematically, such values have to satisfy
dWD/dh'D = 0 at
h'D = hD,
dWS/dh'S = 0 at
h'S = hS and
dWL/dh'L = 0 at
h'L = hL. Equations 2a-c,
unfortunately, do not yield an analytical solution. The evolutionarily stable
effort values 
and 
are therefore obtained by iteration,
where new effort values are a weighted sum of the previous prevailing effort
and the new best response [e.g., hD'' =
hD + (1 - )h'D.
In practice, the iteration converges quickly (e.g., with = 0.5), and
this value was used in the calculations.
When does rent-paying apply?
A dominant may potentially demand more help from a subordinate, whose
voluntary help effort equals 
(note that
may equal 0). We consider the rule where a
dominant evicts a subordinate if its effort falls below H. Renting
can be stable only if it is more beneficial for a subordinate to stay and
spend the effort, H, than to leave. To evaluate the stability of
renting, there are hence two values of effort that need to be specified: (1)
What is the smallest value of H that the dominant accepts
(Hmin)? (2) What is the highest value of H that
the subordinate agrees to pay (Hmax)?
To find Hmin and Hmax, we need to
take into account that the best effort of dominants and of lone individuals
will depend on the help given by subordinates (Equations 2a,c). The
calculation of Hmin and Hmax proceeds
as follows: Substitute the subordinate's effort hS in
Equation 2b,c with H, and let H vary from 0 to 1. Seek the
fitness-maximizing values of hD and hL
according to Equation 2b,c for each H. Equation 1 then yields stable
values of wL, wD, and wS
for each H. We want to find the range of acceptable values of
H from the dominant's and subordinate's point of view. Assuming (as
in reproductive skew models in general;
Johnstone, 2000;
Reeve, 1998) that a dispersing
subordinate finds a breeding vacancy and becomes a lone breeder with
probability x, the dominant benefits from retaining the subordinate
if
 | (3a) |
 | (3b) |
Hmin >
, subordinates accept paying rent.
There are four ways in which renting may be replaced by other types of
solutions. First, renting may be replaced by voluntary helping if the
voluntary effort by the subordinate, ,
satisfies Equations 3a,b. Such solutions are characterized by
. Second, if the minimum
effort required by the dominant exceeds the maximum that subordinates are
willing to pay (Hmin > Hmax),
renting is replaced by three alternatives, depending on the value of
. If at
subordinates and dominants both benefit from dispersal of the subordinate
Equation 3, the group simply disbands. If the subordinate benefits from
staying at (but not at
Hmin), and if the dominant does not benefit from retaining
the subordinate Equation 3, the dominant is expected to evict the subordinate.
Finally, if the subordinate should leave at
(Equation 3b) but the dominant would benefit from retaining the helper
(Equation 3a), the situation has potential for staying incentives. Here, it
may pay for the dominant to give a share of the reproduction to the
subordinate (p > 0), to make staying and helping the preferred
option. As our primary focus is on renting rather than on incentives, we do
not solve the value of concessions p needed to stabilize the group
(these would interact with the evolution of levels of helping, which makes the
solution complicated), nor do we check whether dominants are willing to accept
this increase in p. We merely note that in this last case, renting is
excluded as an outcome, and it is replaced either by incentives or a failure
of group formation.
Comparison to fixed parental effort by dominants
If survival trades off against parental effort, dominants may respond to
increased effort by rent-paying subordinates by reducing their own effort. If
"lazy" dominants survive better, selfish dominants reduce the
subordinates' prospects of territorial inheritance. To investigate whether
this mechanism has a strong effect, we compared group stability in the above
model to a hypothetical case where the dominant can set a minimum effort
Hmin as above, but the dominant's own effort is fixed to
—the best response to a subordinate who
does not pay rent (i.e., uses the effort ).
The subordinate is free to choose any effort greater than
Hmin or to leave if the Hmin is
unacceptable to its. The dominant's effort, ,
is clearly evolutionarily unstable in this scenario because it is not the best
response to the effort actually used by the subordinate. Yet, considering
group formation under the artificial absence of adaptive effort adjustment of
dominants allows us to evaluate the effect that this behavior has on group
stability.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONREFERENCES |
|---|
Stable groups are often found at complete skew with no reproductive concessions (p = 0). For example, consider the case with maximum survival smax = 0.75, ease of helping
= 2,
probability that a lonely breeder gains a helper a = 0.5, effect of
nonhelping group member on productivity h0 = -0.2,
relatedness between dominant and subordinate r = 0.25, and complete
skew p = 0. The model predicts evolutionarily stable effort by
subordinates 
and predicts that the
subordinate would rather pay this than leave the group to breed on its own.
The effort is not sufficient to
compensate completely for the negative effect on productivity,
h0 = -0.2, that the subordinate causes simply by being
present. Yet, because the dominant shares an interest in the related
subordinate's future, it accepts this level of effort rather than evict the
subordinate, whenever the probability that the subordinate finds a breeding
position elsewhere falls below x = 0.50
(Figure 1).
|
That the subordinate stays with no concessions (p = 0) is in line
with the recognition that territorial inheritance or other benefits of
philopatry (see Brown, 1987)
remove or reduce the need for reproductive concessions
(Kokko and Johnstone, 1999;
Ragsdale, 1999). Indeed, the
above example shows that it is the dominant, rather than the subordinate, who
is the first to benefit from the dispersal of the subordinate when the
subordinate's dispersal prospects improve. When the probability of a
subordinate finding a breeding position exceeds x = 0.50, the
dominant would rather have the subordinate dispersing than staying and helping
at , while x has to reach 0.57
before dispersal becomes the preferred option for the subordinate
(Figure 1). This means that our
model captures the essential conflict that precedes paying rent: subordinates
often benefit more strongly from staying in a group than dominants benefit
from retaining subordinates.
Nevertheless, the situation does not immediately translate into paying
rent. Rent is paid if the maximum effort that subordinates are willing to pay,
Hmax, equals or exceeds the minimum that dominants
require, Hmin, and if subordinates would not pay this much
without the rent requirement,
. In
Figure 1, this situation only
applies in a narrow range of ecological constraints, between x = 0.50
and 0.57. When ecological constraints are tighter (x < 0.50), both
the dominant and the subordinate benefit more from group formation than from
the dispersal of the subordinate. The subordinate is allowed to stay, and its
fitness is improved by prospects of territorial inheritance, but helping is
nevertheless voluntary and based on indirect fitness benefits rather than on
renting. On the other hand, when independent breeding is not strongly
constrained (x > 0.57), the subordinate's expected success by
independent breeding is greater than the benefits of staying, and it will not
stay as a helper or even as a nonhelping subordinate
(Figure 1).
We now turn to the effect of different parameters on the prospects of renting.
Renting requires low relatedness or high costs of helping
Voluntary help levels increase with
relatedness, rS, whereas effort requirements,
Hmin, decrease with rD, as dominants
have increasing interest in the subordinate's survival and future
reproduction. Therefore, renting is unlikely in kin groups, as it becomes
replaced by voluntary, kin-selected helping when
exceeds Hmin
(Figure 2).
Figure 2 also shows that
individuals of different relatedness to the dominant may exhibit similar
levels of alloparental effort, but for different reasons. For individuals with
low relatedness to the dominant, voluntary effort is low, but renting may
apply (in Figure 2,
rD = rS = 0 leads to rent 0.2). Highly
related subordinates do not need to pay as much to be allowed to stay, but
indirect fitness benefits can favor an increase in helping effort
(Figure 2:
rD = rS = 0.25 leads to voluntary
helping 
).
|
; Hmax
not shown for clarity). Rent is only required
when relatedness is low,
and full compensation hS = 0.2 is only required if
dominant-to-subordinate relatedness (rD) = 0. Three
specially marked points indicate help levels in sister—sister
association under diploidy (square: rD =
rS = 0.5, 
,
Hmin = 0.102), sister—sister association under
haplodiploidy (dot: rD = rD = 0.75,
, Hmin = 0.078), and
mother—daughter association under either diploidy or haplodiploidy, when
the mother is monogamously mated (star: rD = 0.5,
rS = 1, hS = 0.866,
Hmin = 0.124).
Because voluntary helping increases with the subordinate's relatedness to
the dominant (rS), renting is especially unlikely in
mother—daughter associations, where relatedness asymmetry increases the
effective relatedness of the daughter to the mother. If the mother is still
mated to the daughter's father and hence produces full sibs for the daughter,
the daughter's relatedness to the mother is effectively rS
= 1, while the mother's relatedness to the daughter remains at
rD = 0.5 (Reeve and
Keller, 1995). We conclude that mother—daughter groups
should exhibit higher voluntary helping effort than sister—sister
associations, and they should be less likely to require renting for the
maintenance of group stability. With sufficiently monogamous mothers, this
applies even if sisters are more related to each other than mothers to their
daughters (Figure 2).
Highly related subordinates may, however, be required to pay rent if their
willingness to provide help voluntarily is reduced. Such a reduction may be
caused by high costs of helping behavior. Subordinates and dominants are
asymmetric in their prospects of current versus future fitness. In
subordinates that help while waiting to inherit a breeding position, the
future represents a major fitness component, and they are expected to be more
sensitive to survival costs of current helping effort. Therefore, if survival
costs of helping effort are high ( is low), the model predicts a strong
asymmetry in the amount of care provided: reproductive dominants show much
more effort than the nonreproductive subordinates. If some degree of effort is
relatively cheap (indicated by high ), effort is more evenly distributed
among reproductive and non-reproductive group members
(Figure 3). To summarize,
decreasing the cost of helping (increasing ) shifts some part of
parenting effort from dominants to helpers. Because minimum effort
requirements by dominants do not appear to respond strongly to costs of
helping (Figure 3), the net
effect is that renting in kin groups is more likely if helping is very
costly.
|
Renting requires high survival
According to life-history theory, a long life span means that the relative
importance of future fitness increases compared with the current reproductive
event (e.g., Roff, 1992). This
has several implications for the evolution of renting. Benefits of philopatry,
such as the prospects of inheriting a territory, are of greater importance in
species with high survival (Kokko and
Johnstone, 1999; Pen and
Weissing, 2000). Therefore, the willingness of subordinates to
stay and queue for breeding positions is stronger if survival,
Smax, is high. In addition, renting requires that the
voluntary effort by subordinates is low, as it is otherwise replaced by
voluntary helping. Life-history theory predicts that individuals with a long
life span should be less willing to trade off their survival for a fixed
current benefit (Roff, 1992).
Thus, both parental and alloparental effort decrease with increasing survival
(Figure 4), which enhances
prospects for renting in long-lived species. Yet rent requirements may also
decrease when survival improves: in long-lived species, a decrease in current
productivity becomes less important for the parent than ensuring that the
(related) subordinate has good prospects to inherit the territory. However,
this drop in the minimum effort requirement is less strong than changes in
voluntary effort levels, so that the net effect is that long-lived species are
more likely to exhibit renting behavior
(Figure 4).
|
at
intermediate survival, 0.34 < Smax < 0.66, and rent
is paid In long-lived species, potential benefits of increasing life span are also greatest, and we might expect that long-lived dominants reduce their own effort as a response to rent payment by subordinates. However, with the parameters of Figure 4, dominants would spend maximally only 5% more effort if they were unable to adjust their behavior to increased help by subordinates (at Smax = 0.9; for clarity, evolutionary unstable efforts are not shown in Figure 4). Likewise, the maximum effort that subordinates are willing to pay would increase by less than 6%. With relatively tight ecological constraints as in Figure 4, adaptively lazy dominants do not therefore threaten group formation.
Renting requires tight or moderate ecological constraints in non-kin
groups and moderate ecological constraints in kin groups
Tight ecological constraints (low x) describe a situation in which
dispersing individuals face difficulties in finding a breeding position. If
constraints are tight, subordinates are willing to pay more to be allowed to
stay. However, in the case where dominants are related to their subordinates,
dominants will demand less rent if the subordinate's dispersal chances are
poor. This is because dominants compare the benefits of retaining a
subordinate to the benefits of evicting it, and the latter diminishes if a
related, evicted subordinate fares badly. Therefore, for related individuals,
voluntary helping will exceed rent requirements at tightest ecological
constraints (low x), whereas benefits of dispersal will exceed
benefits of staying if independent breeding is unconstrained (high
x). Renting, if any, will be required at intermediately strong
constraints. Because the minimum acceptable rent is set by the dominants, the
highest rent appears at relatively good dispersal prospects (i.e., constraints
that are mild [high x] but not mild enough to lead to dispersal to
the subordinate; Figure 5; see
also Figure 1).
|
) or rD =
rS = 0.5 (solid lines). Other parameter values: 
0.32 when group members are
unrelated, but only between x = 0.25 and x = 0.32 when
relatedness equals 0.5. At low x (tight constraints), voluntary
helping of related subordinates exceeds the rent requirement. The conflict
over staying, where subordinates would be willing to stay though not
compensating fully for the negative effect of their presence, extends up to
x = 0.47 if subordinates are unrelated to the dominant, but only to
x = 0.32 if subordinates are related to the dominant.In unrelated subordinates, voluntary helping is absent (but see Discussion), and renting is not replaced by voluntary helping at the tightest constraints. Also, in cases where the subordinate benefits from staying and the dominant benefits from its dispersal, subordinates are more tenacious if unrelated, as they do not need to take the dominant's fitness into account. Therefore, the conflict that underlies renting applies at a wider range of values (both low and moderate) of ecological constraint if relatedness between dominant and subordinate is low (Figure 5).
Renting requires that nonhelping subordinates are harmful
If it is beneficial for a dominant to have subordinates even if these do
not help (i.e., if h0 > 0); dominants will not require
any rent-paying. From related subordinates, rent will not be required even if
they are slightly harmful for the dominant
(Figure 6). Unrelated
subordinates are required to fully compensate for the harm
(h0 + hS 0), to be accepted as
group members, whereas partial compensation suffices for related subordinates.
Additionally, because of the voluntary helping effort by related subordinates,
subordinates that are kin will only need to pay rent if they are extremely
harmful to the dominant's reproduction, and even then they do not need to
fully compensate for the harm caused. This is seen in
Figure 6, where
h0 + Hmin = 0 for an unrelated
subordinate (full compensation), but h0 +
Hmin < 0 for a related subordinate (i.e., the dominant
tolerates some reduction in its own fitness). Kin groups are therefore
expected to be less productive overall if help is rent-based. However, when
help by related individuals is voluntary, kin groups are more productive than
non-kin (Figure 6).
|
), and
only fail to do so when their harmful effect is very great (lowest
h0). Renting will apply to non-kin as soon as they will
otherwise harm group productivity, but related subordinates will pay rent only
if retaining them is otherwise extremely harmful for the dominant. Where
related subordinates provide much voluntary help, kin groups are more
productive than non-kin groups; otherwise, the reverse is true because kin pay
less rent.
Summary of results
We have shown that renting can be expected in some situations but can be
overridden by several alternatives such as voluntary helping or eviction.
Especially in kin groups, payment of rent is only stable under rather
restrictive conditions. Figure
7 summarizes the effect of survival Smax
relatedness, r (assuming symmetrical relatedness
rD = rS = r), ecological
constraint, x, and passive subordinate effect,
h0, on the solutions. Payment of rent requires tight or
moderate ecological constraints in non-kin and moderate constraints in kin,
and is most widely established if survival is high. Related subordinates pay
rent only if they would otherwise be very harmful for the dominant
(h0 << 0), whereas rent is required from unrelated
subordinates at the slightest negative effect h0 < 0.
At the borderline between stable, cooperative groups and instability of group
formation, there may be a region in which rent-paying groups can only exist if
dominants are not adaptively lazy (i.e., if they do not reduce their own
parental effort as a response to help by the subordinate). If dominants are
lazy, the subordinates' benefits of staying are reduced to such a degree that
they opt for dispersal instead. Dominant laziness, however, seldom hinders
group formation (Figure 7;
regions marked with UR are small).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONREFERENCES |
|---|
This model confirms arguments for the evolutionary stability of a pay-to-stay system of helping (Gaston, 1978
; Kazem and Wright, in
press). By exploring this possibility within a reproductive skew
framework, we have been able to formulate predictions concerning the existence
of rent paying in cooperative groups. Rent paying applies most widely when the
relatedness between the subordinate and the dominant is low, but related
subordinates may also be required to pay rent under restricted conditions.
This happens especially if helping is costly, in which case the voluntary
effort of related subordinates will be low. There is a possibility that
dominants may destroy the stability of rent paying by selfishly decreasing
their parental effort in favor of their own survival (thereby diminishing
prospects of territorial inheritance), but this only applies to borderline
cases in which mild ecological constraints begin to favor independent breeding
of subordinates. We may thus expect to see rent being paid in groups where
relatedness between group members is low, subordinates potentially reduce the
fitness of dominants unless they compensate by helping, ecological constraints
are moderately tight, and survival from one breeding season to the next is
high.
High survival also means that breeding vacancies occur less often in the
environment. We have modeled the probability of acquiring a breeding vacancy,
x, and the probability of acquiring a new subordinate, a, as
independent parameters, whereas in reality they interact with the population
parameters, including survival (Arnold and
Owens, 1998; Kokko and
Lundberg, 2001; Pen and
Weissing, 2000). Because subordinates are more willing to stay
under tight ecological constraints (low x), our conclusion that high
survival favors staying and renting would have been strengthened even more if
our model had included a link between low x and low mortality of
territory owners.
Our conclusions appear to be broadly consistent with examples of pay to
stay in the cooperative bird literature. These tend to involve unrelated male
helpers in potential reproductive competition with the breeding male, being
tolerated only because they are needed, and benefiting via access to one of a
limited number of dominant breeding positions in subsequent breeding seasons
(e.g., Dunn et al., 1995;
Kazem and Wright, in preparation; Mulder
and Langmore, 1993; Reyer,
1980,
1984). However, rent payment
explanations may only have been invoked in exactly these cases where a lack of
relatedness and reproductive concessions already excluded the more obvious
possibility of fitness benefits from kin-selected helping. It is only through
models such as ours that rent payment can be identified as one of a number of
factors responsible for the level of helping seen in a particular system. We
have outlined the conditions under which renting should be observed. In
empirical tests of the pay-to-stay hypothesis, renting can potentially be
distinguished from other forms of helping by its involuntary nature, where too
little help leads to punishment.
In agreement with earlier results
(Kokko and Johnstone, 1999;
Ragsdale, 1999), our model
predicts that concessions are not necessarily very important in groups with
prominent future fitness benefits: despite the possibility of concessions, our
model often predicts complete skew. Helping in many species is not restricted
to nonbreeding individuals, however (Bourke
and Franks, 1995; Brown,
1987; Emlen,
1991), and the evolution of such cooperative breeding is
associated with longevity (Arnold and
Owens, 1998). That concessions are evolutionarily unstable in
long-lived species (see also Kokko and
Johnstone, 1999) suggests that any observations of incomplete skew
in real cooperative systems might be the result of a lack of dominant control
over reproduction (see also Clutton-Brock
et al., 2001). A natural extension of the present model would
therefore be to solve the optimal allocation of parental and alloparental care
in groups without dominant control of reproduction. This would require one to
focus on the trade-off between helping and an individual's own breeding
effort, instead of, or in addition to, the trade-off between helping and
survival. The former has been addressed, albeit indirectly, in a recent model
of reproductive skew (Cant and Johnstone,
1999), which incorporates a link between the division of
reproduction among group members and total productivity; the latter is the
focus of results presented here. The joint treatment of helping and skew may
be particularly relevant to payment of rent because the amount of reproduction
that a subordinate claims will strongly influence the level of help that it
must give to render tolerance profitable for the dominant
(Johnstone and Cant,
1999).
How general are our conclusions, given that we have shown the evolutionary stability of rent paying in nonconcessive groups only? Our model may provide a conservative view on rent paying: the model shows that rent paying is evolutionarily stable but relevant only under fairly restricted conditions. Imperfect dominant control over reproduction may mean that subordinates have the option of breeding and may do best by claiming some reproduction for themselves, while simultaneously offering help in order to defray the costs their breeding imposes on dominants. Payment of rent could, in this case, prove to be significant in a wider range on natural systems than has previously been acknowledged.
We conclude by discussing two other processes not included in our model
that may influence the prevalence of rent paying: nonevictive enforcement of
helping behavior and delayed benefits of helping. Enforcement of helping
behavior (Clutton-Brock and Parker,
1995; Reeve, 1992)
might be a widely used strategy by dominants. The levels of effort calculated
in the present model represent the minimum effort required to persuade a
dominant to accept otherwise harmful subordinates. If independent breeding is
constrained, this is often less than the maximum that subordinates will
tolerate rather than leaving the group
(Johnstone and Cant, 1999).
Indeed, in our model, the maximum help, Hmax, accepted by
the subordinate is often much higher than the minimum,
Hmin, that enables group stability: related subordinates
should often accept even suicidal help levels (Hmax = 1).
This means that subordinates can be coerced to pay more than the minimum if
dominants use an eviction rule with a large H, even in cases where
renting is not required for group stability. Still, such eviction rules may be
unlikely to spread because it would not be beneficial for a single mutant
dominant to evict subordinates if they are helping sufficiently to have a
positive net effect on the dominant's fitness. However, it seems that a more
detailed treatment of the use of behaviors such as punishment without evicting
(e.g., Mulder and Langmore,
1993) could lead to higher rent levels. Therefore, if dominants
can potentially punish subordinates that want to stay (by means other than
eviction), they could potentially enforce higher help levels from
subordinates. A game-theoretic negotiation approach
(McNamara et al., 1999) would
shed more light on this issue.
Finally, we note that any process that increases voluntary help levels will
reduce the need for rent paying. We have considered kin selection as the only
reason to provide help voluntarily, but delayed benefits of helping, such as
pseudoreciprocity or group augmentation, may provide a reason to help even for
unrelated subordinates (Brown,
1983,
1987; Connor,
1986,
1995;
Kokko et al., 2001;
Ligon, 1981; Woolfenden and
Fitzpatrick, 1978,
1984). Hence, with the
inclusion of direct fitness benefits from increasing group size, the prospects
of renting would decrease.
|
ACKNOWLEDGEMENTS |
|---|
We thank Sigal Balshine, Mike Cant, Andrew Cockburn, Franck Courchamp, Peter Dunn, Jeremy Field, Alasdair Houston, Laurent Keller, Haven Wiley, and an anonymous referee for their helpful comments on this manuscript at its various stages. This study was funded by the TMR programme of the European Commission and by the Royal Society.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONREFERENCES |
|---|
Arnold KE, Owens PF, 1998. Cooperative breeding in birds: a comparative test of the life history hypothesis. Proc R Soc Lond B 265: 739-745.
Balshine-Earn S, Neat FC, Reid H, Taborsky M, 1998.
Paying to stay or paying to breed? Field evidence for direct benefits of
helping behavior in a cooperatively breeding fish. Behav Ecol
9: 432-438.
Bourke AFG, Franks NR, 1995. Social evolution in ants. Princeton, New Jersey: Princeton University Press.
Brown JL, 1983. Cooperation: a biologist's dilemma. Adv Study Behav 13: 1-37.
Brown JL, 1987. Helping and communal breeding in birds: ecology and evolution. Princeton, New Jersey: Princeton University Press.
Cant MA, Johnstone RA, 1999. Costly young and the
partitioning of reproduction in animal societies. Behav Ecol
10: 178-184.
Clutton-Brock TH, 1991. The evolution of parental care. Princeton, New Jersey: Princeton University Press.
Clutton-Brock TH, Brotherton PNM, Russell AF, O'Riain MJ, Gaynor D,
Kansky R, Griffin A, Manser M, Sharpe L, McIlrath GM, Small T, Moss A, Monfort
S, 2001. Cooperation, control, and concession in meerkat groups.
Science 291:
478-481.
Clutton-Brock TH, O'Riain MJ, Brotherton PNM, Gaynor D, Kansky R,
Griffin AS, Manser M, 1999. Selfish sentinels in cooperative
mammals. Science 284:
1640-1644.
Clutton-Brock TH, Parker GA, 1995. Punishment in animal societies. Nature 373: 209-215.[Medline]
Cockburn A, 1998. Evolution of helping behavior in cooperatively breeding birds. Annu Rev Ecol Syst 29: 141-177.[Web of Science]
Connor RC, 1986. Pseudo-reciprocity: investing in mutualism. Anim Behav 34: 1562-1566.
Connor RC, 1995. The benefits of mutualism: a conceptual framework. Biol Rev 70: 427-457.
Crespi BJ, Ragsdale JE, 2000. A skew model for the evolution of sociality via manipulation: why it is better to be feared than loved. Proc R Soc Lond B 267: 821-828.[Medline]
Dunn PO, Cockburn A, Mulder RA, 1995. Fairy-wren
helpers often care for young to which they are unrelated. Proc R Soc
Lond B 259:
339-343.
Ekman J, Bylin A, Tegelström H, 1999. Increased
lifetime reproductive success for Siberian jay Perisoreus infaustus
males with delayed dispersal. Proc R Soc Lond B
266: 911-915.
Emlen ST, 1991. Evolution of cooperative breeding in birds and mammals. In: Behavioural ecology: an evolutionary approach (Krebs JR, Davies NB, eds). Oxford: Blackwell Scientific; 301-337.
Emlen ST, 1997. Predicting family dynamics in social vertebrates. In: Behavioural ecology: an evolutionary approach, 3rd ed. (Krebs JR, Davies NB, eds). Oxford: Blackwell Scientific; 228-253.
Emlen ST, Wrege PH, 1989. A test of alternate hypotheses for helping in white-fronted bee-eaters of Kenya. Behav Ecol Sociobiol 25: 303-319.[Web of Science]
Gaston AJ, 1978. The evolution of group territorial behavior and cooperative breeding. Am Nat 112: 1091-1100.[Web of Science]
Goldstein JM, Woolfenden GE, Hailman JP, 1998. A same-sex step-parent shortens a prebreeder's duration on the natal territory: tests of two hypotheses in Florida scrub-jays. Behav Ecol Sociobiol 44: 15-22.
Hamilton WD, 1964. The genetical theory of social behaviour. J Theor Biol 7: 1-52.[Web of Science][Medline]
Hamilton WD, 1971. Geometry for a selfish herd. J Theor Biol 31: 295-311.[Web of Science][Medline]
Hatchwell BJ, 1999. Investment strategies of breeders in avian cooperative breeding systems. Am Nat 154: 205-219.
Hatchwell BJ, Russell AF, 1996. Provisioning rules in
cooperatively breeding long-tailed tits Aegithalos caudatus: an
experimental study. Proc R Soc Lond B
263: 83-88.
Houston AI, Davies NB, 1985. The evolution of co-operation and life history in the dunnock, Prunella modularis. In: Behavioural ecology: ecological consequences of adaptive behaviour (Sibly RM, Smith RH, eds). Oxford: Blackwell Scientific; 471-487.
Johnstone RA, 2000. Models of reproductive skew: a review and synthesis. Ethology 106: 5-26.
Johnstone RA, Cant MA, 1999. Reproductive skew and the
threat of eviction: a new perspective. Proc R Soc Lond B
266: 275-279.
Kazem AJN, Wright J, in press. Provisioning as a signal: providing food for thought? J Avian Biol.
Kokko H, Johnstone RA, 1999. Social queuing in animal
societies: a dynamic model of reproductive skew. Proc R Soc Lond
B 266:
571-578.
Kokko H, Johnstone RA, Clutton-Brock TH, 2001. The evolution of cooperative breeding through group augmentation. Proc R Soc Lond B 268: 187-196.[Medline]
Kokko H, Lundberg P, 2001. Dispersal, migration and offspring retention in saturated habitats. Am Nat 157: 188-202.[Web of Science][Medline]
Legge S, 2000. The effect of helpers on reproductive success in the laughing kookaburra. J Anim Ecol 69: 714-724.
Ligon JD, 1981. Demographic patterns and communal breeding in the green woodhoopoe, Phoeniculus purpureus. In: Natural selection and social behavior: recent research and new theory (Alexander RD, Tinkle DW, eds). New York: Chiron Press; 231-243.
Maynard Smith J, 1964. Group selection and kin selection. Nature 201: 1145-1147.[Web of Science]
McNamara JM, Gasson CE, Houston AI, 1999. Incorporating rules for responding into evolutionary games. Nature 401: 368-371.[Medline]
Motro U, 1993. Helpers at parents' nest: a game theoretic approach. J Theor Biol 163: 127-134.
Mulder RA, Langmore NE, 1993. Dominant males punish helpers for temporary defection in superb fairy-wrens. Anim Behav 45: 830-833.[Web of Science]
Pen I, Weissing FJ, 2000. Towards a unified theory of cooperative breeding: the role of ecology and life history re-examined. Proc R Soc Lond B 267: 2411-2418.[Medline]
Queller DC, Zacchi F, Cervo R, Turillazzi S, Henshaw MT, Santorelli LA, Strassmann JE, 2000. Unrelated helpers in a social insect. Nature 405: 784-787.[Medline]
Ragsdale JE, 1999. Reproductive skew theory extended: the effect of resource inheritance on social organization. Evol Ecol Res 1: 859-874.
Reeve HK, 1992. Queen activation of lazy workers in colonies of the eusocial naked mole-rat. Nature 358: 147-149.[Medline]
Reeve HK, 1998. Game theory, reproductive skew, and nepotism. In: Game theory and animal behaviour (Dugatkin LA, Reeve HK, eds). Oxford: Oxford University Press; 118-145.
Reeve HK, Keller L, 1995. Partitioning of reproduction in mother-daughter versus sibling associations: a test of optimal skew theory. Am Nat 145: 119-132.[Web of Science]
Reyer HU, 1980. Flexible helper structure as an ecological adaptation in the pied kingfisher (Ceryle rudis). Behav Ecol Sociobiol 6: 219-227.[Web of Science]
Reyer HU, 1984. Investment and relatedness: a cost/benefit analysis of breeding and helping in the pied kingfisher (Ceryle rudis). Anim Behav 32: 1163-1178.
Roff DA, 1992. The evolution of life histories. London: Chapman & Hall.
Stacey PB, Ligon JD, 1991. The benefits-of-philopatry hypothesis for the evolution of cooperative breeding: variation in territory quality and group-size effects. Am Nat 137: 831-846.[Web of Science]
Trivers RL, 1972. Parental investment and sexual selection. Sexual selection and the descent of man, 1871-1971 (Campbell B, ed). Chicago: Aldine Press; 136-179.
Vehrencamp SL, 1979. The roles of individual, kin, and group selection in the evolution of sociality. In: Handbook of behavior and communication, vol. 3 (Marler P, Vandenbergh JG, eds). New York: Plenum Press; 351-394.
Vehrencamp SL, 1983. Optimal degree of skew in reproductive societies. Am Zool 23: 327-335.
Woolfenden GE, Fitzpatrick JW, 1978. The inheritance of territory in group-breeding birds. Bioscience 28: 104-108.
Woolfenden GE, Fitzpatrick JW, 1984. The Florida scrub jay: demography of a cooperative-breeding bird. Princeton, New Jersey: Princeton University Press.
Wright J, 1997. Helping-at-the-nest in Arabian babblers: signalling social status or sensible investment in chicks? Anim Behav 54: 1439-1448.[Web of Science][Medline]
Wright J, Cuthill I, 1989. Manipulation of sex differences in parental care. Behav Ecol Sociobiol 25: 171-181.[Web of Science]
Wright J, Dingemanse N, 1999. Parents and helpers compensate for experimental changes in the provisioning effort of others in the Arabian babbler. Anim Behav 58: 345-350.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. M. Johns, K. J. Howard, N. L. Breisch, A. Rivera, and B. L. Thorne Nonrelatives inherit colony resources in a primitive termite PNAS, October 13, 2009; 106(41): 17452 - 17456. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. McDonald, A. J.N. Kazem, M. F. Clarke, and J. Wright Helping as a signal: does removal of potential audiences alter helper behavior in the bell miner? Behav. Ecol., September 1, 2008; 19(5): 1047 - 1055. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Pacheco, P. G. McDonald, J. Wright, A. J.N. Kazem, and M. F. Clarke Helper contributions to antiparasite behavior in the cooperatively breeding bell miner Behav. Ecol., February 7, 2008; (2008) arm163v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. F. Russell, N. E. Langmore, A. Cockburn, L. B. Astheimer, and R. M. Kilner Reduced Egg Investment Can Conceal Helper Effects in Cooperatively Breeding Birds Science, August 17, 2007; 317(5840): 941 - 944. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Y.L Wong, P. M Buston, P. L Munday, and G. P Jones The threat of punishment enforces peaceful cooperation and stabilizes queues in a coral-reef fish Proc R Soc B, April 22, 2007; 274(1613): 1093 - 1099. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. Reeve and S.-F. Shen A missing model in reproductive skew theory: The bordered tug-of-war PNAS, May 30, 2006; 103(22): 8430 - 8434. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Zink and H. K. Reeve Predicting the temporal dynamics of reproductive skew and group membership in communal breeders Behav. Ecol., September 1, 2005; 16(5): 880 - 888. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A Stiver, P. Dierkes, M. Taborsky, H Lisle Gibbs, and S. Balshine Relatedness and helping in fish: examining the theoretical predictions Proc R Soc B, August 7, 2005; 272(1572): 1593 - 1599. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Brouwer, D. Heg, and M. Taborsky Experimental evidence for helper effects in a cooperatively breeding cichlid Behav. Ecol., May 1, 2005; 16(3): 667 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. S. Griffin and S. A. West Kin Discrimination and the Benefit of Helping in Cooperatively Breeding Vertebrates Science, October 24, 2003; 302(5645): 634 - 636. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. S. Griffin, J. M. Pemberton, P. N. M. Brotherton, G. McIlrath, D. Gaynor, R. Kansky, J. O'Riain, and T. H. Clutton-Brock A genetic analysis of breeding success in the cooperative meerkat (Suricata suricatta) Behav. Ecol., July 1, 2003; 14(4): 472 - 480. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||