Behav Ecol Sociobiol

DOI 10.1007/s00265-015-1921-1

ORIGINAL PAPER

Exploration behavior and parental effort in wild great tits:

partners matter
Morgan David 1 & Rianne Pinxten 1,2 & Tine Martens 1 & Marcel Eens 1

Received: 20 February 2015 / Revised: 3 April 2015 / Accepted: 7 April 2015

# Springer-Verlag Berlin Heidelberg 2015

Abstract The extended pace-of-life syndrome (POLS) hypothesis suggests that variation in boldness-like behaviors
has co-evolved with variation in life-history strategies within
populations, yet both theoretically driven experiments and
evidence for phenotypic correlations between boldness-like
behaviors and reproduction-related activities are scarce. Here
we test the prediction that more exploratory individuals should
be willing to provide more effort into current reproduction
than less exploratory ones by investigating the association
between exploration behavior and parental effort in wild great
tits (Parus major). To this end, we assessed exploration behavior following a standardized assay. Then, we estimated
individual willingness to provide parental effort into brood
provisioning as (1) individual increase in nest visit rate after
the brood had been artificially enlarged and (2) individual
latency to return to the nest after this manipulation. Fast male
explorers were quicker than slow explorers to return to the
nest after the manipulation. Males paired with a partner of
similar exploration scoreeither a fast or slow female explorer increased their nest visit rate more than males paired with
Communicated by S. Pruett-Jones
* Morgan David
[email protected]
Rianne Pinxten
[email protected]
Tine Martens
[email protected]
Marcel Eens
[email protected]
1

Department of BiologyEthology, University of Antwerp, Drie

Eiken Campus, Universiteitsplein 1, 2610, Wilrijk Antwerp, Belgium

Department of Instructional and Educational Sciences, Research Unit

Didactica, University of Antwerp, Antwerp, Belgium

a partner of dissimilar exploration score. The relationship between exploration and parental effort then depended on ones
partners behavior. Our test thus provides only partial support
for the extended POLS hypothesis and highlights the potential
importance of the social environment in shaping the relationship between boldness-like behaviors and fitness-maximizing
traits.
Keywords Brood size manipulation . Pace-of-life . Parus
major . POLS . Parental care . PIT tags . Provisioning .
Partners compatibility

Introduction
The extended pace-of-life syndrome (POLS) hypothesis proposes that between-individual variation in boldness-like behaviors such as exploration, aggressiveness, or risk-taking
has coevolved with between-individual variation in lifehistory strategies (Biro and Stamps 2008; Rale et al. 2010).
Life-history/behavior correlations are thus thought of as being
linked with variation in individual productivity and lifehistory trade-offs (Biro and Stamps 2008). For instance, the
extended POLS hypothesis predicts that more exploratory,
active, or aggressive individuals should show lower survival
prospects (Nicolaus et al. 2012; Auclair et al. 2013) but a
higher metabolism (Careau et al. 2011), higher energy intake
(Carter et al. 2010; David et al. 2011a), higher growth rate
(Biro et al. 2014), and higher short-term reproductive success
(Patterson and Schulte-Hostedde 2011).
So far, the extended POLS hypothesis has received mixed
support with studies showing unexpected opposite patterns
(Adriaenssens and Johnsson 2009, 2011; Smith and
Blumstein 2010; David et al. 2011b; Le Galliard et al.
2013; Bridger et al. 2015), calling its generalizability into

Behav Ecol Sociobiol

question. In addition, evidence for significant covariation

between boldness-like behaviors and reproduction-related
activities are scarce (Duckworth 2006; Barnett et al. 2012;
Mutzel et al. 2013), while theoretically driven tests are virtually lacking (but see Patrick and Browning (2011)). For
instance, the question on whether variation in boldness-like
behaviors has coevolved with variation in parental care is
left open. In addition, less is known about how the social
environment can shape the relationship between boldnesslike behaviors and fitness-maximizing traits (Webster and
Ward 2011). This is especially important for biparental
care species where both partners of a breeding pair show
common fitness prospects but also conflicts about parental duties (Lessells and McNamara 2012; Johnstone et al.
2014). Behavioral compatibility within breeding pairs has
been shown to be crucial for reproductive fitness in such
species (Spoon et al. 2006; Schuett et al. 2011; Mariette
and Griffith 2012, 2015; Harris and Siefferman 2014),
yet it remains unknown whether the importance of behavioral compatibility within a pair may confound and/or
override the association between boldness-like behaviors
and parental effort expected under the extended POLS
hypothesis.
In this study, we use wild great tits (Parus major) to
test the phenotypic relationship between exploration behavior and willingness to provide provisioning effort,
while taking partners exploration behavior into account.
Exploration behavior has been demonstrated to be heritable and repeatable in this species (e.g., Dingemanse et al.
2002), related to life-history traits (e.g., Dingemanse et al.
2003, 2004) and other behaviors such as aggressiveness
(Verbeek et al. 1996; Mutzel et al. 2013). Exploration
behavior thus reflects an ecologically relevant behavioral
dimension in great tits. In the present study, willingness to
provide parental provisioning effort was assessed as the
increase in nest visit rate following a temporary artificial
enlargement of brood size (Patrick and Browning 2011).
To this end, we monitored parental nest visits using a
system of PIT-tag antennas placed at the entrance of artificial nestboxes. This procedure allows to investigate individuals willingness to provide more or less effort into
parental provisioning when stimulated to do so, and not
genuinely the investment that a bird is a priori expected to
provide given its behavioral profile (see Fawcett et al.
(2013) for a more thorough discussion on the difference
between evolutionary and behavioral decisions). Following the predictions from a recent mathematical model
(Wolf et al. 2007), we expect fast explorers to invest more
effort into current reproduction and thus be more willing
to increase parental effort than slow explorers. This is
because fast explorers are thought of as having lower
survival prospects and thus prioritizing current over future
reproduction (Wolf et al. 2007; Nicolaus et al. 2012). Fast

explorers should thus show a higher increase in nest visit

rate after the experimental brood size enlargement compared to slow explorers.

Methods
Study subjects
The data were collected from a suburban great tit population
located on the Drie Eiken Campus of the University of Antwerp, Belgium (51944 N, 42415 E). Circa 140 nest boxes are provided for great tits to reproduce (Rivera-Gutierrez
et al. 2010). Great tits can be identified using metal leg rings
that they receive as nestlings (day 9/10 post-hatching) or using
a unique combination of colored leg rings for adults (RiveraGutierrez et al. 2012). For all adults, one colored ring bears a
passive integrated transponder (PIT) tag (IB Technology,
Aylesbury, UK), allowing for further bird identification using
antennas (Dorset Identification B.V., The Netherlands). Before the reproductive seasons onset, nest boxes were regularly
checked for nest building indications and then checked daily
before the anticipated start of egg-laying until the last egg
hatches.
Individual parents were caught at their nest box for a short
period of time when feeding the chicks at day 9 post-hatching.
Nestlings and unringed or untagged parents were then banded
and PIT-tagged. The number of chicks present inside each nest
(thereafter termed Bbrood size^) was determined on day 9. On
average, brood size in focal nests was 7.00.3 (SE).
Assessment of exploration behavior
Exploration behavior was assessed from the 15th to the 27th
of February 2013, during the winter preceding the reproductive season when the parental effort experiment was conducted. Great tits were taken out of the nest box within which they
were sleeping at night. Birds were immediately brought to an
experimental room following capture and placed alone in a
cage (lwh: 0.830.40.5 m) comprising a small nest box
and ad libitum access to mealworms, sunflower seeds, and
water. The room temperature was 52 C and kept under a
natural light/dark cycle. On the morning following capture,
birds were tested for their exploration behavior in a novel
environment room (lwh: 4.02.42.3 m) comprising five
artificial trees (height diameter: 1.5 0.04 m) with four
branches each (two at 5 cm and two at 25 cm below the
top). This is a standard procedure for assessing exploration
behavior in great tits (Dingemanse et al. 2002; van Overveld
and Matthysen 2010; Patrick and Browning 2011; Nicolaus
et al. 2012). A sliding door providing a direct access from the
cage to the novel environment room was opened by the experimenter while the lights were still off inside the latter. Then,

Behav Ecol Sociobiol

lights were turned off in the room where the cages were held
and turned on in the novel environment room, which stimulates birds to enter it. Individual behavior was then recorded
for 2 min during which the number of movements between
trees and between branches of the same tree was measured
(thereafter referred to as Bexploration score^). Birds with high
exploration scores are thereafter called Bfast explorers^,
whereas those with low scores are called Bslow explorers^.
Lights from the novel environment room were turned off
again, which makes the bird Bfreeze^ and easy to catch by
the experimenter. Birds were then immediately taken back to
and released at the place they had been caught the day before.

Assessment of parental effort

The amount of effort into chick provisioning was assessed
using parental nest visit rate (Wilkin et al. 2009; Auclair
et al. 2014) following an artificial brood size enlargement
(Patrick and Browning 2011). In the present study, we do
not present results about the observed relationship between
exploration and nest visit rate during the pre-enlargement period. A more thorough dataset using the same individuals
recorded over a longer time period is the subject of a future
publication elsewhere. Antennas were placed at each selected
nest box in the afternoon of day 10 post-hatching. Focal nest
boxes were selected on the basis that, when possible, both
parents were PIT-tagged and other nest boxes from which to
take chicks of the same age for the artificial enlargement were
available. Overall, we used 14 nest boxes where both parents
were tested for exploration behavior, four nest boxes for
which only the male was exploration-tested, two nest boxes
for which only the female was exploration-tested, and one nest
box for which neither parent was exploration-tested. This latter nest box was obviously only used in analyses that did not
involve exploration scores (i.e., between-period change in nest
visit rate). Overall, 34 birds were tested for exploration behavior, whereas eight were not. However, provisioning behavior
was recorded for all of them. Each setup consisted of a box
(43339 cm) placed on the ground close to the tree to which
the nest box was attached. It contained all the electronic devices, including data logger, the battery, and the USB stick on
which data about nest visits were stored. This box was linked
with a thin wire to a circular antenna with cycles of ten detections per second, placed inside the nest box, and fitted with the
entrance hole. The actual antenna replaced a dummy antenna
that had been set before the start of the breeding season in
order for birds to familiarize with it. The installation of the
whole antenna system by the experimenter (MD) was not to
last more than a couple of minutes, thus minimizing disturbance at the nest. The artificial brood size enlargement occurred on the morning of day 13 post-hatching, when chicks
feeding frequency peaks in great tits (Barba et al. 2009).

On day 13 post-hatching, some nest boxes that were not

used in the experiment yet and for which chicks were of the
same age as those of the monitored focal nest boxes were
selected. The experimenter (MD) collected two chicks from
these nest boxes. They were then carried and added to a focal
nest (enlargements starting timeSE, 9:23 am 4 min). They
remained in the focal nest box for 2:30 h, after which they
were placed back in their original nest. The antenna setup was
removed from the focal nest box shortly afterwards.
The brood size manipulation caused parents to temporarily
fly away from the nest box. We considered the time delay to
return to the nest box after the enlargement as an indication of
a birds willingness to provide parental effort despite a potential hazard. Individuals returning faster to the nest box after the
manipulation are thus thought of as being more willing to
provide parental effort than individuals returning later.

Data processing
We computed a similarity index indicating to what extent both
partners of a given breeding pair are similar in terms of exploration behavior following the formula:
Similarity index
jfemale exploration scoremale exploration scorej
In our sample, similarity indices range from 3 (pairs of
individuals with highly similar exploration scores) to 26 (pairs
of individuals with highly dissimilar exploration scores).
Nest visit rate was computed as the number of minutes per
hour that an individual was detected at the nest divided by the
number of minutes within an hour (i.e., 60). This method has
previously been used in studies of great tits provisioning behavior using the same antennas apparatus (Wilkin et al. 2009;
Patrick and Browning 2011) and is considered to reliably reflect chicks feeding rate (Wilkin et al. 2009). To compute the
difference in nest visit rate before versus after brood size enlargement (respectively pre- and post-enlargement period),
thus controlling for between-pair initial differences in nest
visit rate, we quantified nest visit rate during 2:15 h before
and after the manipulation. The measurement of nest visit rate
after brood size enlargement started 15 min after the manipulation to allow parents to recover from the associated disturbance (see Limbourg et al. (2013)). We chose this duration as
birds were re-detected by the antenna after a median delay of
7.5 min (interquartile range, [3,14]). Eighty-eight percent of
the birds (37 out of 42) were re-detected by the antenna within
the 15-min period following the manipulation. Analyses conducted only with these individuals that were re-detected by the
antenna within the 15-min period substantially yield the same
results (not shown). For each parent, we recorded the time
delay (in number of minutes) it took to come back to its nest

Behav Ecol Sociobiol

after the manipulation as a measure of willingness to provide

parental effort. The relative change in nest visit rate following

brood size enlargement was computed for each individual as

follows:

Relative change in nest visit rate

nest visit rate during the post enlargement periodnest visit rate during the pre enlargement period

Nest visit rate during the pre enlargement period

The computation of this index indeed allows for a control

of potential initial differences in nest visit rate between pairs.
Also, it enables us to reduce the number of predictors in our
models, thus diminishing the risk of overloading our models
with many parameters given our small sample size. The relative change in nest visit rate was log-transformed for every
analysis so as to reach normality (Shapiro test W=0.96, P=
0.16) and be bounded by minus and plus infinity rather than
1 and plus infinity otherwise.
Statistical analyses
Statistical models were built to, first, investigate any potential
link between time delay to return to the nest after the brood size
enlargement and exploration behavior; second, determine
whether brood size enlargement was efficient at stimulating
higher nest visit rates; third, examine whether fast explorers
increased their nest visit rate more following brood size enlargement than slow explorers; and fourth, test whether the
increase in nest visit rate was linked to the similarity in exploration behavior between members of a pair. In the present paper, we did not intend to test the relationship between brood
size and exploration, which will be the subject of another publication elsewhere with a more thorough dataset. However, we
added the variable Bbrood size^ as a co-variate in most of our
models. Models were, otherwise stated, computed using the
Bglmer^ command from the Blme4^ R package (Bates et al.
2014). We proceeded to stepwise backward deletion of the least
significant term until only significant variables remained in the
model with a significance threshold set at P=0.05. We provided
Cliffs (Cliff 1996; Torchiano 2014), Pearsons (r), and
Spearmans (rs) correlation coefficients as effect sizes with the
associated sample size (Nakagawa and Cuthill 2007).
In a first generalized linear mixed model (GLMM), we set
time delay (i.e., number of minutes between the manipulation
and bird detection at the nest) as a response variable. Predictors were brood size and the triple interaction between sex,
exploration score, and partners exploration score. BPair^ was
added as a random intercept variable to account for the nonindependence of females and males behavior within pairs.
We set the model error structure to negative binomial as our
response variable showed signs of overdispersion. We therefore used the BglmmADMB^ R package to analyze this model
(Fournier et al. 2012; Skaug et al. 2014).

In a second GLMM, we set nest visit rate as a response variable. To this end, we used the cbind command including both the number of minutes per hour that
a given bird had been detected by the antenna and the
number of minutes per hour that the bird had not been
detected by the antenna (see Crawley 2007). Brood size
and the interaction between period (pre- and post-enlargement) and sex were added as predictors. In order
to control for the non-independence of males and females behavior within the same pair and for the repeated measures of the same individual across the two periods, individual ID, nested within pair, was set as a
random intercept factor. We also tested this model with
Bperiod^ added as a random slope factor, as recommended by Schielzeth and Forstmeier (2009). The error structure was set to binomial.
Using correlation analyses, we investigated the potential
relationship between nest visit rate before brood size enlargement on day 13 and the change in nest visit rate following
brood size enlargement following Tu and Gilthorpes (2007)
standardized procedure (David et al. 2012). This is to test
whether the increase in provisioning effort following brood
size enlargement depends on the amount of provisioning provided beforehand. We may indeed expect individuals provisioning at a high rate to be less capable of increasing their
effort following brood size enlargement.
In a third GLMM, we set the log-transformed relative
change in nest visit rate (see above) as the response variable.
Brood size and the interaction between exploration score,
partners exploration score, and sex were added as predictors.
To account for the non-independence of males and females
behavior within the same pair, we added Bpair^ as a random
intercept variable. The error structure was set to Gaussian. We
could not add ones partner relative change in nest visit rate as
a predictor in this model as, due to its very random effect
structure, it would have prevented it from converging properly. We have thus tested the correlation between relative change
in nest visit rate and ones partners aside.
In a fourth GLMM, we tested the relationship between the
relative change in nest visit rate and the similarity index (see
above). The log-transformed relative change in nest visit rate
was defined as the response variable, and similarity index was
set as a predictor. We added Bpair^ as a random intercept
variable and set the model error structure to Gaussian.

Behav Ecol Sociobiol

Table 1 Results of the models
testing the relationship between
the time delay taken by birds to
return to the nest after brood size
enlargement and sex, brood size,
exploration score, and partners
exploration score (significant
terms are highlighted in italic)

Time delay to return to the nest

z

Brood size

1.98

0.047

Partners exploration score

Sexexploration score
Sexpartners exploration score
Exploration scorepartners exploration score
Sexexploration scorepartners exploration score

1.46
2.39
1.13
0.77
1.96

0.15
0.050
0.26
0.44
0.73

Results

Brood size enlargement triggers higher nest visit rate

Fast explorers returned quicker to the nest

after the manipulation

Nest visit rate significantly increased between the preenlargement and the post-enlargement period (pre-enlargement period: rateSE=0.350.02; post-enlargement
period: rate SE = 0.39 0.03; b SE = 0.18 0.04, Cliffs
= 0.23, 95%CI = (0.14,0.36); 2 = 19.85, df = 1,
P<0.0001; Fig. 2), irrespective of sex (interaction sex
period: 2 =0.77, df=1, P=0.38). However, the effect of
the period no longer remained significant when Bperiod^
was added as a random slope factor in the model (b
SE=0.150.09; 2 =2.54, df=1, P=0.11). Brood size had
a significant positive effect on nest visit rate (b SE =
0.30 0.08; 2 = 11.37, df = 1, P < 0.001; r s (84) = 0.48,
95%CI=(0.28,0.64)). The change in nest visit rate following brood size enlargement was not related to nest
visit rate before the enlargement (r(42) = 0.18,
95%CI=(0.46,0.14), P=0.27; see BMethods^) nor was
it significantly predicted by time delay to come back to
the n est after the m anipulation ( r s ( 4 2 ) = 0 .05 ,
95%CI=(0.36,0.27), P=0.78).

Time delay to come back to the nest after brood size

enlargement was affected by the interaction between
sex and exploration score (z=1.96, P=0.050; Table 1)
(females: Spearmans rs(14)=0.05, 95 % confidence interval (CI) = (0.73, 0.60); m ales: r s (1 4) = 0. 67,
95%CI=(0.95,0.18); Fig. 1; Table 1), and brood size
(regressions slope estimate b SE = 0.22 0.11, P =
0.047, rs(28) = 0.46, 95%CI = (0.73,0.12); Table 1),
but not by partners exploration score (b SE = 0.03
0.02, P = 0.15, rs(28)= 0.20, 95%CI= (0.55,0.18); Table 1). Fast male explorers were quicker to return to
the nest after the manipulation. No other interaction
terms were found to be significant (Table 1). Both partners time delay to return to the nestbox were positively
correlated (rs(21)=0.47, 95%CI=(0.07,0.78), P=0.025).

Partners combined exploration scores influence parental

effort

Fig. 1 Influence of exploration score on time delay to return to the nest

(in number of minutes) after brood size enlargement

The relative change in nest visit rate was significantly

affected by the interaction between sex, exploration
score, and ones partner exploration score ( 2 = 5.36,
df=1, P =0.021) but not by brood size (bSE = 0.06;
2 = 0.86, df = 1, P = 0.35). Taking each sex separately,
the interaction between exploration score and ones partner exploration score was significant in males (t10 =3.48,
P=0.006; Fig. 3a) but not in females (t10 =1.21, P=0.25;
Fig. 3b). Males paired with a female of similar exploration score increased their nest visit rate more than males
paired with a female of dissimilar exploration score
(Fig. 3a). In females, the relative change in nest visit rate
was not related to exploration score (bSE=0.010.01,
t11 = 0.91, P = 0.38) nor partners exploration score (b
SE = 0.01 0.01, t 1 2 = 0.83, P = 0.42). However, the

Behav Ecol Sociobiol

Fig. 2 Variation in nest visit rate
between pre- and post-brood size
enlargement periods at day 13.
Each line represents a single
individual. Females are
represented in red (light lines) and
males in blue (dark lines). The
dashed line is the sample average.
Note that the difference between
the two periods no longer remains
significant when Bperiod^ is set as
a random slope factor in the
model

similarity index was negatively linked to the relative

change in nest visit rate in females (b SE = 0.02
0.01, t12 = 2.47, P = 0.030, rs(14) = 0.59,
95%CI=(0.80,0.27); Fig. 4), as well as in males (b
SE = 0.04 0.01, t 12 = 2.89, P = 0.014, r s (14) = 0.60,
95%CI=(0.88,0.10); Fig. 4), indicating that effort into

provisioning indeed was higher in pairs of birds with

similar exploration behavior.
Individual relative change in nest visit rate was found
to be positively and significantly related to ones partner s (r(21) = 0.46, 95%CI = (0.04,0.74), P = 0.033;
Fig. 5).

Fig. 3 Relationship between individual exploration score, partners exploration score, and relative change in nest visit rate in males (a) and females (b).
Regression planes represent the models predictions. The relative change in nest visit rate was log-transformed (see BMethods^)

Behav Ecol Sociobiol

Fig. 4 Negative relationship
between the relative change in
nest visit rate and the similarity
index. Individuals increased their
nest visit rate after brood size
enlargement more when paired
with a more similar partner in
terms of exploration behavior.
The relative change in nest visit
rate was log-transformed (see
BMethods^)

Discussion
The extended POLS hypothesis provides clear predictions about the expected relationship between withinpopulation variation in boldness-like behaviors and lifehistory traits. In the present paper, we conducted a test of
the relationship between exploration behavior and willingness to provide parental effort in a wild great tit population. Below we discuss the scope and limitations of
Fig. 5 Positive relationship
between an individuals relative
change in nest visit rate and its
partners. Each point corresponds
to both partners of a given
breeding pair. The relative change
in nest visit rate was logtransformed (see BMethods^)

our findings and their relevance and implication in relation with theory.
The artificial enlargement of brood size was designed to
stimulate individuals to put more effort into chicks provisioning than they would have normally done. Also, the computation of the relative increase in nest visit rate controlled for
potential initial between-pair differences in provisioning rate,
providing a Bcontrol^ measure to the nest visit rate as quantified after the enlargement. Despite this control, it remains

Behav Ecol Sociobiol

possible that the relative increase in nest visit rate may have
been influenced by any natural dynamic daily pattern of parental provisioning, something that we were not able to control in this experiment. However, if provisioning rate has been
shown to peak at dawn and dusk in great tits (Patrick and
Browning 2011), the daily pattern of provisioning was shown
not to differ between individuals with varying exploration
behaviors (Patrick and Browning 2011). We thus think that
it is unlikely that the observed relationship between the relative increase in nest visit rate and exploration was the mere
outcome of variation in daily patterns of provisioning. We
believe that the brood size manipulation was appropriate to
identify which individuals and to what extent they were willing to put more effort into parental duties and thus to place
more weight onto current reproduction than others.
Nest visit rate was computed from the proportion of minutes per hour that a bird was detected by the antenna placed
at the nest entrance (Patrick and Browning 2011). Although
this method does not give an exact measure of how much time
parents spend at the nest or any clue about the quantity and
quality of food brought to the chicks, it has already been used
in previous studies (Patrick and Browning 2011) and shown to
reliably reflect chicks provisioning rate (Wilkin et al. 2009).
We are thus confident that our estimate of nest visit rate is a
reliable proxy for parental provisioning rate and possibly for
the amount of food brought to the offspring. The possibility
yet remains that parents visiting more also bring lower-quality
food than others. Unfortunately, this issue cannot be investigated using the present experimental procedures and would
deserve further research.
The average increase in nest visit rate between the two
periods was found to be rather low (4 %) and even nonsignificant when Bperiod^ was added as a random slope factor
in the model, providing at best only little evidence that birds
increased their provisioning rate after the brood size enlargement. Also, several individuals did not increase but decreased
or kept their nest visit rate stable across the two periods (see
Fig. 2). The possibility exists that some individuals may simply have not responded to the manipulation for several reasons: on the one hand, they may not have had sufficient time
to detect and respond to the manipulation (2 h and 15 min). On
the other hand, some individuals may not have responded to
the enlargement because they had already reached their maximum level of effort into provisioning before the manipulation. However, we did not find any significant correlation
between nest visit rate before the manipulation and the increase resulting from the manipulation, which makes the latter
possibility unlikely. It may also be that some birds did not
respond to the manipulation because their partner responded
sufficiently (the reproductive compensation hypothesis;
Gowaty et al. 2007), but we think that this also is unlikely
given the positive relationship that we found between an individuals relative change in nest visit rate and its partners

(Hunt and Simmons 2002), which on the contrary is more

indicative of a social facilitation phenomenon. Further studies
would be needed to understand why some birds are responsive
and some are not.
The importance of the social environment when testing
the POLS hypothesis
In accordance with the extended POLS hypothesis, we found
that fast male explorers were quicker to come back to the nest
after brood size enlargement than slow explorers, a finding
reminiscent of a recent comparable result during the incubation stage in the same species (Cole and Quinn 2014). In
addition, the relative change in nest visit rate was influenced
by the interaction between an individuals exploration score
and its partners, at least in males. Individuals within assorted
pairs for exploration scores (fastfast explorers and slowslow
explorers) showed a higher increase in nest visit rate than
individuals within non-assorted pairs, irrespective of their
own exploration behavior. In females, we detected no effect
of exploration behavior on the relative increase in nest visit
rate. Yet, based on the correlation between the similarity index
and the relative increase in nest visit rate, females paired with
a male of similar exploration behavior were found to be more
likely to increase their nest visit rate than in dissimilar pairs.
We believe that this discrepancy may come from our small
sample size and a lack of statistical power in the first mentioned linear model. We indeed acknowledge that the sample
size of our study is pretty low and further studies would be
needed to test the generalizability of these results. That said,
our findings provide only partial support for the extended
POLS hypothesis which predicts that fast explorers should,
all else being equal, provide more effort into current reproduction than slow explorers. Indeed, in the present study fast
explorers provisioned their brood more when paired with a
fast partner than when paired with a slow one. Conversely,
slow explorers provisioned their offspring less when paired
with a fast explorer than when paired with a slow explorer.
The importance of combined behavioral types within breeding
pairs is reminiscent of previous findings that assortment for
behavioral traits affects reproductive fitness in pairs of great
tits (Dingemanse et al. 2004), zebra finches (Taeniopygia
guttata) (Schuett et al. 2011), Stellers jays (Cyanocitta
stelleri) (Gabriel and Black 2012), eastern bluebirds (Sialia
sialis) (Harris and Siefferman 2014), and guppies (Poecilia
reticulata) (Ariyomo and Watt 2013). The functional consequences and the reasons why it may be optimal for individuals
to breed with partners of similar behavioral type remain unclear so far (Kralj-Fier et al. 2013). Specific combinations of
behavioral types within breeding pairs may affect brood provisioning efficiency (Mutzel et al. 2013), for instance, through
provisioning synchronization (Mariette and Griffith 2012; see
van Rooij and Griffith (2013)). The extent to which (1) pairs

Behav Ecol Sociobiol

of partners with similar exploration behavior are better synchronized and (2) better synchronization leads to higher reproductive success remains to be determined. The positive correlation between both partners time delay to return to the nest
and between both partners relative change in nest visit rate,
while confirming previous studies (Hinde 2006; Westneat
et al. 2011), also suggests that such synchronization phenomenon could be at work in our study. Indeed it may be that
social facilitation leads partners to change their nest visit rate
to the same extent and that any sort of social facilitation effect
is more salient when partners are of similar exploration behavior. However, this interpretation remains speculative and more
work is needed to disentangle the complex interplay among
partners behavioral types, provisioning behavior, and reproductive success (Mutzel et al. 2013).
The interaction between individual exploration score and
partners exploration score in determining brood provisioning
effort suggests the importance of the social environment in
shaping the relationship between boldness-like behaviors
and reproduction-related activities, and fitness-maximizing
traits in general (Bergmller and Taborsky 2010; Webster
and Ward 2011). It is especially important as an individuals
reproductive success greatly depends on its partners investment into breeding in biparental care species. Testing for a
positive relationship between boldness-like behaviors and reproductive effort may thus become inconclusive because of
the social environments influence (in the present case, ones
partners behavior). In particular, the importance of behavioral
compatibility for reproductive fitness within breeding pairs
(Spoon et al. 2006) may override the positive association between exploration and provisioning effort. Boldness-like behaviors are known to be substantially affected by the social
environment (Mainwaring et al. 2011; Webster and Ward
2011). We believe that it would be elusive to ignore its influence (be it a constraint or a facilitator) when testing predictions of the extended POLS hypothesis in social species,

Fig. 6 Integration of various reproduction-related behaviors into a single

test of the extended POLS hypothesis. Insofar as different components of
reproduction can be related with one another, the overall individual effort
into reproduction can be assessed by taking the residuals of the regression
from the effort provided into one component on the other (principal components analyses may be considered when more than two components are

especially in a reproductive context where both partners fitness prospects converge. Taking the social environment into
account should involve studying pairing patterns with respect
to boldness-like behaviors and determining the functional relationships between boldness-like traits and fitness (i.e.,
questioning what the factors mediating the link between both
are) (Patrick and Browning 2011; Mutzel et al. 2013).
POLS hypothesis and the multidimensionality
of reproductive investment
Overall, our results provide, at best, partial support for the
extended POLS hypothesis with possible sex effects. Our
findings differ from a previous correlational study in wild blue
tits (Cyanistes caeruleus) that showed a positive relationship
between exploration behavior and brood provisioning rate in
females only (Mutzel et al. 2013). Our results also differ from
another study using a similar experimental design that did not
identify any significant link between exploration and brood
provisioning in great tits (Patrick and Browning 2011). The
possibility remains that various wild great tit populations exhibit different patterns of behavioral correlations depending
on the specific selective pressures or the constraints they face
or on their particular life-histories (Adriaenssens and Johnsson
2009; Patrick and Browning 2011). Finally, in western bluebirds (Sialia mexicana), male provisioning rate has been
found to be negatively related to aggressiveness (Duckworth
2006). In this latter study, the direction of the relationship
between parental effort and boldness-like traits goes against
the extended POLS hypothesis predictions. However, it is
noteworthy that aggressiveness may be related to several functional behaviors that a given male may have to trade off
against one another. In the case of western bluebirds, for example, more aggressive males spend more time defending
their nest against potential predators and competitors
(Duckworth 2006). This can reasonably be considered as

involved). According to the extended POLS hypothesis, proactive individuals should overall invest more into reproduction-related activities
than reactive individuals. In this hypothetical example of a trade-off between parenting and mating effort, proactive individuals are then on average expected to have higher residual values than reactives

Behav Ecol Sociobiol

investment into a reproduction-related activity as a males

reproductive fitness likely depends on keeping the nest safe
away from potential hazards. This is because investment into
current reproduction may concern various dimensions of reproductive behavior, such as nest defense (Hollander et al.
2008), extra-pair sexual behavior (Patrick et al. 2012), or parental care (Barnett et al. 2012), that the link between
boldness-like behavioral variation and brood provisioning effort may differ among populations or species. This possibility
argues in favor of the necessity to confirm, generalize, and
extend the present findings to other species and/or other populations of the same species. This is important in order to
refine the extended POLS hypothesis, and understand unexpected associations between variables (David et al. 2011b;
Adriaenssens and Johnsson 2011). Also, we encourage the
simultaneous study of multiple behavioral dimensions within
the same functional context (e.g., in a reproductive context:
signaling, parenting, and so on) insofar as these various
dimensions may not be all maximized at the same time and
may be traded off against one another (Fig. 6). These tradeoffs may contribute to confound the relationships between
boldness-like behaviors and reproduction-related activities expected under the extended POLS hypothesis framework.

Conclusion
To conclude, our findings indicate that both an individuals
behavioral type and its partners can be critical in its decision
to provide more or less effort into brood provisioning. This
interaction is supposed to have a great impact on the relationship between boldness-like behaviors and reproductionrelated activities and thus on the testing of the extended POLS
hypothesis predictions. Future studies should then carefully
consider the social environment (Bergmller and Taborsky
2010) when testing predictions of the extended POLS hypothesis. Further investigations are also needed to integrate various
reproduction-related activities together into a single test of the
extended POLS hypothesis.
Acknowledgments We thank Yannick Auclair, Vincent Careau, FX
Dechaume-Moncharmont, Niels Dingemanse, Claire Doutrelant, Geert
Eens, Wolfgang Forstmeier, Luc-Alain Giraldeau, Carsten Lucass, Karine
Monceau, Wendt Mller, Kees van Oers, Samantha Patrick, Denis Rale,
and Peter Scheys for their help or their comments at any stage of this
study. Financial support was provided by the University of Antwerp
(TOPBOF) and FWO-Flanders (Research Projects G.0130.07 and
G.0280.10) to ME and RP.
Ethical standards This study was approved by the ethical committee
of the University of Antwerp (ID number 2011-31), and it was performed
in accordance with Belgian and Flemish laws. The Belgian Royal Institute for Natural Sciences (Koninklijk Belgisch Instituut voor
Natuurwetenschappen) provided ringing licences for authors and technical personnel.

References
Adriaenssens B, Johnsson JI (2009) Personality and life-history productivity: consistent or variable association. Trends Ecol Evol 24:179
180
Adriaenssens B, Johnsson JI (2011) Shy trout grow faster: exploring the
links between personality and fitness-related traits in the wild.
Behav Ecol 22:135143
Ariyomo TO, Watt PJ (2013) Disassortative mating for boldness decreases reproductive success in the guppy. Behav Ecol 24:1320
1326
Auclair Y, Knig B, Lindholm AK (2013) A selfish genetic element
influencing longevity correlates with reactive behavioural traits in
female house mice (Mus domesticus). PLoS ONE 8:e67130
Auclair Y, Knig B, Ferrari M, Perony N, Lindholm AK (2014) Nest
attendance of lactating females in a wild house mouse population:
benefits associated with communal nesting. Anim Behav 92:143
149
Barba E, Atinzar F, Marn M, Monrs JS, Gil-Delgado JA (2009)
Patterns of nestling provisioning by a single-prey loader bird, great
tit Parus major. Bird Study 56:187197
Barnett CA, Thompson CF, Sakaluk SK (2012) Aggressiveness, boldness
and parental food provisioning in male house wrens (Troglodytes
aedon). Ethology 118:984993
Bates D, Maechler M, Bolker B, Walker S (2014) lme4: linear mixedeffects models using Eigen and S4. R package version 1.1-7
Bergmller R, Taborsky M (2010) Animal personality due to social niche
specialisation. Trends Ecol Evol 25:504511
Biro PA, Stamps JA (2008) Are animal personality traits linked to lifehistory productivity? Trends Ecol Evol 23:361368
Biro PA, Adriaenssens B, Sampson P (2014) Individual and sexspecific differences in intrinsic growth rate covary with consistent individual differences in behaviour. J Anim Ecol 83:
11861195
Bridger D, Bonner SJ, Briffa M (2015) Individual quality and personality:
bolder males are less fecund in the hermit crab Pagurus bernhardus.
Proc R Soc Lond 282:20142492
Careau V, Thomas D, Pelletier F, Turki L, Landry F, Garant D, Rale D
(2011) Genetic correlation between resting metabolic rate and exploratory behaviour in deer mice (Peromyscus maniculatus). JE Vol
biol 24:21532163
Carter AJ, Goldizen AW, Tromp SA (2010) Agamas exhibit behavioral
syndromes: bolder males bask and feed more but may suffer higher
predation. Behav Ecol 21:655661
Cliff N (1996) Ordinal methods for behavioral data analysis. Routledge
Cole EF, Quinn JL (2014) Shy birds play it safe: personality in captivity
predicts risk responsiveness during reproduction in the wild. Biol
Lett 10:20140178
Crawley, MJ (2007) Proportion data. In: The R Book. Wiley, Chichester
David M, Auclair Y, Czilly F (2011a) Personality predicts social dominance in female zebra finches, Taeniopygia guttata, in a feeding
context. Anim Behav 81:219224
David M, Czilly F, Giraldeau L-A (2011b) Personality affects zebra
finch feeding success in a producerscrounger game. Anim Behav
82:6167
David M, Auclair Y, Czilly F (2012) Assessing short- and long-term
repeatability and stability of personality in captive zebra finches
using longitudinal data. Ethology 118:932942
Dingemanse NJ, Both C, Drent PJ, van Oers K, van Noordwijk AJ (2002)
Repeatability and heritability of exploratory behaviour in great tits
from the wild. Anim Behav 64:929938
Dingemanse NJ, Both C, van Noordwijk AJ, Rutten AL, Drent PJ (2003)
Natal dispersal and personalities in great tits (Parus major). Proc R
Soc Lond B 270:741747

Behav Ecol Sociobiol

Dingemanse NJ, Both C, Drent PJ, Tinbergen JM (2004) Fitness consequences of avian personalities in a fluctuating environment. Proc R
Soc Lond B 271:847852
Duckworth RA (2006) Behavioral correlations across breeding contexts
provide a mechanism for a cost of aggression. Behav Ecol 17:1011
1019
Fawcett TW, Hamblin S, Giraldeau L-A (2013) Exposing the behavioral
gambit: the evolution of learning and decision rules. Behav Ecol 24:
211
Fournier DA, Skaug HJ, Ancheta J, Ianelli J, Magnusson A, Maunder M,
Nielsen A, Sibert J (2012) AD Model Builder: using automatic
differentiation for statistical inference of highly parameterized complex nonlinear models. Optim Methods Softw 27:233249
Gabriel PO, Black JM (2012) Behavioural syndromes, partner compatibility and reproductive performance in Stellers jays. Ethology 118:
7686
Gowaty PA, Anderson WW, Bluhm CK, Drickamer LC, Kim YK, Moore
AJ (2007) The hypothesis of reproductive compensation and its
assumptions about mate preferences and offspring viability. P Natl
Acad Sci USA 104:1502315027
Harris MR, Siefferman L (2014) Interspecific competition influences fitness benefits of assortative mating for territorial aggression in eastern bluebirds (Sialia sialis). PLoS ONE 9, e88668
Hinde CA (2006) Negotiation over offspring care?a positive response
to partner-provisioning rate in great tits. Behav Ecol 17:612
Hollander FA, van Overveld T, Tokka I, Matthysen E (2008) Personality
and nest defence in the great tit (Parus major). Ethology 114:405
412
Hunt J, Simmons LW (2002) Behavioural dynamics of biparental care in
the dung beetle Onthophagus taurus. Anim Behav 64:6575
Johnstone RA, Manica A, Fayet AL, Caswell Stoddard M, RodriguezGirons MA, Hinde CA (2014) Reciprocity and conditional cooperation between great tit parents. Behav Ecol 25:216222
Kralj-Fier S, Mostajo GAS, Preik O, Pekr S, Schneider JM (2013)
Assortative mating by aggressiveness type in orb weaving spiders.
Behav Ecol 24:824831
le Galliard JF, Paquet M, Cisel M, Montes-Poloni L (2013) Personality
and the pace-of-life syndrome: variation and selection on exploration, metabolism and locomotor performances. Funct Ecol 27:136
144
Lessells CM, McNamara JM (2012) Sexual conflict over parental investment in repeated bouts: negotiation reduces overall care. Proc R Soc
Lond B279:15061514
Limbourg T, Mateman AC, Lessells CM (2013) Parental care and UV
coloration in blue tits: opposite correlations in males and females
between provisioning rate and mates coloration. J Avian Biol 44:
1726
Mainwaring MC, Beal JL, Hartley IR (2011) Zebra finches are bolder in
an asocial, rather than social, context. Behav Process 87:171175
Mariette MM, Griffith SC (2012) Nest visit synchrony is high and correlates with reproductive success in the wild zebra finch Taeniopygia
guttata. J Avian Biol 43:131140
Mariette MM, Griffith SC (2015) The adaptive significance of provisioning and foraging coordination between breeding partners. Am Nat
185:270280
Mutzel A, Dingemanse NJ, Araya-Ajoy YG, Kempenaers B (2013)
Parental provisioning behaviour plays a key role in linking personality with reproductive success. Proc R Soc Lond 280:20131019
Nakagawa S, Cuthill IC (2007) Effect size, confidence interval and statistical significance: a practical guide for biologists. Biol Rev 82:
591605

Nicolaus M, Tinbergen JM, Bouwman KM, Michler SPM, Ubels R, Both

C, Kempenaers B, Dingemanse NJ (2012) Experimental evidence
for adaptive personalities in a wild passerine bird. Proc R Soc Lond
B 279:48854892
Patrick SC, Browning LE (2011) Exploration behaviour is not associated
with chick provisioning in great tits. PLoS ONE 6, e26383
Patrick SC, Chapman JR, Dugdale HL, Quinn JL, Sheldon BC (2012)
Promiscuity, paternity and personality in the great tit. Proc R Soc
Lond B279:17241730
Patterson LD, Schulte-Hostedde AI (2011) Behavioural correlates of parasitism and reproductive success in male eastern chipmunks, Tamias
striatus. Anim Behav 81:11291137
Rale D, Garant D, Humphries MM, Bergeron P, Careau V, Montiglio PO
(2010) Personality and the emergence of the pace-of-life syndrome
concept at the population level. Philos TRoy Soc B 365:40514063
Rivera-Gutierrez HF, Pinxten R, Eens M (2010) Multiple signals for
multiple messages: great tit, Parus major, song signals age and
survival. Anim Behav 80:451459
Rivera-Gutierrez HF, Pinxten R, Eens M (2012) Tuning and fading voices
in songbirds: age-dependent changes in two acoustic traits across the
life span. Anim Behav 83:12791283
Schielzeth H, Forstmeier W (2009) Conclusions beyond support: overconfident estimates in mixed models. Behav Ecol 20:416420
Schuett W, Dall SRX, Royle NJ (2011) Pairs of zebra finches with similar
personalities make better parents. Anim Behav 81:609618
Skaug H, Fournier D, Bolker B, Magnusson A, Nielsen A (2014)
Generalized linear mixed models using AD Model Builder. R package version 0.8.0
Smith BR, Blumstein DT (2010) Behavioral types as predictors of survival in Trinidadian guppies (Poecilia reticulata). Behav Ecol 21:
919926
Spoon TR, Millam JR, Owings DH (2006) The importance of mate
behavioural compatibility in parenting and reproductive success by cockatiels, Nymphicus hollandicus. Anim Behav 71:
315326
Torchiano M (2014) effsize: efficient effects size computation. R Package
Version 0.4.1
Tu YK, Gilthorpe MS (2007) Revisiting the relation between change and
initial value: a review and evaluation. Stat Med 26:443457
van Overveld T, Matthysen E (2010) Personality predicts spatial responses to food manipulations in free-ranging great tits (Parus
major). Biol Lett 6:187190
van Rooij EP, Griffith SC (2013) Synchronized provisioning at the nest:
parental coordination over care in a socially monogamous species.
Peer J 1, e232
Verbeek MEM, Boon A, Drent PJ (1996) Exploration, aggressive behaviour and dominance in pair-wise confrontations of juvenile male
great tits. Behaviour 133:945963
Webster MM, Ward AJW (2011) Personality and social context. Biol Rev
86:759773
Westneat DF, Hatch MI, Wetzel DP, Ensminger AL (2011) Individual
variation in parental care reaction norms: integration of personality
and plasticity. Am Nat 178:652667
Wilkin TA, King LE, Sheldon BC (2009) Habitat quality, nestling diet,
and provisioning behaviour in great tits Parus major. J Avian Biol
40:135145
Wolf M, van Doorn GS, Leimar O, Weissing FJ (2007) Life-history tradeoffs favour the evolution of animal personalities. Nature 447:581
585