ArticlePDF Available

Early life adversity increases foraging and information gathering in European starlings, Sturnus vulgaris

Authors:

Abstract and Figures

Animals can insure themselves against the risk of starvation associated with unpredictable food availability by storing energy reserves or gathering information about alternative food sources. The former strategy carries costs in terms of mass-dependent predation risk, while the latter trades off against foraging for food; both trade-offs may be influenced by an individual's developmental history. Here, we consider a possible role of early developmental experience in inducing different mass regulation and foraging strategies in European starlings. We measured the body mass, body condition, foraging effort, food consumption and contrafreeloading (foraging for food hidden in sand when equivalent food is freely available) of adult birds (!10 months old) that had previously undergone a subtle early life manipulation of food competition (cross-fostering into the highest or lowest ranks in the brood size hierarchy when 2 e12 days of age). We found that developmentally disadvantaged birds were fatter in adulthood and differed in foraging behaviour compared with their advantaged siblings. Disadvantaged birds were hy-perphagic compared with advantaged birds, but only following a period of food deprivation, and also spent more time contrafreeloading. Advantaged birds experienced a trade-off between foraging success and time spent contrafreeloading, whereas disadvantaged birds faced no such trade-off, owing to their greater foraging efficiency. Thus, developmentally disadvantaged birds appeared to retain a phenotypic memory of increased nestling food competition, employing both energy storage and information-gathering insurance strategies to a greater extent than their advantaged siblings. Our results suggest that subtle early life disadvantage in the form of psychosocial stress and/or food insecurity can leave a lasting legacy on foraging behaviour and mass regulation even in the absence of food insufficiency during development or adulthood.
Content may be subject to copyright.
Early life adversity increases foraging and information gathering in
European starlings, Sturnus vulgaris
Clare Andrews
a
,
*
,J
er
emie Viviani
a
,
b
, Emily Egan
a
, Thomas Bedford
a
, Ben Brilot
a
,
c
,
Daniel Nettle
a
, Melissa Bateson
a
a
Centre for Behaviour and Evolution, Institute of Neuroscience and Newcastle University Institute of Ageing, Newcastle University, Newcastle upon Tyne,
U.K.
b
D
epartement de Biologie,
Ecole Normale Sup
erieure de Lyon, Universit
e de Lyon, Lyon, France
c
School of Biological Sciences, Plymouth University, Plymouth, U.K.
article info
Article history:
Received 1 May 2015
Initial acceptance 2 June 2015
Final acceptance 22 July 2015
Available online
MS. number: 15-00365R
Keywords:
body mass regulation
contrafreeloading
developmental stress
early life adversity
European starling
food insecurity
foraging
Sturnus vulgaris
Animals can insure themselves against the risk of starvation associated with unpredictable food avail-
ability by storing energy reserves or gathering information about alternative food sources. The former
strategy carries costs in terms of mass-dependent predation risk, while the latter trades off against
foraging for food; both trade-offs may be inuenced by an individual's developmental history. Here, we
consider a possible role of early developmental experience in inducing different mass regulation and
foraging strategies in European starlings. We measured the body mass, body condition, foraging effort,
food consumption and contrafreeloading (foraging for food hidde n in sand when equivalent food is freely
available) of adult birds (10 months old) that had previously undergone a subtle early life manipulation
of food competition (cross-fostering into the highest or lowest ranks in the brood size hierarchy when 2
e12 days of age). We found that developmentally disadvantaged birds were fatter in adulthood and
differed in foraging behaviour compared with their advantaged siblings. Disadvantaged birds were hy-
perphagic compared with advantaged birds, but only following a period of food deprivation, and also
spent more time contrafreeloading. Advantaged birds experienced a trade-off between foraging success
and time spent contrafreeloading, whereas disadvantaged birds faced no such trade-off, owing to their
greater foraging efciency. Thus, developmentally disadvantaged birds appeared to retain a phenotypic
memory of increased nestling food competition, employing both energy storage and information-
gathering insurance strategies to a greater extent than their advantaged siblings. Our results suggest
that subtle early life disadvantage in the form of psychosocial stress and/or food insecurity can leave a
lasting legacy on foraging behaviour and mass regulation even in the absence of food insufcienc y during
development or adulthood.
© 2015 The Authors. Published on behalf of The Association for the Study of Animal Behaviour by Elsevier
Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Food availability is often unpredictable in the wild. One means
for animals to reduce the risk of starvation is to store energy re-
serves, while another is to gather information about alternative
food sources which could be useful later (Dall & Johnstone, 2002;
Mathot & Dall, 2013). Animals facing increased starvation risk or
reduced access to food due to competition maintain more stored
reserves (Houston & McNamara, 1993; Witter & Swaddle, 1995),
which they must forage to obtain. However, they must trade off
carrying reserves against increased costs of locomotion or mass-
dependent predation (Macleod, Gosler, & Cresswell, 2005).
Environmental stochasticity also impacts optimal sampling of the
environment to gain information (Dunlap & Stephens, 2012;
Keasar, Motro, & Shmida, 2013; Shettleworth, Krebs, Stephens, &
Gibbon, 1988). With limited time and energy to invest, animals
will also face a trade-off between the need to gain food and the
need to gather information about alternative food sources. These
trade-offs governing foraging for food or gathering information,
and storing energy reserves, may be inuenced by developmental
history. For example, developmental history may impact locomotor
ability (O'Hagan, Andrews, Bedford, Bateson, & Nettle, 2015) and
hence predation risk. Understanding the impact of early environ-
mental conditions on foraging decisions and information use is
critical to evolutionary biologists studying phenotypic plasticity
and the causes of individual behavioural differences (Dall, Bell,
* Correspondence: C. Andrews, Institute of Neuroscience, Newcastle University,
Henry Wellcome Building, Framlington Place, Newcastle NE2 4HH, U.K.
E-mail address: clare.andrews@ncl.ac.uk (C. Andrews).
Contents lists available at ScienceDirect
Animal Behaviour
journal homepage: www.elsevier.com/locate/anbehav
http://dx.doi.org/10.1016/j.anbehav.2015.08.009
0003-3472/© 2015 The Authors. Published on behalf of The Association for the Study of Animal Behaviour by Elsevier Ltd. This is an open access article under the CC BY
license (http://creativecommons.org/licenses/by/4.0/).
Animal Behaviour 109 (2015) 123e132
Bolnick, & Ratnieks, 2012; Mathot & Dall, 2013). It may also aid
understanding of the developmental origins of food-related con-
ditions such as obesity and metabolic syndrome in humans (e.g.
Cottrell & Ozanne, 2008; Gluckman & Hanson, 2007; Khlat, Jusot, &
Ville, 2009; Levin, 2006). We consider here the possible role of early
developmental history in altering an individual's mass regulation,
foraging behaviour and information gathering.
Optimal foraging theory (Stephens & Krebs, 1986) predicts that
animals should preferentially forage on patches offering the highest
net rate of energy intake. Paradoxically, in many species including
humans, animals sometimes choose to forage in patches in which
effort is required to exploit the food resource despite the existence
of alternative patches containing identical food available for mini-
mum effort (Coulton, Waran, & Young, 1997; Inglis, Forkman, &
Lazarus, 1997; Larson & Tarte, 2013; Lindqvist & Jensen, 2008;
Neuringer, 1969; Ogura, 2011; Robertson & Anderson, 1975;
Vasconcellos, Harumi, & Ades, 2012). This behaviour whereby an-
imals work for food even when the same food is freely available is
known as contrafreeloading (CFL). The information hypothesis
offers a functional explanation for CFL, suggesting that it is a form of
sampling to gather information that may be useful in future
foraging attempts (Bean, Mason, & Bateson, 1999; Forkman, 1996;
Inglis & Ferguson, 1985). Animals performing CFL demonstrate
that they know the location of the most protable foraging patches
when the previously most protable patch is removed, or while
they are more hungry (Bean et al., 1999; Forkman, 1996; Lindqvist,
Schütz, & Jensen, 2002). Current energetic state appears to affect
the trade-off between foraging and gathering information, with
food deprivation decreasing CFL in starlings (Bean et al., 1999; Inglis
& Ferguson, 1985) and chickens, Gallus gallus domesticus (Lindqvist
et al., 2002). Yet despite growing interest in the drivers of individual
differences in information use (Mathot & Dall, 2013), the potential
impact of past environment on this trade-off, and hence informa-
tion gathering via CFL, has not yet been investigated.
Adverse environmental conditions experienced early in life can
strongly impact adult survival and health, in humans and other
animals (Lindstr
om, 1999; Lummaa & Clutton-Brock, 2002). The
relationship between prenatal or early postnatal environment and
obesity is an active research
eld (Cottrell & Ozanne, 2008;
Gluckman & Hanson, 2007; Khlat et al., 2009; Levin, 2006),
although there has been limited attention on the evolutionary and
associated proximate behavioural mechanisms involved. Evidence
from rodent and avian models suggests that adult foraging de-
cisions may be inuenced by developmental history. Early food
restriction of rodents induces an increased drive to obtain and
consume food in adulthood, with rats, Rattus norvegicus, whose
mothers were fed restricted diets in the perinatal period being
hyperphagic long beyond weaning (Coup
e, Grit, Darmaun, &
Parnet, 2009; Orozco-S
olis et al., 2009; Qasem et al., 2012;
Vickers et al., 2000). In starlings, increased competition for food
in the nest affects foraging decisions in adulthood in terms of the
trade-off between the benets of obtaining nutrients and the costs
of ingesting toxins by consuming toxic prey (Bloxham, Bateson,
Bedford, & Nettle, 2014). The propensity to contrafreeload may
also be impacted by growth early in life. In broiler chickens
(selected for high feed conversion efciency) a reduction in CFL
compared with layer chickens or the ancestral red junglefowl,
Gallus gallus, has been interpreted as a selected energy-saving
response to the demands of rapid growth (Lindqvist & Jensen,
2008; Lindqvist et al., 2002; Lindqvist, Zimmerman, & Jensen,
2006; Schütz & Jensen, 2001). However, no study has yet experi-
mentally investigated how developmental history may inuence
the amount of CFL individuals choose to perform.
In the current study, we used a recently developed subtle early
life manipulation of food competition in wild European starlings to
investigate the effect of early life adversity on adult foraging
behaviour and body mass. The European starling is a species known
to reliably display CFL, showing considerable individual variation in
this behaviour (Bean et al., 1999; Inglis & Ferguson, 1985). To con-
trol for genetic effects, we cross-fostered siblings on day 2 of life
until day 12, to a nest in which they were either slightly larger than
the other chicks (the advantaged treatment) or slightly smaller (the
disadvantaged treatment). Previous similar manipulations suggest
that disadvantaged nestlings would have had to beg more in order
to be fed, and would have been jostled to more peripheral positions
in the nest (Cotton, Wright, & Kacelnik, 1999; Kacelnik, Cotton,
Stirling, & Wright, 1995). Although our manipulation did not
affect growth, there was evidence that it did increase develop-
mental stress. Indeed, telomere attrition, an acknowledged
biomarker of developmental stress exposure, was greater in the
disadvantaged than the advantaged nestlings, yet they did not
differ in growth (Nettle, Monaghan, et al., 2015). Following the
manipulation, we brought the birds into captivity and reared them
to adulthood under uniform conditions. When they were between
10 and 13 months of age we studied their energy reserves and
foraging on freely available food versus contrafreeloading via
searching for food hidden in sand.
In light of theory and previous ndings in rodents, we predicted
that developmentally disadvantaged birds would be hyperphagic
and carry greater energetic reserves than the advantaged birds.
Since our manipulation did not detectably impact growth (Nettle,
Monaghan, et al., 2015), we might assume that developmentally
disadvantaged starlings would behave unlike fast-growing broiler
chickens (which show reduced CFL to conserve energy). Instead, we
predicted that disadvantaged birds would show increased CFL as a
means to gather information as insurance against future food
insecurity.
METHODS
Overview
Subjects were 37 European starlings (23 male, 14 female; 19
advantaged, 18 disadvantaged) from 10 natal families. These birds
were subject to a developmental manipulation of early life
competition as nestlings in the eld (day 2e12 posthatching)
whereby they were cross-fostered in sibling pairs into 20 different
host nests in which they were either the smallest (disadvantaged
treatment) or largest (advantaged treatment) nestlings in the brood
size hierarchy, after which time they were raised under identical
laboratory conditions. Birds were 10e13 months old at the time of
the current experiment. For the experiment, birds were housed
individually and pretrained to forage for food hidden in sand. Once
a stable level of food consumption was achieved, birds entered an
experimental phase during which we presented them with a choice
between foraging for freely available food or identical food hidden
in sand during a 2 h trial, daily for 4 days. We measured foraging
effort and consumption during trials, as well as food consumption
outside trials when food was freely available.
Developmental Manipulation
The developmental manipulation is described in detail else-
where (Nettle, Monaghan, et al., 2015). Briey, in 2013 on post-
hatching day 2 (henceforth day 2), we removed 12 quartets of
siblings from the natal nest and cross-fostered them to two
different host nests per quartet; the two randomly selected nes-
tlings in the advantaged (ADV) condition were fostered to a nest
where they were (mean ± SD) 4.9 ± 1.9 g larger than all other
nestlings, and the two in the disadvantaged (DIS) condition to a
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132124
nest where they were 4.8 ± 2.2 g smaller than the other nestlings.
The composed brood size ranged from four to six chicks but was
matched for the ADV and the DIS half of a focal quartet. These size
differences against an average day 2 weight of 13 g were less
extreme than the average difference between the smallest and
largest chicks in unmanipulated broods (10.4 ± 0.6 g against an
average day 4 weight of 20 g; Andrews et al., n.d.). On day 12, cross-
fostered nestlings were removed to the laboratory (complete sur-
viving quartets only) where the natal families were reconstituted
and nestlings hand-reared to independence at around 6 weeks of
age (for details of hand-rearing methods see Feenders & Bateson,
2011), whereafter they lived in communal aviaries. ADV and DIS
nestlings did not differ signicantly in body mass at any age during
development (measured at days 3, 4, 7, 12, 15, 18, 21 and 24),
although the ADV nestlings remained signicantly heavier than
their unrelated host nest competitors at day 12, while the DIS
nestlings remained signicantly smaller than their host nest com-
petitors (Nettle, Monaghan, et al., 2015). Wing lengths did not differ
signicantly by treatment at day 12 or af ter edging on day 24
(Nettle, Monaghan, et al., 2015).
Housing and Husbandry
When not in experimental procedures, we housed birds in
mixed-sex groups of up to 20 in two indoor aviaries (215 340 cm
and 220 cm high; ca. 18
C; 40% humidity; 15:9 h light:dark cycle).
Birds were provided with environmental enrichment (foraging
substrate, water baths, multilevel rope perches, suspended card-
board boxes as cover), clean drinking water, and were fed ad libi-
tum on domestic chick crumb (Special Diets Services Poultry
Starter (HPS)), supplemented with cat biscuits (Royal Canin Ltd.
Fit), dried insect food (Orlux insect p
^
at
e), live mealworms and
fruit. The birds were maintained in nonbreeding condition by
means of an unchanging light:dark cycle of long days. For the
experiment, birds were taken into the experimental laboratory in
groups of eight, with all four members of a natal family in the same
group. Birds were individually housed in wire-mesh cages
(75 45 cm and 45 cm high) tted with two wooden perches, two
water bottles, and a central wooden divider (8.5 cm high) across the
centre of the cage. Bowls of water for bathing were provided daily
for 20 min. The birds were maintained under a 15:9 h light:dark
cycle, at ca. 18
C; 40% humidity. Birds were never acoustically
isolated, but were visually isolated from one another during the
experimental sessions by means of opaque barriers between cages.
Following this study, birds were either kept for further experiments
at Newcastle University or rehomed to outdoor aviaries.
Pretrial Training
Birds were initially acclimatized to individual cages and foraging
for food (domestic chick crumb) hidden in sand. This was done in
order to remove any potential effects of differential neophobia to
the apparatus or variation in the time course of a stress response to
handling, and to achieve stable levels of food consumption prior to
the experiment. Birds were already familiar with eating freely
available crumb from ad libitum feeding in aviaries. We mixed food
with sand to increase the amount of effort required to access the
food. Starlings will forage for food in sand and previous experi-
ments conrm that they are not able to assess food quantity in sand
using the sense of smell (Bean et al., 1999). Previous studies of avian
CFL have employed similar methods, requiring birds to search for
food mixed with sand or wood shavings (Bean et al.,1999; Lindqvist
& Jensen, 2008; Lindqvist et al., 2002, 2006; Schütz & Jensen,
2001).
Birds were caught from the aviary, weighed and placed indi-
vidually in cages. Visual contact was maintained for the initial 3
days to aid habituation to cages, with barriers then placed between
cages overnight (1700e0930 hours) to habituate birds to these
prior to their use in experimental trials. Daily at noon following
husbandry, we placed a bowl (18 cm diameter, 4 cm deep) con-
taining 100 g of crumb (ad libitum quantity) and 200 g of sand into
each cage, alternating on which side of the central divider the bowl
was positioned. Birds also received four live mealworms (Tenebrio
sp.) in a separate bowl. To obtain accurate food consumption
measures, we sieved the crumb prior to use to remove dust or
smaller pieces. We removed the bowl from the previous 24 h period
and sieved the contents to collect the remaining crumb, which we
weighed to measure food consumption. The training period
continued until all eight birds ate at least 15 g of crumb in 24 h
(maximum 9 days; mean 6.3 days).
Experimental Phase
We measured CFL and free food consumption on 4 consecutive
days because levels of CFL uctuate within individual starlings
(Bean et al., 1999). On the day preceding an experimental trial, we
positioned barriers to visually isolate the birds at 1700 hours. On
the morning of an experimental trial, we removed the ad libitum
food bowl at 0800 hours, and weighed the remaining crumb to
measure food consumption outside experimental trials (nontrial
food consumption). Birds were then food deprived from 0800 to
1000 hours, after which we turned off the lights and placed two
bowls in each cage. One bowl (CFL bowl) contained 39.5 g of
crumb covered with 200 g of sand, with 0.5 g of crumb sprinkled
on top of the sand. The other bowl (free food bowl) contained
200 g of crumb. The side position of the bowls was alternated
between days. Trials began by switching on the lights once the
experimenters had left the room. They lasted 2 h and were lmed
using video cameras (Sony Handycam DCR-SR32) positioned
outside the cages. At the end of the trial, any spilt food was placed
in the appropriate bowl (determined by which side of the divider
it was on) and both bowls were then removed. The remaining
crumb was weighed to measure food consumption from each bowl
(CFL consumption; free food consumption), following sieving of
the CFL bowl to remove the sand. We then provided each cage
with ad libitum food consisting of a bowl containing 200 g of
crumb and four mealworms, and removed visual barriers until
1700 hours.
An observer blind to treatment scored by continuous obser-
vation of all trials on the video footage the duration of time spent
in, perched on or with head over each bowl (CFL duration; free
feeding duration). We used duration as a measure of foraging/CFL
effort since this metric relates to the opportunity costs entaile d.
We measured these behaviours to conrm that f oraging ef-
cien cy was lowered by the presence of sand in the CFL bowl, and
because acquiring information about patch qua lity may involve
not only ingesting food but also probing in sand without
ingesting food (Bean et al., 1999; Lindqvist et al., 2002).
Following the fourth trial, birds were caught and weighed. We
used the mean of cage entry body mass and experiment end
body mass for analyses of body mass. To examine whether hy-
perphagia occurr ed during trials, we summed for each day CFL
consumption and free food consumption to give trial consump-
tion. To examine whether hyperphagia oc curred overall, we
calculated for each day the mass of crumb consumed (total food
consumption) in 22 h as the sum of trial consumption during the
2 h trial plus nontrial food consumption during the 20 h
foll owing the trial.
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132 125
Ethical Note
Our study adhered to the ASAB/ABS Guidelines for the Use of
Animals in Research, and was approved by Newcastle University
local ethical review committee. Work was conducted under U.K.
Home Ofce project licence number PPL60/4073, and the removal
of starlings from the wild was authorized by Natural England
(licence number 20121066). Fieldwork was carried out with the
permission of landowners, with the number and duration of nest
disturbances minimized. One chick of 48 that we cross-fostered
died between cross-fostering and the next morning; this is no
greater than the expected rate of mortality this early in life. All
other cross-fostered chicks gained weight between fostering and
the next morning, suggesting rapid recovery from transport and
acceptance in host nests. The manipulation was intended to in-
crease developmental stress in the disadvantaged group. However,
the level of within-brood size discrepancy created is within the
natural range observed in starling nests (see Developmental
Manipulation). Thus, the level of developmental stress was likely to
have been within the naturally experienced range. The manipula-
tion was also equally likely to improve a nestling's position within
its nest as to make it worse. The mean weights for the disadvan-
taged birds were not signicantly lower than for the advantaged
birds at any weighing point during development; nor was the
variance greater (Nettle, Monaghan, et al., 2015). Two disadvan-
taged birds and three advantaged birds died before day 12; this
suggests that our disadvantageous manipulation did not result in
excessive mortality and is in line with natural mortality rates in
starlings (Feare, 1984). Sex was determined by molecular sexing as
part of a related study using blood samples of 75
m
l taken from the
alar vein (see Nettle, Monaghan, et al., 2015). A small as possible
needle was used and volume taken, and antiseptic cream was
applied to the puncture site to minimize risk of infection. Stress due
to catching adult birds was minimized by doing so in a darkened
room using torchlight, with up to three persons catching simulta-
neously to minimize the time taken, and holding birds in cloth bags
for the shortest possible time. The acute stress response is relatively
short lived (Rich & Romero, 2005) and food consumption stabilized
during the several days of pretrial training; thus variation in stress
responses is likely to have had minimal impact on the experimental
phase. The CFL experiment is likely to have induced short-term
stress due to visual social isolation and the unfamiliar foraging
environment. Birds were returned to their aviary within 22 days of
removal. None showed any subsequent adverse effects.
Statistical Analysis
From body mass measured after the nal trial, we calculated
body condition index as the residual mass corrected for tarsus
length (mean of left and right tarsus measured on day 24, by which
time the tarsus is fully grown) using an equation (body
condition ¼ mass 2.9158 tarsus þ 18.1278) derived from a
simple regression of mass measured at day 115e123 (when birds
were group-housed in aviaries for a sufcient period to obtain
stable body mass) on day 24 tarsus. Residual condition indices offer
a reasonable proxy for fat mass (Labocha & Hayes, 2012).
Data were analysed using general linear mixed models in R (R
Development Core Team, 2011), using the base statistical pro-
cedures and package nlme (Pinheiro, Bates, DebRoy, Sarkar, & R
Core Team, 2015). Model estimation was by maximum likelihood,
and whether parameters differed signicantly from 0 was deter-
mined by a likelihood ratio test (LRT) on the difference in model
deviance (
c
2
distributed) when the parameter was removed from
the model. The degrees of freedom for this test equals the differ-
ence in the number of free parameters of the models that include
and exclude the term of interest, which for the purposes of our
study equated to 1 in all instances. We assumed a criterion for
signicance of P < 0.05 throughout; results with P < 0.10 are also
reported and discussed as marginally nonsignicant trends.
We describe the main results relevant to the experimental hy-
potheses below and provide detailed specication of statistical
models and output in Appendix Table A1, to which we refer by
model number below. The raw data and R script are available as
Supplementary Material. The response variables we examined
were body mass (Appendix Table A1, Model 1), body condition
index (Model 2), total food consumption (Model 3), trial con-
sumption (Models 4 and 9), CFL consumption (Model 5), free food
consumption (Model 6), CFL duration (Model 7), foraging rate
(Model 8) and nontrial consumption (Model 10). The basic model
for each response variable we studied included xed effects for sex
(since numbers were unbalanced between treatments), develop-
mental treatment (ADV/DIS) and the sex)treatment interaction.
Models included, where appropriate, nested random effects for
individual bird identity (since the same individuals were measured
for multiple days) and natal family (since quartets of birds were
siblings). We did not include mass or body condition as model
covariates in models of foraging behaviour since these could be
consequences, rather than causes, of variation in foraging behav-
iour. The distribution of the outcome variables CFL consumption
and foraging rate were zero inated and truncated, requiring
transformation to meet model assumptions. We experimented
with different error structures and transformations and present
here those giving satisfactory distribution of residuals (Gaussian
error for all models; log transform of CFL consumption; square root
transform of foraging rate). The sample size for all models was 37
birds.
RESULTS
Mass and Body Condition
We examined the effect of developmental treatment on body
mass and condition (Models 1 and 2). ADV and DIS birds were of
equivalent body mass at the time of the present experiment (Fig. 1a,
Appendix Table A1, Model 1). Birds were, on average, lighter at the
time of our experiment than when we calculated the body condi-
tion index equation (see Statistical Analysis); thus mean body
condition indices were negative in our experiment (Fig. 1a). It is
usual for birds to be lighter when individually caged than when in
aviary groups (Feenders & Bateson, 2013), possibly due to reduced
competition when alone (Witter & Swaddle, 1995). At the time of
our experiment, DIS birds had signicantly higher body condition
indices than ADV birds (Fig. 1b, Appendix Table A1, Model 2).
Food Consumption
We compared the amount of food consumed by DIS and ADV
birds both during trials (Model 4) and in total (Model 3) since
treatment could potentially inuence sensitivity to food depriva-
tion which preceded the trials as well as overall hyperphagia. Total
food consumption was marginally nonsignicantly greater for DIS
birds than ADV birds (Fig. 2, Appendix Table A1
, Cohen's d ¼ 0.469,
Model 3). There was a stronger effect of developmental treatment
on trial consumption, with DIS birds eating signicantly more food
than ADV birds (Fig. 2, Appendix Table A1, Cohen's d ¼ 0.691, Model
4).
Both DIS and ADV birds consumed more food from the free food
bowl than by CFL (Fig. 2; greater free food consumption than CFL
consumption in 132 of 148 trials). We compared CFL consumption
(log transformed) of ADV and DIS birds while statistically
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132126
controlling for free food consumption (Model 5). We included free
feeding as a covariate because the amount consumed from the CFL
bowl (assuming a xed total energy requirement) is logically not
independent from that at the alternative bowl. Indeed, CFL and free
food consumption were negatively related for both ADV and DIS
birds (estimate ± SE for free food consumption ¼0.194 ± 0.029;
Appendix Table A1, Model 5). DIS birds consumed marginally
nonsignicantly more food by CFL than ADV birds (Fig. 2, Appendix
Table A1, Model 5). Comparing free food consumption while con-
trolling for CFL consumption as a covariate (Model 6), we found that
DIS birds ate more free food than ADV birds (Appendix Table A1,
Model 6).
Foraging and Information Gathering Effort
We assessed the effort put into foraging for food or information
gathering in terms of the time spent at the free food bowl or CFL
bowl, respectively. Both ADV and DIS birds spent more time
foraging on freely available food than CFL (Fig. 3; mean free feeding
duration > mean CFL duration at every trial for both ADV and DIS
birds). We compared CFL effort of ADV and DIS birds in terms of the
time spent foraging at the CFL bowl while statistically controlling
for free feeding duration as covariate (Model 7). We did this
because time spent on CFL (given the xed trial length) is logically
not independent from that at the alternative bowl. Indeed, CFL
duration was negatively related to the time spent foraging on free
food (estimate ± SE for free feeding duration ¼0.127 ± 0.059;
Appendix Table A1, Model 7). DIS birds spent marginally nonsig-
nicantly more time on CFL than ADV birds (Fig. 3, Appendix
Table A1, Model 7). There was a signicant interaction between
free feeding duration and treatment, indicating DIS birds spent
more time on CFL for a given free foraging effort than ADV birds
(Appendix Table A1, Model 7).
Foraging Efciency and Trade-offs
To conrm whether foraging was less efcient in the CFL bowl
than when free feeding, as we had intended in our design, we used
as a measure of foraging rate the food consumption per unit time
(g/s) spent at each bowl (square-root transformed), including bowl
(CFL or free food) as a xed factor along with its interaction with
treatment (Model 8). As intended, foraging in sand in the CFL bowl
was less efcient than foraging in the free food bowl for both ADV
and DIS birds (Fig. 4, Appendix Table A1, Model 8). On average,
foraging in the CFL bowl resulted in a 40% reduction in foraging rate
compared with free feeding. There was also a signicant interaction
between bowl and developmental treatment (Fig. 4,
Appendix
Table A1, Model 8). Although DIS birds foraged more efciently
than ADV birds on free food (mean DIS foraging efciency > mean
ADV foraging efciency in three of four trials), the difference be-
tween treatments was not present for CFL (Fig. 4).
0
20
40
80
DIS ADV
DIS
(a)
10
7.5
5
2.5
0
DIS
ADV DIS
(b)
60
Body mass (g)Body condition index
ADV
ADV
Female Male
Figure 1. (a) Body mass, and (b) body condition index (residual mass) of develop-
mentally advantaged (ADV) and disadvantaged (DIS) male and female starlings.
Means ± 1 SE of raw data are shown.
ADV ADV DIS ADV DIS ADV DISDIS
Free food Nontrial TotalCFL
0
10
20
Food consumed (g)
Figure 2. Mean mass of crumb consumed per day by developmentally advantaged
(ADV) and disadvantaged (DIS) starlings from the contrafreeloading (CFL) bowl and
free food bowl during the trial, the nontrial food consumption and total food con-
sumption (i.e. the sum of the other measures). Means ± 1 SE of raw data are shown.
0
100
DIS
ADV DIS
ADV
Free feedin
g
Contrafreeloadin
g
200
Time spent foraging (s)
Figure 3. Foraging duration at free food and contrafreeloading bowls by advantaged
(ADV) and disadvantaged (DIS) starlings. Means ± 1 SE of raw data are shown.
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132 127
We expected CFL to occur at the expense of a reduction in
overall foraging success. To examine the trade-off between CFL and
foraging success, we analysed trial consumption including CFL
duration and its interaction with treatment as a covariate (Model
9). If a trade-off occurs then birds that spend more time on CFL
would on average consume less food in total during the trial. The
predicted trade-off between foraging and information gathering
was present for ADV birds but not for DIS birds, as indicated by a
negative relationship between trial consumption and the time
spent on CFL for ADV birds only (Fig. 5, Appendix Table A1, Model
9). Since it is possible in our experiment that birds that spent more
time on CFL during trials might compensate by consuming addi-
tional free food outside trials, we also analysed nontrial food con-
sumption including CFL duration and its interaction with treatment
as a covariate (Model 10). There was no relationship between
nontrial food consumption and time spent on CFL (Appendix
Table A1, Model 10).
DISCUSSION
Early life competitive disadvantage exerted a long-lasting in-
uence on starlings' motivation to forage for food and gather in-
formation about food sources. Despite the absence of an effect of
developmental treatment on early growth (Nettle, Monaghan, et al.,
2015), developmentally disadvantaged birds were fatter in adult-
hood at the time of our study than advantaged birds (Fig. 1) and
showed differences in their foraging behaviour. Disadvantaged
starlings were hyperphagic following food deprivation (Fig. 2) and
foraged at a faster rate on freely available food (Fig. 4) than their
advantaged siblings. They also spent more time gathering infor-
mation about food via CFL (Fig. 3), and consumed marginally more
food by CFL, than advantaged birds (Fig. 2). The expected trade-off
between foraging and CFL was conrmed for advantaged starlings,
since those that contrafreeloaded more also consumed less food
during trials (Fig. 5). However, there was no trade-off for disad-
vantaged birds, probably because of their higher foraging efciency
on freely available food. We predicted that developmentally
disadvantaged starlings would contrafreeload more than advan-
taged birds, this being a means to gather information that could
serve as insurance against future changes in food availability. As
expected, we found that disadvantaged birds spent more time on
CFL and consumed marginally more food by CFL than advantaged
birds. Also as predicted, disadvantaged birds consumed more food
than advantaged birds, but signicantly so only following food
deprivation. Disadvantaged birds also carried more energy reserves
(fat), as expected. Thus developmentally disadvantaged starlings
appeared to employ both energy storage and information gathering
(CFL) foraging strategies to a greater extent than advantaged birds.
Since our manipulation had no detectable impact on the timing of
weight gain, or weight or wing length at edging (Nettle,
Monaghan, et al., 2015), the impact of developmental history on
foraging and information gathering in our birds as adults appears
not to be wholly due to variation in the overall pattern of growth
(i.e. to food insufciency). Our study thus demonstrates that early
life disadvantage can have long-lasting effects on foraging and mass
regulation even in the absence of food insufciency during devel-
opment. Instead, psychosocial stress and/or food insecurity
resulting from competition in early life appears to have been suf-
cient to cause these differences. Below, we defend our measure of
CFL then discuss our results in terms of adaptive developmental
plasticity, and consider the extent to which our ndings support
the idea that subtly developmentally disadvantaged birds show a
memory of hunger (
Bloxham et al., 2014).
Several lines of evidence support the assumption that our
experiment measured CFL, that is, working for food when equiva-
lent food is freely available, in order to gather information. First,
foraging in sand entailed a 40% reduction in food intake rate
compared with free feeding, as expected if work was entailed.
Second, at least for advantaged birds, the more time a bird spent
searching for food in sand, the less food it consumed during a trial;
thus CFL carried a foraging opportunity cost. As with laboratory
studies generally, it is difcult to translate such a decit into a
tness cost. In our experiment, birds that spent more time on CFL
did not appear to compensate for lost foraging opportunity by
increasing food intake outside trials when only free food was
available. In the wild, CFL opportunities may be present throughout
the day, so the trade-off against free feeding could have potentially
greater impact on overall energy intake. Third, a previous study
found that starlings search for food in sand more if the surface is
visually occluded, suggesting that information gathering is a driver
0
0.01
0.02
ADV ADV DISDIS
Contrafreeloadin
g
Free feedin
g
0.03
Foraging rate (g/s)
Figure 4. Foraging rate (food consumption per unit time spent at bowl) in contra-
freeloading and free food bowls by developmentally advantaged (ADV) and disad-
vantaged (DIS) starlings. Means ± 1 SE of raw data are shown.
3
4
5
6
8
0 100 150 200
50
CFL duration (s)
7
Trial consumption (g)
Figure 5. Trade-off between foraging and contrafreeloading (CFL) for developmentally
advantaged (ADV) and disadvantaged (DIS) starlings. Linear regression lines are shown
for ADV (red) and DIS (black) birds; data points are means per bird over the four trials.
Dashed lines show mean trial food consumption or contrafreeloading duration for ADV
(red) and DIS (black) birds.
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132128
(Bean et al., 1999) rather than the behaviour simply representing
exploration or activity per se without information gain. CFL may in
fact be a form of exploration since exploration has been considered
as a means to gain information about the environment (Inglis,
Langton, Forkman, & Lazarus, 2001).
Unlike rodent models of early life adversity, our disadvantaged
starlings were not strongly hyperphagic overall, since their daily
food consumption was only marginally increased compared with
their advantaged siblings. Disadvantaged birds did show signicant
hyperphagia, however, following 2 h of food deprivation. Hyper-
phagia in rats is especially pronounced for very high-quality foods
(Vickers et al., 2000), whereas in the current study starlings were
foraging on a lower quality (low-protein) food. Consistent with this
explanation for limited hyperphagia overall, in a previous study,
starlings raised in enlarged brood sizes were hyperphagic only for
high-protein foods (mealworms) and not for a lower-protein
(crumb) diet (Bloxham et al., 2014). To elucidate the factors inu-
encing hyperphagia, future work should examine interactions be-
tween current energetic state and reserves, dietary quality and
developmental history on foraging and information gathering.
As evidenced by their greater foraging efciency on free food as
well as increased CFL effort, developmentally disadvantaged star-
lings showed greater motivation for both foraging for food and
information gathering. Thus they appear to retain a phenotypic
memory of increased early food competition. Increased foraging in
response to current food restriction has been reported in a range of
species, including humans (Garner, 1997; Redman & Ravussin,
2011; Vitousek, Manke, Gray, & Vitousek, 2004). Increased body
mass index is also associated with current food insecurity in
humans (Adams, Grummer-Strawn, & Chavez, 2003; Martin-
Fernandez, Caillavet, Lhuissier, & Chauvin, 2014; Townsend, Peer-
son, Love, Achterberg, & Murphy, 2001). Our study adds to this by
demonstrating a lasting effect of prior disadvantage in food
competition on adult foraging behaviour and energy reserves. The
results are consistent with our previous study in which we found
that starlings reared in enlarged broods also showed a phenotypic
memory of hunger, in the form of reduced dietary selectivity of
toxic prey (Bloxham et al., 2014). From an adaptive perspective, it
may be benecial for animals faced with high competition and/or
low food reliability to increase their foraging activity and infor-
mation gathering, thus increasing the relative priority of getting
food and information concerning sources of food over other
competing activities. Wild animals often reduce the performance of
maintenance behaviours, play and afliative behaviour when food
is scarce (D'Eath, Tolkamp, Kyriazakis, & Lawrence, 2009). It re-
mains to be investigated what cost, if any, such foraging prioriti-
zation may entail for disadvantaged starlings. Possibilities
warranting further study include reduced vigilance or feather
maintenance.
In terms of CFL, disadvantaged starlings behaved unlike birds
experiencing acute current food restriction, the latter reducing CFL
following 8 h of food deprivation (Bean et al., 1999). Reduction in
CFL has been interpreted as an energy-conserving strategy
(Lindqvist et al., 2002). Therefore the difference may be because our
developmental treatment is likely to have altered perceptions of
food uncertainty as opposed to birds' current energy intake
requirement, as birds in our study were minimally food deprived
(2 h) during trials and disadvantaged birds were equal in body mass
and fatter than their advantaged siblings. The greater CFL by
disadvantaged birds might alternatively be in part a consequence of
their apparent higher foraging motivation overall. However, moti-
vational differences cannot completely account for the greater CFL
by disadvantaged birds since even for a given free foraging effort,
disadvantaged birds spent more time on CFL. Since disadvantaged
birds spent both absolutely and proportionally more time on CFL
than their advantaged siblings they invested more in information
gathering directly. An alternative nonadaptive possibility is that
disadvantaged birds spent more time on CFL (and as a result
consumed marginally more contrafreeloaded food), owing to a
reduced learning capacity making them less efcient in acquiring
information about the foraging patch. Learning performance is
reduced in zebra nches, Taeniopygia guttata, experiencing low-
quality early nutrition (Brust, Krüger, Naguib, & Krause, 2014).
Reduced learning ability has been suggested to explain the reduced
CFL of rats reared in enriched environments (Coburn & Tarte, 1976).
However, we consider an explanation based on differential learning
ability unlikely since independent measures of learning did not
differ between these same advantaged and disadvantaged birds
(Nettle, Andrews, et al., 2015).
Under an adaptive developmental plasticity framework, it is
possible that our disadvantaged birds were less robust to food
deprivation in a way not captured by our gross measures of body
mass or condition. That hyperphagia was more pronounced in
disadvantaged birds following food deprivation is at least consis-
tent with this. Within-brood competition (Verhulst, Holveck, &
Riebel, 2006) and early exposure to the stress hormone cortico-
sterone (Spencer & Verhulst, 2008) can raise adult metabolic rate.
Thus, is it possible that disadvantaged birds have a differing
metabolic requirement as adults, perhaps altering their sensitivity
to food shortages and hence making information about alternative
foraging patches more valuable to them or less costly to obtain, or
reducing the utility of a given level of reserves (Mathot & Dall,
2013). Other cryptic physiological differences are possible, for
example, physiological constraints such as digestive capacity can be
inuenced by feeding history in nestlings (Wright et al., 2002),
although the persistence of such effects into adulthood is unknown.
The optimal level of sampling the environment to gain infor-
mation and hence reduce uncertainty over food supply theoreti-
cally depends on the relative costs of missing good foraging
opportunities versus the costs of sampling, and the rate at which
the environment changes. These factors may be inuenced by
developmental history, suggesting an alternative adaptive expla-
nation for increased CFL by disadvantaged starlings. Birds experi-
encing a poor start may be disadvantaged in social competition as
adults owing to a lower phenotypic quality, making their foraging
environment effectively more uncertain. Experimentally elevated
corticosterone in zebra nch nestlings inuences social network
position in adulthood (Boogert, Farine, & Spencer, 2014). Disad-
vantaged birds could possibly be more vulnerable to food unpre-
dictability induced through competition, again making information
on alternative food sources more valuable to them. In the wild,
developmental disadvantage could be predictive of a more uncer-
tain environment per se independent of competition, perhaps
selecting for developmentally plastic strategies to insure against or
reduce uncertainty via sampling the environment by CFL. In op-
position to these hypotheses, CFL has been shown to be reduced by
increased environmental uncertainty (Forkman, 1991) as well as by
increased effort to obtain earned food (Rutter & Nevin, 1990), both
of which are likely to result from inferiority in social competition.
Our sibling birds experienced only 10 days of differential
experience as nestlings while living their subsequent lives in the
same environment, and did not differ in early growth. Even such
limited difference in exposure to competition appears sufcient to
have multiple, long-lasting effects: not only did it impact a
biomarker of stress (telomere attrition; Nettle, Monaghan, et al.,
2015) and ight performance (O'Hagan et al., 2015), it also
induced increased motivation to forage for food and gather infor-
mation about sources of food and, perhaps consequently, to carry
more body fat. Thus, our ndings add to previous evidence in birds
and other taxa (Bateson, Brilot, Gillespie, Monaghan, & Nettle,
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132 129
2015; Bloxham et al., 2014; Boogert et al., 2014; Boogert, Zimmer, &
Spencer, 2013; Criscuolo, Monaghan, Nasir, & Metcalfe, 2008; Felitti
et al., 1998; Lindstr
om, 1999; Nettle, Monaghan, et al., 2015;
Nowicki, Searcy, & Peters, 2002; O'Hagan et al., 2015; Tschirren,
Rutstein, Postma, Mariette, & Grifth, 2009; Verhulst et al., 200 6;
Zimmer, Boogert, & Spencer, 2013) that even subtle early life ma-
nipulations can induce enduring alterations in physical and
behavioural phenotypes. The question why such lasting inuences
of early environment endure into adulthood even when current
energetic or environmental conditions apparently no longer differ
has been the topic of much theoretical debate (Dall et al., 2012;
Frankenhuis & Panchanathan, 2011). The effects in adulthood
may be a nonadaptive consequence of behaviour that was advan-
tageous during the nestling or immediate postedging period
when selection is most intense. Instead, we may plausibly interpret
the altered phenotype as remaining benecial in adulthood for
birds that experienced a poor start in life, perhaps due to reduced
certainty over the future availability of, or access to, foraging
patches. To this end, the social dominance of developmentally
disadvantaged birds and their relative foraging success in unpre-
dictable food environments remains to be examined. Understand-
ing the underlying causes of individual variation in foraging
behaviour and resource or information acquisition may help us to
elucidate behavioural variation typically thought of as personality
variation (Dall et al., 2012) and could improve our understanding of
the ontogeny of food-related conditions such as obesity in humans
(Cottrell & Ozanne, 2008).
Acknowledgments
This research was funded by the Biotechnology and Biological
Sciences Research Council under grant BB/J016446/1. We thank
Michelle Waddle for support in looking af ter the birds, and three
anonymous referees for helping us improve the manuscript. We
also thank the farmers whose farms house our starling colonies,
and Kirkley Hall Zoo for giving the birds a permanent home at
retirement.
Supplementary Material
Supplementary material associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.anbehav.
2015.08.009.
References
Adams, E. J., Grummer-Strawn, L., & Chavez, G. (2003). Community and interna-
tional nutrition food insecurity is associated with increased risk of obesity in
California. Journal of Nutrition, 133(August 2002), 1070e1074.
Andrews, C., Viviani, J., Egan, E., Bedford, T., Brilot B., Nettle, D., et al. (n.d.). [Data set
of chick weights of manipulated and unmanipulated broods of starlings]. Un-
published raw data.
Bateson, M., Brilot, B. O., Gillespie, R., Monaghan, P., & Nettle, D. (2015). Develop-
mental telomere attrition predicts impulsive decision-making in adult starlings.
Proceedings of the Royal Society B: Biological Sciences, 282, 20142140.
Bean, D., Mason, G. J., & Bateson, M. (1999). Contrafreeloading in starlings: testing
the information hypothesis. Behaviour, 136(10), 1267e1282. http://dx.doi.org/
10.1163/156853999500712.
Bloxham, L., Bateson, M., Bedford, T., & Nettle, D. (2014). The memory of hunger:
developmental plasticity of dietary selectivity in the European starling, Sturnus
vulgaris. Animal Behaviour, 91,33e40.
Boogert, N. J., Farine, D. R., & Spencer, K. A. (2014). Developmental stress predicts
social network position. Biology Letters, 10, 20140561.
Boogert, N. J., Zimmer, C., & Spencer, K. A. (2013). Pre- and post-natal stress have
opposing effects on social information use. Biology Letters, 9(2), 20121088.
http://dx.doi.org/10.1098/rsbl.2012.1088.
Brust, V., Krüger, O., Naguib, M., & Krause, E. T. (2014). Lifelong consequences of
early nutritional conditions on learning performance in zebra nches (Taenio-
pygia guttata). Behavioural Processes, 320e326. http://dx.doi.org/10.1016/
j.beproc.2014.01.019.
Coburn, J. F., & Tarte, R. D. (1976). The effect of rearing environments on the con-
trafreeloading phenomenon in rats. Journal of the Experimental Analysis of
Behavior, 26(2), 289e294.
Cotton, P. A., Wright, J., & Kacelnik, A. (1999). Chick begging strategies in relation to
brood hierarchies and hatching asynchrony. The American Naturalist, 153(4),
412e420.
Cottrell, E. C., & Ozanne, S. E. (20 08). Early life programming of obesity and meta-
bolic disease. Physiology and Behavior, 94,17e28. http://dx.doi.org/10.1016/
j.physbeh.2007.11.017.
Coulton, L. E., Waran, N. K., & Young, R. J. (1997). Effects of foraging enrichment on
the behaviour of parrots. Animal Welfare e Potters Bar, 6,357e363.
Coup
e, B., Grit, I., Darmaun, D., & Parnet, P. (2009). The timing of catch-up growth af-
fects metabolism andappetite regulation in male ratsbornwith intrauterinegrowth
restriction. American Journal of Physiology. Regulatory, Integrative and Comparative
Physiology, 297(3), R813eR824. http://dx.doi.org/1 0.11 52/ajpregu.00201.2009.
Criscuolo, F., Monaghan, P., Nasir, L., & Metcalfe, N. B. (2008). Early nutrition and
phenotypic development:catch-upgrowth leads to elevated metabolic rate in
adulthood. Proceedings of the Royal Society B: Biological Sciences, 27(1642),
1565e1570. http://dx.doi.org/10.1098/rspb.2008.0148.
Dall, S. R. X., Bell, A. M., Bolnick, D. I., & Ratnieks, F. L. W. (2012). An evolutionary
ecology of individual differences. Ecology Letters, 15(10), 1189e1198. http://
dx.doi.org/10.1111/j.1461-0248.2012.01846.x.
Dall, S. R. X., & Johnstone, R. A. (2002). Managing uncertainty: information and
insurance under the risk of starvation. Philosophical Transactions of the Royal
Society of London. Series B, Biological Sciences, 357,1519e1526. http://dx.doi.org/
10.1098/rstb.2002.1061.
Dunlap, A. S., & Stephens, D. W. (2012). Tracking a changing environment: optimal
sampling, adaptive memory and overnight effects. Behavioural Processes, 89(2),
86e94. http://dx.doi.org/10.1016/j.beproc.2011.10.005.
D'Eath, R. B., Tolkamp, B. J., Kyriazakis, I., & Lawrence, A. B. (2009). Freedom from
hunger and preventing obesity: the animal welfare implications of reducing
food quantity or quality. Animal Behaviour, 77(2), 275e288. http://dx.doi.org/
10.1016/j.anbehav.2008.10.028.
Feare, C. (1984). The starling. Oxford, U.K.: Oxford University Press.
Feenders, G., & Bateson, M. (2011). Hand-rearing reduces fear of humans in Euro-
pean starlings, Sturnus vulgaris. PLoS One, 6(2), e17466. http://dx.doi.org/
10.1371/journal.pone.0017466.
Feenders, G., & Bateson, M. (2013). Hand rearing affects emotional responses but
not basic cognitive performance in European starlings. Animal Behaviour, 86(1),
127e138. http://dx.doi.org/10.1016/j.anbehav.2013.05.002.
Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V.,
et al. (1998). Relationship of childhood abuse and household dysfunction to
many of the leading causes of death in adults: the adverse childhood experi-
ences (ACE) study. American Journal of Preventive Medicine, 14(4), 245e258.
http://dx.doi.org/10.1016/S0749-3797(98)00017-8.
Forkman, B. (1991). Some problems with current patch-choice theory: a study on
the Mongolian gerbil. Behaviour, 117(3), 243e
254. http://dx.doi.org/10.1163/
156853991X00553.
Forkman, B. (1996). The foraging behaviour of Mongolian gerbils: a behavioural
need or a need to know? Behaviour, 133(1), 129e143.
Frankenhuis, W. E., & Panchanathan, K. (2011). Balancing sampling and specializa-
tion: an adaptationist model of incremental development. Proceedings of the
Royal Society B: Biological Sciences, 278(April), 3558e3565. http://dx.doi.org/
10.1098/rspb.2011.0055.
Garner, D. M. (1997). Psychoeducational principles in treatment. In D. M. Garner, &
P. E. Garnkel (Eds.), Handbook of treatment for eating disorders (pp. 145e177).
New York, NY: Guilford Press.
Gluckman, P. D., & Hanson, M. A. (2007). Developmental plasticity and human
disease: research directions. Journal of Internal Medicine, 261,461e471 . http://
dx.doi.org/10.1111/j.1365-2796.2007.01802.x.
Houston, A. I., & McNamara, J. M. (1993). A theoretical investigation of the fat re-
serves and mortality levels of small birds in winter. Ornis Scandinavica, 24(3),
205e219. http://dx.doi.org/10.2307/3676736.
Inglis, I. R., & Ferguson, N. J. K. (1985). Starlings search for food rather than eat
freely-available, identical food. Animal Behaviour, 32(2), 614e617.
Inglis, I. R., Forkman, B., & Lazarus, J. (1997). Free food or earned food? A review and
fuzzy model of contrafreeloading. Animal Behaviour, 53,1171e1191.
Inglis, I. R., Langton, S., Forkman, B., & Lazarus, J. (2001). An information primacy
model of exploratory and foraging behaviour. Animal Behaviour, 62(3), 543e557.
http://dx.doi.org/10.1006/anbe.2001.1780.
Kacelnik, A., Cotton, P. A., Stirling, L., & Wright, J. (1995). Food allocation among
nestling starlings: sibling competition and the scope of parental choice. Pro-
ceedings of the Royal Society B: Biological Sciences, 259(1356), 259e263.
Keasar, T., Motro, U., & Shmida, A. (2013). Temporal reward variability promotes
sampling of a new ower type by bumblebees. Animal Behaviour, 86(4),
747e753. http://dx.doi.org/10.1016/j.anbehav.2013.07.010.
Khlat, M., Jusot, F., & Ville, I. (20 09). Social origins, early hardship and obesity: a
strong association in women, but not in men? Social Science and Medicine, 68(9),
1692e1699. http://dx.doi.org/10.1016/j.socscimed.2009.02.024.
Labocha, M. K., & Hayes, J. P. (2012). Morphometric indices of body condition in
birds: a review.
Journal of Ornithology, 153(1), 1e22. http://dx.doi.org/10.1007/
s10336-011-0706-1.
Larson, L. D., & Tarte, R. D. (2013). The effects of training and effortfulness on rats'
choice behavior in a modied T-maze. Bulletin of the Psychonomic Society, 7(6),
506e508. http://dx.doi.org/10.3758/BF03337265.
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132130
Levin, B. E. (2006). Metabolic imprinting: critical impact of the perinatal environ-
ment on the regulation of energy homeostasis. Philosophical Transactions of the
Royal Society of London. Series B, Biological Sciences, 361(June), 1107e1121 . http://
dx.doi.org/10.1098/rstb.2006.1851.
Lindqvist, C. E. S., & Jensen, P. (2008). Effects of age, sex and social isolation on
contrafreeloading in red junglefowl (Gallus gallus ) and White Leghorn fowl.
Applied Animal Behaviour Science, 114,419e428. http://dx.doi.org/10.1016/
j.applanim.2008.03.002.
Lindqvist, C. E. S., Schütz, K. E., & Jensen, P. (2002). Red jungle fowl have more
contrafreeloading than white leghorn layers: effect of food deprivation and
consequences for information gain. Behaviour, 139(9), 1195e1209.
Lindqvist, C. E. S., Zimmerman, P., & Jensen, P. (2006). A note on contrafreeloading in
broilers compared to layer chicks. Applied Animal Behaviour Science, 101(1e2),
161e166. http://dx.doi.org/10.1016/j.applanim.2006.01.006.
Lindstr
om, J. (1999). Early development and tness in birds and mammals. Trends in
Ecology & Evolution, 14(9), 343e348. http://dx.doi.org/10.1016/S0169-5347(99)
01639-0.
Lummaa, V., & Clutton-Brock, T. (2002). Early development, survival and repro-
duction in humans. Trends in Ecology & Evolution, 17(3), 141e147. http://
dx.doi.org/10.1016/s0169-5347(01)02414-4.
Macleod, R., Gosler, A. G., & Cresswell, W. (2005). Diurnal mass gain strategies and
perceived predation risk in the great tit Parus major. Journal of Animal Ecology,
74(5), 956e964. http://dx.doi.org/10.1111/j.1365-2656.2005.00993.x .
Martin-Fernandez, J., Caillavet, F., Lhuissier, A., & Chauvin, P. (2014). Food insecurity,
a determinant of obesity?eAn analysis from a population-based survey in the
paris metropolitan area, 2010. Obesity Facts, 7,120e129. http://dx.doi.org/
10.1159/000362343.
Mathot, K. J., & Dall, S. R. X. (2013). Metabolic rates can drive individual differences
in information and insurance use under the risk of starvation. The American
Naturalist, 182(5), 611e620. http://dx.doi.org/10.1086/673300.
Nettle, D., Andrews, C. P., Monaghan, P., Brilot, B. O., Bedford, T., Gillespie, R., et al.
(2015). Developmental and familial predictors of adult cognitive traits in the
European starling.
Animal Behaviour, 107, 239e248.
Nettle, D., Monaghan, P., Gillespie, R., Brilot, B., Bedford, T., & Bateson, M. (2015). An
experimental demonstration that early-life competitive disadvantage acceler-
ates telomere loss. Proceedings of the Royal Society B: Biological Sciences,
282(1798), 20141610. http://dx.doi.org/10.1098/rspb.2014.1610.
Neuringer, A. J. (1969). Animals respond for food in the presence of free food. Sci-
ence, 166(i), 399e401.
Nowicki, S., Searcy, W. A., & Peters, S. (2002). Brain development, song learning and
mate choice in birds: a review and experimental test of the nutritional stress
hypothesis. Journal of Comparative Physiology. A, Neuroethology, Sensory, Neural,
and Behavioral Physiology, 188(11e12), 1003e1014. http://dx.doi.org/10.1007/
s00359-002-0361-3.
Ogura, T. (2011). Contrafreeloading and the value of control over visual stimuli in
Japanese macaques (Macaca fuscata). Animal Cognition, 14(3), 427e431. http://
dx.doi.org/10.1007/s10071-010-0377-y.
Orozco-S
olis, R., Lopes de Souza, S., Barbosa Matos, R. J., Grit, I., Le Bloch, J.,
Nguyen, P., et al. (2009). Perinatal undernutrition-induced obesity is indepen-
dent of the developmental programming of feeding. Physiology & Behavior,
96(3), 481e492. http://dx.doi.org/10.1016/j.physbeh.2008.11.016.
O'Hagan, D., Andrews, C. P., Bedford, T., Bateson, M., & Nettle, D. (2015). Early life
disadvantage strengthens ight performance trade-offs in European starlings,
Sturnus vulgaris. Animal Behaviour, 102, 141e148. http://dx.doi.org/10.1016/
j.anbehav.2015.01.016.
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & R Core Team. (2015). R Package nlme.
Retrieved from http://cran.r-project.org/package¼nlme.
Qasem, R. J., Yablonski, E., Li, J., Tang, H. M., Pontiggia, L., & D'mello, A. P. (2012).
Elucidation of thrifty features in adult rats exposed to protein restriction during
gestation and lactation. Physiology & Behavior, 105(5), 1182e1193. http://
dx.doi.org/10.1016/j.physbeh.2011.12.010.
R Development Core Team. (2011). R: A language and environment for statistical
computing. Vienna, Austria: R Foundation for Statistical Computing.
Redman, L. M., & Ravussin, E. (2011). Caloric restriction in humans: impact on
physiological, psychological, and behavioral outcomes. Antioxidants & Redox
Signaling, 14(2), 275e287. http://dx.doi.org/10.1089/ars.2010.3253.
Rich, E. L., & Romero, L. M. (2005). Exposure to chronic stress downregulates
corticosterone responses to acute stressors. American Journal of Physiology.
Regulatory, Integrative and Comparative Physiology, 288(6), R1628eR1636. http://
dx.doi.org/10.1152/ajpregu.00484.2004.
Robertson, L. C., & Anderson, S. C. (1975). The effects of differing type and magni-
tude of reward on the contrafreeloading phenomenon in rats. Animal Learning
& Behaviour, 3(4), 325e328.
Rutter, S., & Nevin, J. A. (1990). Long-term contrafreeloading in rats during
continuous sessions. Bulletin of the Psychonomic Society, 28(6), 556e558. http://
dx.doi.org/10.3758/BF03334080.
Schütz, K. E., & Jensen, P. (2001). Effects of resource allocation on behavioural
strategies: a comparison of red junglefowl (Gallus gallus ) and two domesticated
breeds of poultry. Ethology, 765, 753e765.
Shettleworth, S. J., Krebs, J. R., Stephens, D. W., & Gibbon, J. (1988). Tracking a
uctuating environment: a study of sampling. Animal Behaviour, 36(1), 87e105.
http://dx.doi.org/10.1016/S0003-3472(88)80252-5.
Spencer, K. A., & Verhulst, S. (2008). Post-natal exposure to corticosterone affects
standard metabolic rate in the zebra nch (Taeniopygia guttata). General and
Comparative Endocrinology, 159(2e3), 250e256. http://dx.doi.org/10.1016/
j.ygcen.2008.09.007.
Stephens, D. W., & Krebs, J. R. (1986). Foraging theory. Chichester, U.K.: Princeton
University Press.
Townsend, M. S., Peerson, J., Love, B., Achterberg, C., & Murphy, S. P. (2001). Food
insecurity is positively related to overweight in women. The Journal of Nutrition,
131,1738e1745.
Tschirren, B., Rutstein, A., Postma, E., Mariette, M., & Grifth, S. C. (2009). Short- and
long-term consequences of early developmental conditions: a case study on
wild and domesticated zebra nches. Journal of Evolutionary Biology, 22(2),
387e395. http://dx.doi.org/10.1111/j.1420-9101.2008.01656.x.
Vasconcellos, S., Harumi, C., & Ades, C. (2012). Contrafreeloading in maned wolves:
implications for their management and welfare. Applied Animal Behaviour Sci-
ence, 140(1e2), 85e91 . http://dx.doi.org/10.1016/j.applanim.2012.04.012.
Verhulst, S., Holveck, M.-J., & Riebel, K. (2006). Long-term effects of manipulated
natal brood size on metabolic rate in zebra nches. Biology Letters, 2,478e480.
http://dx.doi.org/10.1098/rsbl.2006.0496.
Vickers, M. H., Breier, B. H., Cuteld, W. S., Hofman, P. L., Gluckman, P. D., Mark, H.,
et al. (2000). Fetal origins of hyperphagia, obesity, and hypertension and
postnatal amplication by hypercaloric nutrition. American Journal of Physi-
ology. Endocrinology and Metabolism, 279, E83eE87.
Vitousek, K. M., Manke, F. P., Gray, J. A., & Vitousek, M. N. (2004). Caloric restriction
for longevity: IIeThe systematic neglect of behavioural and psychological out-
comes in animal research. European Eating Disorders Review, 12, 338e360.
http://dx.doi.org/10.1002/erv.604.
Witter, M. S., & Swaddle, J. P. (1995). Dominance, competition, and energetic re-
serves in the European starling, Sturnus vulgaris. Behavioral Ecology, 6(3),
343e348. http://dx.doi.org/10.1093/beheco/6.3.343.
Wright, J., Hinde, C., Fazey, I., Both, C., Fazey, L., & Both, C. (2002). Begging signals
more than just short-term need: cryptic effects of brood size in the pied
ycatcher (Ficedula hypoleuca). Behavioural Ecology and Sociobiology, 52(1),
74e83. http://dx.doi.org/10.1007/s00265-002-0478-y.
Zimmer, C., Boogert, N. J., & Spencer, K. A. (2013). Developmental programming:
cumulative effects of increased pre-hatching corticosterone levels and post-
hatching unpredictable food availability on physiology and behaviour in
adulthood. Hormones and Behavior, 64(3), 494e500. http://dx.doi.org/10.1016/
j.yhbeh.2013.07.002.
Appendix
Table A1
Output of statistical models
Model no. Response variable Random effects Fixed effects LRT P
Mass and body condition
1 Body mass Natal nest Sex
f
11.62 0.001
Treatment
f
1.77 0.183
Sex)Treatment 0.16 0.685
2 Body condition index Natal nest Sex
f
5.32 0.021
Treatment
f
6.83 0.009
Sex)Treatment 0.02 0.893
Food consumption
3 Total food consumption Natal nest/Bird Sex
f
8.73 0.003
Treatment
f
2.98 0.085
Sex)Treatment 0.00 0.9500
(continued on next page)
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132 131
Table A1 (continued )
Model no. Response variable Random effects Fixed effects LRT P
4 Trial consumption Natal nest/Bird Sex
f
7.18 0.007
Treatment
f
6.27 0.012
Sex)Treatment 0.01 0.927
5 log(CFL consumption) Natal nest/Bird Sex
f
7.11 0.008
Treatment
f
3.79 0.052
Sex)Treatment 0.39 0.531
Free food consumption
c
61.07 <0.001
Free food consumption)Treatment 2.12 0.147
6 Free food consumption Natal nest/Bird Sex
f
6.66 0.010
Treatment
f
5.79 0.016
Sex)Treatment 0.00 0.948
CFL consumption
c
62.18 <0.001
CFL consumption)Treatment 3.10 0.078
CFL effort
7 CFL duration Natal nest/Bird Sex
f
0.31 0.577
Treatment
f
3.45 0.063
Sex)Treatment 0.43 0.510
Free feeding duration
c
26.97 <0.001
Free feeding duration)Treatment 6.16 0.013
Foraging efciency
8 Foraging rate Natal nest/Bird Sex
f
4.60 0.032
Treatment
f
0.04 0.838
Sex)Treatment 0.00 0.998
Bowl
f
46.33 <0.001
Bowl)Treatment 5.16 0.023
Trade-off between CFL and foraging success
9 Trial consumption Natal nest/Bird Sex 7.06 0.008
Treatment 5.47 0.019
Sex)Treatment 0.00 0.962
CFL duration
c
1.42 0.234
CFL duration)Treatment 6.00 0.014
10 Nontrial food consumption Natal nest/Bird Sex 8.64 0.003
Treatment 1.29 0.256
Sex)Treatment 0.01 0.915
CFL duration
c
1.53 0.216
CFL duration)Treatment 0.94 0.332
This table provides the output from the statistical models described in the Results. All models are general linear mixed with Gaussian error structure, run using package nlme
in R. LRT is the likelihood ratio test statistic comparing models with and without the xed effect of interest (see Statistical Analysis); P is the P value for this likelihood ratio test.
All xed and random effects included in the model are shown. All models included a random nest-of-origin effect (Natal nest); where the data contained multiple mea-
surements from individual birds a random effect of individual (Bird) nested in Natal nest was also included. Body mass is the mean of cage entry body mass and experiment
end body mass. Body condition is the residual mass corrected for tarsus length (see Statistical Analysis for details). Total food consumption is the daily summed mass of food (g)
consumed outside the experimental trial (nontrial consumption) and during the trial (trial consumption) from both free food and CFL bowls. CFL consumption and free food
consumption is the mass of food (g) consumed from the CFL bowl or free food bowl, respectively during a daily trial. Trial consumption is the sum of CFL consumption and free
food consumption. CFL duration and free feeding duration are the daily times (s) spent in, perched on or with head over the CFL or free food bowl, respectively. Foraging rate is
the food consumption per unit duration (g/s) i.e. CFL consumption/CFL duration or free food consumption/free feeding duration. The xed factor treatment refers to whether a
bird experienced the ADV or DIS developmental manipulation of early life competition as a nestling. The xed factor Bowl (Model 7) refers to foraging carried out in the free
food or CFL bowl.
f
Denotes a xed factor;
c
denotes a covariate.
C. Andrews et al. / Animal Behaviour 109 (2015) 123e132132
... Open Sci. 9: 220102 birds [65]. Similarly, rats show greater sensitivity to resource allocation in predictable environments compared with unpredictable environments [66]. ...
Article
Full-text available
We report the results of a pre-registered analysis of data from the English Longitudinal Study of Ageing that was designed to test the hypothesis that economic scarcity is associated with individual differences in decision-making. We tested this hypothesis by comparing time preferences for different socio-economic groups and in geographical areas ranging from the most deprived to the least deprived in England using the English indices of multiple deprivation. The data supported this hypothesis: people in the most deprived areas were more likely to prefer smaller-sooner rewards than people from the least deprived areas. Similarly, people in technical or routine occupations tended to prefer smaller-sooner rewards than people in professional or intermediate occupations. In addition, we found that gender, cognitive function and subjective social status also predicted time preferences. We discuss these results in the context of theoretical models of scarcity-based models of choice behaviour and decision-making.
... The effect of early childhood experience reported by Griskevicius et al. is a very plausible source of individual differences in time preferences. There is abundant evidence that impulsivity has a large environmental component in its heritability (Bezdjian et al., 2011) and evidence from behavioural ecology suggest that scarcity is a plausible mechanism that drives impulsivity in other species (Andrews et al., 2015). Similarly, mortality salience is a plausible motivating factor in decisions that have long time horizons, although one could develop an argument consistent with either an increase or a decrease in discount rates. ...
Article
Full-text available
We outline three attempts to replicate experiments that reported priming effects on time preferences measured by delay discounting. Experiment 1 tested the claim that images of poverty prime impulsive choice in people from less affluent backgrounds compared to people from more affluent backgrounds. Experiment 2 tested the claim that mortality salience – thinking about death – primes people to place more value on the future than people who thought about dental surgery. Experiment 3 tested the claim that an episodic foresight manipulation primes greater discounting than no episodic foresight. Experiments 1 and 2 failed to replicate the effects of priming on discount rates. Experiment 3 was a successful and very close replication of the effect of episodic foresight on discount rates.
... Intriguingly, the association between social position and body fatness is not limited to humans. For example, subordinate female rhesus monkeys had higher caloric consumption than dominant monkeys (Wilson et al., 2008), and starlings with a disadvantageous early life environment were fatter in adulthood and spent more effort for food gathering than those with more advantaged early life environments (Andrews et al., 2015). Even when it would be attractive to consider that these parallel results in different species would suggest deep neurophysiological pathways, it is noteworthy that the inverse association between social position and BMI may not be universal, even in industrialized societies. ...
Article
Obesity has dramatically increased during the last decades and is currently one of the most serious global health problems. We present a hypothesis that obesity is a neuro-behavioral disease having a strong genetic background mediated largely by eating behavior and is sensitive to the macro-environment; we study this hypothesis from the perspective of genetic research. Genetic family and genome-wide-association studies have shown well that body mass index (BMI, kg/m2) is a highly heritable and polygenic trait. New genetic variation of BMI emerges after early childhood. Candidate genes of BMI notably express in brain tissue, supporting that this new variation is related to behavior. Obesogenic environments at both childhood family and societal levels reinforce the genetic susceptibility to obesity. Genetic factors have a clear influence on macro-nutrient intake and appetite-related eating behavior traits. Results on the gene-by-diet interactions in obesity are mixed, but emerging evidence suggests that eating behavior traits partly mediate the effect of genes on BMI. However, more rigorous prospective study designs controlling for measurement bias are still needed.
... Subordinate female rhesus monkeys had higher caloric consumption than dominant monkeys (42). In addition, starlings with a disadvantaged early life environment were fatter in adulthood and spent more effort for food gathering than those with more advantaged early life environments (43). Although comparisons between species should be treated with caution, these results may suggest deep neurophysiological pathways between early life social position and later eating behavior. ...
Article
Objective The objective of this study was to analyze how parental education modifies the genetic and environmental variances of BMI from infancy to old age in three geographic‐cultural regions. Methods A pooled sample of 29 cohorts including 143,499 twin individuals with information on parental education and BMI from age 1 to 79 years (299,201 BMI measures) was analyzed by genetic twin modeling. Results Until 4 years of age, parental education was not consistently associated with BMI. Thereafter, higher parental education level was associated with lower BMI in males and females. Total and additive genetic variances of BMI were smaller in the offspring of highly educated parents than in those whose parents had low education levels. Especially in North American and Australian children, environmental factors shared by co‐twins also contributed to the higher BMI variation in the low education level category. In Europe and East Asia, the associations of parental education with mean BMI and BMI variance were weaker than in North America and Australia. Conclusions Lower parental education level is associated with higher mean BMI and larger genetic variance of BMI after early childhood, especially in the obesogenic macro‐environment. The interplay among genetic predisposition, childhood social environment, and macro‐social context is important for socioeconomic differences in BMI.
... There is a clear body of evidence linking differences in earlylife developmental conditions to the social behaviour, and resulting social structure [21,28,86,87]. Developmental history could be a key factor underpinning consistent differences in individual behaviour [88][89][90][91][92][93]. For example, differences in brood size during early life can generate effects that are carried over into the following generations [94]. ...
Article
Early-life experience can fundamentally shape individual life-history trajectories. Previous research has suggested that exposure to stress during development causes differences in social behaviour later in life. In captivity, juvenile zebra finches exposed to elevated corticosterone levels were less socially choosy and more central in their social networks when compared to untreated siblings. These differences extended to other aspects of social life, with ‘stress-exposed’ juveniles switching social learning strategies and juvenile males less faithfully learning their father's song. However, while this body of research suggests that the impacts of early-life stress could be profound, it remains unknown whether such effects are strong enough to be expressed under natural conditions. Here, we collected data on social associations of zebra finches in the Australian desert after experimentally manipulating brood sizes. Juveniles from enlarged broods experienced heightened sibling competition, and we predicted that they would express similar patterns of social associations to stress-treated birds in the captive study by having more, but less differentiated, relationships. We show striking support for the suggested consequences of developmental stress on social network positions, with our data from the wild replicating the same results in 9 out of 10 predictions previously tested in captivity. Chicks raised in enlarged broods foraged with greater numbers of conspecifics but were less ‘choosy’ and more central in the social network. Our results confirm that the natural range of variation in early-life experience can be sufficient to predict individuals' social trajectories and support theory highlighting the potential importance of developmental conditions on behaviour.
... However, a widespread conclusion in developmental programming research is that small changes to the nature, timing, or severity of developmental treatments can produce different or even opposite effects on the adult phenotype (Love and Williams 2008;Kriengwatana et al. 2014). In our own work, for example, we have observed different types of early-life adversity producing both lower and higher long-term adult body weights (Andrews et al. 2015;Dunn et al. 2018). Thus, it may be that our hand-rearing paradigm, rather than being a more controlled model of a brood-size manipulation in the wild, actually has qualitatively different effects. ...
Article
Full-text available
Judgement bias tasks are designed to provide markers of affective states. A recent study of European starlings (Sturnus vulgaris) demonstrated modest familial effects on judgement bias performance, and found that adverse early experience and developmental telomere attrition (an integrative marker of biological age) both affected judgement bias. Other research has shown that corticosterone levels affect judgement bias. Here, we investigated judgement bias using a modified Go/No Go task in a new cohort of starlings (n = 31) hand-reared under different early-life conditions. We also measured baseline corticosterone and the corticosterone response to acute stress in the same individuals. We found evidence for familial effects on judgement bias, of a similar magnitude to the previous study. We found no evidence that developmental treatments or developmental telomere attrition were related to judgement bias per se. We did, however, find that birds that experienced the most benign developmental conditions, and birds with the greatest developmental telomere attrition, were significantly faster to probe the learned unrewarded stimulus. We also found that the birds whose corticosterone levels were faster to return towards baseline after an acute stressor were slower to probe the learned unrewarded stimulus. Our results illustrate the potential complexities of relationships between early-life experience, stress and affectively mediated decision making. For judgement bias tasks, they demonstrate the importance of clearly distinguishing factors that affect patterns of responding to the learned stimuli (i.e. response inhibition in the case of the Go/No Go design) from factors that influence judgements under ambiguity.
... However, a widespread conclusion in developmental programming research is that small changes to the nature, timing, or severity of developmental treatments can produce different or even opposite effects on the adult phenotype (Love and Williams 2008;Kriengwatana et al. 2014). In our own work, for example, we have observed different types of early-life adversity producing both lower and higher long-term adult body weights (Andrews et al. 2015;Dunn et al. 2018). Thus, it may be that our hand-rearing paradigm, rather than being a more controlled model of a brood-size manipulation in the wild, actually has qualitatively different effects. ...
Article
Full-text available
Judgement bias tasks are designed to provide markers of affective states. A recent study of European starlings (Sturnus vulgaris) demonstrated modest familial effects on judgement bias performance, and found that adverse early experience and developmental telomere attrition (an integrative marker of biological age) both affected judgement bias. Other research has shown that corticosterone levels affect judgement bias. Here, we investigated judgement bias using a modified Go/No Go task in a new cohort of starlings (n = 31) hand-reared under different early-life conditions. We also measured baseline corticosterone and the corticosterone response to acute stress in the same individuals. We found evidence for familial effects on judgement bias, of a similar magnitude to the previous study. We found no evidence that developmental treatments or developmental telomere attrition were related to judgement bias per se. We did, however, find that birds that experienced the most benign developmental conditions, and birds with the greatest developmental telomere attrition, were significantly faster to probe the learned unrewarded stimulus. We also found that the birds whose corticosterone levels were faster to return towards baseline after an acute stressor were slower to probe the learned unrewarded stimulus. Our results illustrate the potential complexities of relationships between early-life experience, stress and affectively mediated decision making. For judgement bias tasks, they demonstrate the importance of clearly distinguishing factors that affect patterns of responding to the learned stimuli (i.e. response inhibition in the case of the Go/No Go design) from factors that influence judgements under ambiguity. Electronic supplementary material The online version of this article (10.1007/s10071-018-1226-7) contains supplementary material, which is available to authorized users.
Article
Criteria for assessing the severity of scientific procedures in laboratory rodents include the loss of body mass. However, guidance is limited for passerine birds and application of criteria developed for mammals risks poor welfare decisions. Here, I ask whether, and how, body mass criteria could be incorporated into laboratory welfare assessment of passerines. Passerine birds strategically adjust their body mass to minimise combined mortality risk from starvation and predation. A systematic literature review found that strategic mass changes can be sizeable (sometimes > 10%) even over short timescales. Many aspects of a bird's current or past environment, including husbandry and experimental procedures, may alter perceived starvation or predation risks and thus drive strategic mass change via evolved mechanisms. Therefore, body mass criteria used for rodents may be too stringent for passerines, potentially leading to over-estimated severity. Strategic mass changes might obscure those stemming from experimental interventions yet could also offer insights into whether birds perceive an intervention or altered husbandry as a threat. Mass criteria for severity assessment should be species- and context-specific in order to balance needs for refinement and reduction. To guide the development of appropriate criteria, a future research priority is for greater data collection and sharing based on standardised routine monitoring of mass variation under a representative range of husbandry conditions and procedures.
Preprint
Full-text available
We report the results of a pre-registered analysis of data from the English Longitudinal Study of Aging that was designed to test the hypothesis that economic scarcity is associated with individual differences in decision-making. We tested this hypothesis by comparing time preferences for different socio-economic groups and in geographical areas ranging from the most deprived to the least deprived in England using the English Indices of Multiple Deprivation. The data supported this hypothesis: People in the most deprived areas were more likely to prefer smaller sooner rewards than people from the least deprived areas. Similarly, people in technical or routine occupations tended to prefer smaller sooner rewards than people in professional or intermediate occupations. In addition, we found that gender, length of education, cognitive function, and subjective social status also predicted time preferences. We discuss these results in the context of theoretical models of scarcity-based models of choice behaviour and decision-making.
Article
Full-text available
Social epigenomics is a new field of research that studies how the social environment shapes the epigenome and how in turn the epigenome modulates behavior. We focus on describing known gene-environment interactions (GEIs) and epigenetic mechanisms in different mammalian social systems. To illustrate how epigenetic mechanisms integrate GEIs, we highlight examples where epigenetic mechanisms are associated with social behaviors and with their maintenance through neuroendocrine, locomotor and metabolic responses. We discuss future research trajectories and open questions for the emerging field of social epigenomics in non-model and naturally occurring social systems. Finally, we outline the technological advances that aid the study of epigenetic mechanisms in the establishment of GEIs and vice versa.
Article
Full-text available
In birds, there is evidence that adult cognitive traits can both run in families and be affected by early developmental influences. However, different studies use different cognitive tasks, which may not be measuring the same traits, and also focus on different developmental factors. We report results from a study in which we administered multiple cognitive tasks (autoshaping, discrimination learning, reversal learning, progressive ratio schedule, extinction learning and impulsivity) to a cohort of 34 European starlings, Sturnus vulgaris, for which several early developmental measures were available. The cohort consisted of siblings raised either apart or together, whose position in the size hierarchy of the rearing brood had been experimentally manipulated. We examined how the different cognitive measures covaried, the extent to which they ran in families, and which of the developmental factors predicted which of the cognitive outcomes. We found that discrimination and reversal learning speeds were positively correlated, as were breakpoint on the progressive ratio schedule and resistance to extinction. Otherwise, the cognitive measures were uncorrelated, suggesting that they reflected different underlying traits. All traits except discrimination and reversal learning speed ran in families to a substantial extent. Using a model selection approach, we found evidence that natal brood size and developmental telomere attrition (the extent to which the birds' erythrocyte telomeres shortened in early life, an integrative measure of developmental stress) were related to several adult cognitive measures. Results are discussed with respect to the best way of measuring avian cognitive abilities, and the utility of developmental telomere attrition as a predictor of adult outcomes.
Article
Full-text available
Developmental stress has been shown to affect adult flight performance in birds, with both negative and positive effects reported in the literature. Previous studies have used developmental manipulations that had substantial effects on patterns of growth. They have also examined mean levels of flight performance per individual, rather than investigating how developmental stress might alter trade-offs between different components of flight performance. We recorded multiple components of escape flight performance in 20 adult European starlings previously subjected to a manipulation likely to have altered levels of developmental stress. Siblings had been cross-fostered to nests where they were either slightly larger (advantaged treatment) or slightly smaller (disadvantaged treatment) than their competitors. The manipulation had no detectable effect on growth. However, developmental treatment affected performance in escape flights a year later by strengthening the trade-offs between different flight parameters. Disadvantaged birds faced a steeper trade-off between take-off speed and take-off angle, and a steeper trade-off between take-off angle and total time in flight, than advantaged birds. The results suggest that even subtle early life adversity that has no obvious effect on growth or size can leave a lasting legacy in the form of constraints on locomotor performance later in life.
Article
Full-text available
Foraging bumblebees focus mainly on one flower species (their 'major'), and occasionally visit other flowers ('minors') that are less rewarding. Minoring allows tracking of potential alternative resources that might become more profitable with time. This is expected to provide an adaptive benefit under unstable foraging conditions, but to reduce foraging efficiency when resources are steady. We tested how predictability of food sources affects minoring and foraging success, using Bombus terrestris workers that fed on sucrose solution in artificial flowers. Bees exposed to three colour-distinct flower types with fluctuating food rewards performed more minoring (visits to a fourth, nonrewarding flower type) than bees that encountered a stable reward schedule. This difference was mostly due to a higher frequency of flower type shifts in the fluctuating-reward treatment. Flowers of the highest-rewarding type were visited less frequently in the fluctuating-reward than in the stable-reward treatment. This suggests that the fluctuating-reward schedule limited the bees' ability to track the most profitable flower type. Intervals between successive visits were longer in the fluctuating-reward than in the stable-reward treatment, possibly because the fluctuating schedule required time-consuming neural processing. As expected, the number of minoring visits correlated negatively with the number of colour shifts and with foraging success in the stable-reward treatment. In the fluctuating-reward treatment, on the other hand, sucrose intake rates were independent of minoring and colour shift frequencies. We suggest that novelty seeking by foragers can evolve when food sources offer fluctuating rewards, which reduce foraging success but also the cost of exploring new resources.
Article
Full-text available
Animals in a poor biological state face reduced life expectancy, and as a consequence should make decisions that prioritize immediate survival and reproduction over long-term benefits. We tested the prediction that if, as has been suggested, developmental telomere attrition is a biomarker of state and future life expectancy, then individuals who have undergone greater developmental telomere attrition should display greater choice impulsivity as adults. We measured impulsive decision-making in a cohort of European starlings (Sturnus vulgaris) in which we had previously manipulated developmental telomere attrition by cross-fostering sibling chicks into broods of different sizes. We show that as predicted by state-dependent optimality models, individuals who had sustained greater developmental telomere attrition and who had shorter current telomeres made more impulsive foraging decisions as adults, valuing smaller, sooner food rewards more highly than birds with less attrition and longer telomeres. Our findings shed light on the biological embedding of early adversity and support a functional explanation for its consequences that could be applicable to other species, including humans.
Article
Full-text available
Adverse experiences in early life can exert powerful delayed effects on adult survival and health. Telomere attrition is a potentially important mechanism in such effects. One source of early-life adversity is the stress caused by competitive disadvantage. Although previous avian experiments suggest that competitive disadvantage may accelerate telomere attrition, they do not clearly isolate the effects of competitive disadvantage from other sources of variation. Here, we present data from an experiment in European starlings (Sturnus vulgaris) that used cross-fostering to expose siblings to divergent early experience. Birds were assigned either to competitive advantage (being larger than their brood competitors) or competitive disadvantage (being smaller than their brood competitors) between days 3 and 12 post-hatching. Disadvantage did not affect weight gain, but it increased telomere attrition, leading to shorter telomere length in disadvantaged birds by day 12. There were no effects of disadvantage on oxidative damage as measured by plasma lipid peroxidation. We thus found strong evidence that early-life competitive disadvantage can accelerate telomere loss. This could lead to faster age-related deterioration and poorer health in later life.