Dominance and social information use in a lizard

Abstract

There is mounting evidence that social learning is not just restricted to group-living animals, but also occurs in species with a wide range of social systems. However, we still have a poor understanding of the factors driving individual differences in social information use. We investigated the effects of relative dominance on social information use in the eastern water skink (Eulamprus quoyii), a species with age-dependent social learning. We used staged contests to establish dominant-subordinate relationships in pairs of lizards and tested whether observers use social information to more quickly solve both an association and reversal learning task in situations where the demonstrator was either dominant or subordinate. Surprisingly, we found no evidence of social information use, irrespective of relative dominance between observer and demonstrator. However, dominant lizards learnt at a faster rate than subordinate lizards in the associative learning task, although there were no significant differences in the reversal task. In light of previous work in this species, we suggest that age may be a more important driver of social information use because demonstrators and observers in our study were closely size-matched and were likely to be of similar age.

Full text

ORIGINAL PAPER
Dominance and social information use in a lizard
Fonti Kar
1,2
•Martin J. Whiting
1
•Daniel W. A. Noble
1,2
Received: 11 October 2016 / Revised: 14 May 2017 / Accepted: 26 May 2017
ÓSpringer-Verlag Berlin Heidelberg 2017
Abstract There is mounting evidence that social learning
is not just restricted to group-living animals, but also
occurs in species with a wide range of social systems.
However, we still have a poor understanding of the factors
driving individual differences in social information use.
We investigated the effects of relative dominance on social
information use in the eastern water skink (Eulamprus
quoyii), a species with age-dependent social learning. We
used staged contests to establish dominant–subordinate
relationships in pairs of lizards and tested whether obser-
vers use social information to more quickly solve both an
association and reversal learning task in situations where
the demonstrator was either dominant or subordinate.
Surprisingly, we found no evidence of social information
use, irrespective of relative dominance between observer
and demonstrator. However, dominant lizards learnt at a
faster rate than subordinate lizards in the associative
learning task, although there were no significant differ-
ences in the reversal task. In light of previous work in this
species, we suggest that age may be a more important
driver of social information use because demonstrators and
observers in our study were closely size-matched and were
likely to be of similar age.
Keywords Social learning Private information Social
status Social rank Reptile
Introduction
The social environment is a rich source of information that
can be used in individual decision-making and learning.
Social information allows observers to shortcut trial-and-
error learning, thereby bypassing the costs associated with
individual learning (Boyd and Richerson 1995; Shettle-
worth 2010 pp. 468). Costs, such as the time and energy
expended acquiring new information and the increased risk
of predation while sampling the environment, should favor
the use of social information (Rieucau and Giraldeau
2011). However, social information use is not inherently
adaptive, and theoretical analyses suggest that individuals
should use social information selectively (Rieucau and
Giraldeau 2011). Socially acquired information also may
be costly to obtain, unreliable or outdated in changing
environments, thus selection may be expected to favor
plastic learning strategies (Laland 2004).
For social information use to be advantageous, indi-
viduals should be selective about whom they learn from
(Laland 2004). Observers may preferentially learn from
certain individuals as the quality and relevance of infor-
mation is predicted to vary between individuals (Coussi-
Korbel and Fragaszy 1995). As a consequence, transmis-
sion of social information in the population can spread at
different rates because information use may be restricted to
a subset of individuals with particular traits. Social cues
such as dominance status, age or size may be indicators of
success, and an observer may use these cues to assess
whether to ‘copy’ an individual’s behavior or not (Galef
and Laland 2005). Dominance status may be indicative of
Electronic supplementary material The online version of this
article (doi:10.1007/s10071-017-1101-y) contains supplementary
material, which is available to authorized users.
&Fonti Kar
fonti.kar@gmail.com
1
Department of Biological Sciences, Macquarie University,
Sydney, NSW 2109, Australia
2
Evolution and Ecology Research Centre, School of
Biological, Earth and Environmental Sciences, University of
New South Wales, Kensington, NSW 2052, Australia
123
Anim Cogn
DOI 10.1007/s10071-017-1101-y

resource monopolization and observers may employ a
‘copy-if-dominant’ strategy to maximize resource gather-
ing opportunities (Laland 2004). However, dominance, age
and size are often confounded, whereby larger individuals
tend to be older and more dominant than smaller individ-
uals. Hence, it becomes difficult to disassociate these
effects (Aplin et al. 2013; Duffy et al. 2009). In order to
understand individual variation in social learning particu-
larly in species that show dominance hierarchies, one must
account for confounding factors such as age and size.
Social information use is most often associated with
group-living species (Lefebvre 2010). Indeed, the role of
dominance in social information use has been extensively
tested in birds and mammals. While reptiles are often
considered to be less socially complex than other verte-
brates; this does not preclude their ability to use social
information (Davis and Burghardt 2011; Kis et al. 2014;
Noble et al. 2014;Pe
´rez-Cembranos and Pe
´rez-Mellado
2015; Whiting and Greeff 1999; Wilkinson et al. 2010).
Moreover, the drivers of variation in social information
use, particularly that of dominance, remain unexplored.
Dominant individuals are predicted to be more salient than
lower ranking animals because subordinates may need to
monitor dominant individuals more closely to avoid
aggressive interactions (Nicol and Pope 1999; Shepherd
et al. 2006). Many studies have found that dominant indi-
viduals are more influential models (Kendal et al. 2014;
Krueger and Heinze 2008; Nicol and Pope 1999 but see
Awazu and Fujita 2000). Other studies show that subor-
dinates are more likely to use social information to solve
novel tasks compared to their dominant counterparts (Aplin
et al. 2013; Benson-Amram et al. 2014; Kavaliers et al.
2005; Kendal et al. 2014; Pongracz et al. 2008; Stahl et al.
2001). Using reptilian models to test questions of how
dominance influences social information use will close this
taxonomic gap, which could be fundamental to understand
the evolution of social learning strategies and any links to
social behavior (Doody et al. 2013).
We investigated whether relative dominance impacts
social learning in eastern water skinks (Eulamprus quoyii).
Eulamprus quoyii perform well on a multitude of cognitive
tasks, and previous work has shown that age is an impor-
tant factor in social information use in this species (Noble
et al. 2014). Males of this species experience contest
competition over territories and exhibit alternative repro-
ductive tactics (Kar et al. 2016; Noble et al. 2013). They
are also known to form feeding hierarchies (Done and
Heatwole 1977), suggesting that social dominance may be
an important driver in social information use. We therefore
only used male lizards to test whether dominant–subordi-
nate relationships between demonstrating and observing
lizards affect their use of social information. We staged
dyadic contests between lizards to establish dominance
relationships between pairs and then conducted social
learning experiments in which the demonstrator and
observer differed in their relative dominance. Given that
lizards continue to grow after sexual maturity (indetermi-
nate growth), age and body size are closely correlated
(Halliday and Verrell 1988). Thus, we attempted to control
for age and body size of the lizards by closely size-
matching demonstrator and observer lizards and random-
izing the body size distribution across our treatments.
Methods
Lizard collection and husbandry
We collected 56 adult male E. quoyii from nine sites in the
Sydney region during September 14–30, 2014 and brought
them back to Macquarie University. We recorded snout-to-
vent length (SVL; from tip of snout to the beginning of the
cloacal opening), total body length (from tip of snout to the
distal tip of the tail) and body mass of all lizards to the
nearest mm.
Apparatus
All trials were conducted in white opaque plastic arenas
measuring 470 (W) 9690 (L) 9455 (H) mm. In the
dominance assays, each lizard occupied half of the arena
separated by a removable wooden divider. During contests,
the divider was removed to allow lizards to interact. Sim-
ilarly, in the cognition trials, the arena was partitioned by a
permanently fixed piece of plexiglass as well as, a
removable wooden divider. At the start of each trial, this
divider was removed to allow pairs of lizards to observe
each other. Volatile chemical cues could be exchanged
through the gaps and cracks of the dividers, but animals
were not able to physically interact during trials.
For our cognition trials, all lizards were trained to dis-
place lids from two black dishes mounted on a wooden
block to access a mealworm (Tenebrio molitor). For more
details of training procedures, see supplementary materials.
Once all lizards had been trained, observer lizards were
given an association and a reversal task. The association
task consisted of one dish that was covered by a blue lid,
while the other was covered with a white lid (incorrect
dish) and required the observer lizard to displace the blue
lid (correct dish) to access a mealworm (Fig. S3). The
reversal task was essentially the same as the association
task, except that the dish containing the accessible meal-
worm was covered by a white lid (correct dish, Fig. S3).
We placed mealworms in both dishes to control for scent
and auditory cues that may differ between the two dishes,
but a piece of cardboard was placed inside the ‘incorrect’
Anim Cogn
123

dish to obstruct the food reward (association task—white,
reversal task—blue). The position (right or left) of the
correct dish was randomized and counter-balanced across
treatment groups to account for differences in lateralization
between lizards. The position of the correct dish remained
consistent within each task after this initial randomization.
We therefore cannot disambiguate whether spatial or color
cues were used to learn the tasks, as our goal was to
determine whether lizards used social information to learn
the task and not specifically test what cue was being used to
learn.
Determining male dominance status
Male contests were carried out between September 22 and
October 12, 2014 in a temperature-controlled room set at
28 °C. Males were size-matched based on SVL (mean size
difference =1 mm, range =0–5 mm). We used a tour-
nament design where individuals participated in up to
seven contests with different opponents (Whiting et al.
2006). On the day of the contest, refuges, water bowls and
dividers were removed to allow opponents to interact.
Contests were closely monitored so that once a clear out-
come was apparent the opponents were immediately sep-
arated. A clear contest outcome occurred when one lizard
fled from his opponent following an aggressive behavior
such as a chase and the lizards were at least half a body
length apart. For more details of the contest setup, see Kar
et al. (2016).
Treatment groups
We obtained 28 demonstrator–observer pairs from contests
that only resulted in a clear outcome. The winner of the
contest was assigned as the dominant lizard, while the loser
became the subordinate lizard. We assessed the stability of
the dominance relationship within pairs by staging another
round of contests, on average, 61 days (range 53–72 days)
after the pair’s initial interaction (n=26, two pairs did not
re-fight). In short, we found that 91% of pairs that inter-
acted in the second round of contests, the dominance
relationship was consistent within each pair. For details on
how we assessed the stability of dominance relationships,
see supplementary materials. We also performed sensitivity
analyses, which showed that the status of demonstrator–
observer pairs that changed in dominance relationships did
not impact our overall results (see supplementary
materials).
The demonstrator–observer pairs were randomly allo-
cated to one of the two treatment groups: (1) a social
treatment group where the observer lizard was allowed to
view the demonstrator performs the task; and (2) a control
group, where the observer lizard was allowed to view a
demonstrator that was not performing the task. In these
treatment groups, the observer of the pair was randomly
chosen to be: (1) the subordinate individual (n=12); or
(2) the dominant individual (n=16). The other individual
of the pair was assigned as the demonstrator such that a
subordinate observer was paired with a dominant demon-
strator and vice versa. While it would have been ideal to
include additional dominant–dominant and subordinate–
subordinate pairs, this was not possible given sample size
and logistical constraints. Overall, we had four treatments
consisting of: (1) subordinate control observers (n=3);
(2) subordinate social observers (n=9); (3) dominant
control observers (n=8); and (4) dominant social obser-
vers (n=8).
Association and reversal task
Social learning trials were carried out in the same room
where contests assays were held. We conducted two trials
per day, in the morning (08:30–10:00 h) and in the after-
noon (12:00–14:00 h) with a minimum interval of 2 h
between trials. At the beginning of each trial, the refuge,
water bowl and wooden divider were removed to provide a
clear view of the demonstrator. The social observer lizards
were first given six trials to view the demonstrator com-
plete the task, while control lizards viewed their demon-
strators for the same amount of time (Fig. S3). During the
six trials, the observer lizards did not receive the apparatus,
and therefore, these six trials did not count toward the total
number of trials taken to learn. Following from this, the
observers viewed the demonstrator on each trial prior to
receiving the apparatus. A lizard was considered to have
learnt the task if it displaced the correct lid 5/6 consecutive
times (a robust learning criterion—see supplementary
materials). Lizards were allowed to continue with the task
even if an incorrect choice was initially made; however,
these trials did not count toward the learning criterion. We
continued to give the task to lizards that had reached cri-
terion in order to evaluate the robustness of our learning
criterion. Lizards were given 18 trials in total to complete
the task. Lizards that did not learn the association task were
excluded from the reversal task (n=1).
In the reversal task, lizards had to reverse previously
learnt contingencies regarding the correct dish (i.e.,
blue reward in association task). All demonstrators were
first trained to displace only the white lid using the same
learning criterion as the association task prior to the com-
mencement of social demonstration. We continued to give
the task to lizards that had reached criterion. Lizards
received more trials in total in the reversal task because
lizards took slightly longer to reach criterion. In total,
lizards received 26 trials to interact with the task. However,
the total number of trials varied slightly in both tasks as
Anim Cogn
123

some lizards did not interact with the task on every trial or
were given an extra trial as they were close to reaching
criterion (association task range 14–19, reversal task range
21–27). All trials of both tasks were filmed using CCTV
cameras and a blind reviewer measured: (1) whether or not
the lizard chose the correct dish first; (2) the latency to
displace the correct lid from the moment the task was
placed inside the lizard’s enclosure; and (3) whether the
lizard displaced the lid from only the correct dish or from
both dishes. A lizard was considered to have made a choice
if it actively displaced the dish with its snout or forelimbs.
Statistical analyses
We explored the robustness of our learning criterion and
also motivation differences due to body condition and
dominance status. These analyses are presented in the
supplementary materials.
We analyzed our data in three different ways that tested
different aspects of learning. First, given that social
learning and trial-and-error learning were occurring con-
currently throughout the trials, we used a Fisher’s exact test
to test whether dominance status influenced the number of
individuals making a correct choice on the first trial in each
treatment group, for both tasks.
Second, we assessed how quickly lizards learnt the
task, based on our learning criterion, by modeling the
mean number of trials it took to learn using a generalized
linear model (GLM) with a negative binomial error dis-
tribution. Lizards that did not reach the learning criterion
were not included in the final GLM analysis (association
task: n=1, reversal task: n=2); however, exclusion of
these lizards did not impact our results. We tested the
significance of a lizard’s dominance status, treatment
group and their interaction using likelihood ratio tests
(LRT).
Given the logistical constraints in obtaining large sam-
ples sizes, which can impact Pvalues, we also calculated a
log response ratio (lnRR) to estimate an effect size (Hedges
et al. 1999). We compared the effect sizes for the mean
number of trials taken to learn the tasks between: (a) con-
trol and social lizards, (b) dominant control and dominant
social lizards, (c) subordinate control and subordinate
social lizards and (d) subordinate social and dominant
social lizards.
We also ran additional analyses to investigate how lizard
cognitive performance changed across trials using gener-
alized linear mixed effect models. We modeled the
‘probability of choosing the correct dish first,’ the proba-
bility of ‘choosing only the correct dish,’ as well as ‘la-
tency to displace the correct lid.’ The results of these
analyses were largely congruent with our GLM results and
are presented in the supplementary materials.
In all models, treatment (social and control) and status
of the observer (dominant =DOM and subordi-
nate =SUB) were coded as two-level factors. We tested
whether differences in cognitive performance depended on
a lizard’s treatment group and/or dominance status by
including an interaction term between treatment and status
because we hypothesized that dominant and subordinate
lizards may use social information differently. We
z-transformed SVL and included it in all models as a
covariate to account for any differences between treatments
in body size that may influence learning. Data for this
study is available from doi:10.6084/m9.figshare.4981958.
Results
Association task
Dominance status did not influence the number of indi-
viduals making a correct choice on the first trial in either
treatment group (P=0.20). Overall, 27 of 28 (96%)
observer lizards learnt the task. All 12 (100%) subordinate
lizards learnt the task (nine social, three controls), whereas
15/16 (94%) of dominant lizards learnt the task (seven
social, eight control).
There were no differences in the mean number of trials
taken to learn between control and social lizards in a model
pooling lizards of both dominance statuses (GLM: esti-
mate =-0.15, SE =0.16, P=0.35; LRT: v2=0.89,
P=0.35). The mean number of trials it took for lizards to
learn depended on a lizard’s dominance status
(Table 1a, LRT: v2=7.92, P=0.01), but not its treatment
group (Table 1a, LRT: v2=0.21, P=0.65), or their
interaction (Table 1a, LRT: v2=0.05, P=0.82). Domi-
nant social lizards learnt the association task in significantly
fewer trials compared to subordinate social lizards (Fig. 1a,
Table 1a). The mean number of trials taken to learn was 20%
smaller in control lizards compared to social lizards
(lnRR =0.18, r2=0.13). The mean number of trials taken
to learn for dominant control lizards was 8% smaller than
dominant social lizards (lnRR =0.08, r2=0.14), whereas
it was 10% smaller in subordinate control lizards compared
to subordinate social lizards (lnRR =0.09, r2=0.37). The
mean number of trials taken to learn was 29% smaller in
dominant social lizards compared to subordinate social
lizards (lnRR =0.34,r2=0.24).
Reversal task
The number of individuals making a correct choice on the
first trial was not associated with dominance status, in
either treatment group (P=0.20). Twenty-six of 28 (93%)
Anim Cogn
123

observer lizards learnt the task. All 12 (100%) subordinate
lizards learnt the task (nine social, three controls), whereas
13/15 (86%) of dominant lizards learnt the task (seven
social and six were controls).
There was a weak significant difference in the mean
number of trials taken to learn between control and
social lizards in a model pooling lizards of both dom-
inance statuses (GLM: estimate =-0.46, SE =0.23,
P=0.05; LRT: v2=4.37, P=0.038). The mean
number of trials it took for lizards to learn did not
depend on treatment group (Table 1b, LRT: v2=2.86,
P=0.09), dominance status (Table 1b, LRT:
v2=0.08, P=0.78) or their interaction
(Table 1b, LRT: v2=0.76, P=0.38). There was a
trend for both dominant and subordinate control lizards
to take fewer trials to learn than their social treatment
counterparts; however, this was not signifi-
cant (Fig. 1b). The mean number of trials taken to learn
for control lizards was 57% smaller compared to social
lizards (lnRR =0.45, r2=0.30). The mean number of
trials taken to learn for dominant control lizards was
33% smaller than dominant control lizards
(lnRR =0.28, r2=0.83), whereas it was 93% smaller
in subordinate control lizards compared to subordinate
social lizards (lnRR =0.66, r2=0.43). The mean
number of trials taken to learn was 20% smaller in
dominant social lizards compared to subordinate social
lizards (lnRR =0.18, r2=1.21).
Table 1 Estimates and standard errors (SE) from a generalized linear
model (GLM) examining the effects of a lizard’s dominance status
(dominant or subordinate), treatment group (social or control) and
standardized SVL xl=r½on the mean number of trials it took for a
lizard to learn the aassociation task (n=27) and bthe reversal task
(n=25)
(a) Association task (b) Reversal task
Estimate SE Estimate SE
Intercept 2.31 0.10 2.40 0.14
Status DOM 20.43 0.15 -0.06 0.21
Treatment control 0.08 0.18 -0.42 0.25
Scaled SVL 0.12 0.08 0 0.12
Status 9treatment -0.07 0.30 0.38 0.44
Bolded estimates are significant. Main effects are presented from a
model without the interaction
05
10 15 20
Mean number of trials taken to learn
Control Social Control Social
DOM SUB
n = 8 n = 7 n = 3 n = 9
(a)
0510
15 20
Mean number of trials taken to learn
Control Social Control Social
DOM SUB
n = 7 n = 6 n = 3 n = 9
(b)
Fig. 1 Raw mean number of
trials and sample sizes to learn
for athe association task and
bthe reversal task for dominant
(DOM) and subordinate (SUB)
lizards in the social
demonstration (gray bars) and
control (white bars) treatments.
Error bars represent standard
error. Note that subordinate
control lizards all achieved the
learning criterion at the same
time and therefore do not have
an error estimate
Anim Cogn
123

Discussion
We show that lizards from the social demonstration treatment
were no more likely to make a correct choice on the first trial
compared to the control group. To our surprise, the social
demonstration treatment did not learn more quickly than the
control group, providing weak evidence that observers were
using social information to learn the association or reversal
task. Our results also suggest that dominant social lizards
learnt the association task in significantly fewer trials, com-
pared to subordinate social lizards. However, there were no
differences in the number of trials required to reachcriterion in
the reversaltask. Given that we did not find evidence of social
information use, this result may reflect underlying differences
in trial-and-error learning between dominant and subordinate
lizards during associative learning.
Lack of social learning in a novel foraging task
Contrary to our predictions, watching a demonstrator exe-
cute the task did not expedite learning in the association and
reversal tasks compared to the control group. This may be
because trial-and-error learning was not particularly costly in
our experiment. Observer lizards have little to lose from
displacing lids from both dishes, as they would still even-
tually be rewarded if they chose the incorrect dish first.
Individuals are predicted to rely on private information if
trial-and-error learning is relatively inexpensive compared
to social information, as it may be more accurate (Boyd and
Richerson 1995; Kendal et al. 2005; Rieucau and Giraldeau
2011). Indeed, naı
¨ve European starlings (Sturnus vulgaris)
have been shown to ignore the sampling behavior of a
demonstrator and rely on private information about the
quality of a food patch when private information was easy to
acquire (Templeton and Giraldeau 1996). However, as the
difficulty of trial-and-error learning increased, naı
¨ve star-
lings were more likely to exploit social information to infer
food patch depletion. Nine-spined sticklebacks (Pungitius
pungitius) initially relied on private information to make
decisions about where to forage, but as private information
becomes less reliable over time, they switched to using social
information (van Bergen et al. 2004). This suggests that the
reliability and difficulty of acquiring private information can
affect the use of social information to acquire new infor-
mation and may explain why we found no evidence of social
information use in our experiment.
Alternatively, we may not have detected social information
use because we size-matched demonstrators and observers
and by doing so, we may have age-matched them as well. In E.
quoyii, young lizards in the presence of larger, older demon-
strators learnt an association task significantly faster than
older lizards watching same-aged demonstrators, suggesting
that older E. quoyii may not use social information when
demonstrators are of a similar age or size (Noble et al. 2014).
Given that our experiment attempted to disassociate age and
dominance, we may have effectively removed age effects and
thus did not detect social information use. Taken together,
these results seem to suggest that age may be the major driver
of social information use (at least on association tasks) in E.
quoyii. The results of these two studies together represent a
unique situation where the confounding effects of age and
dominance have been successfully disassociated in a single
study system. Indeed, naivety can be a strong driver of social
information use in many systems (Duffy et al. 2009; Galef
et al. 2001; Noble et al. 2014). This is not surprising, as
juveniles have much to gain by using social learning during a
vulnerable stage of their lives by exploiting social information
from older, more experienced individuals (Rieucau and Gir-
aldeau 2011). However, we do need to consider that these
effects may be the result of the low statistical power in our
study and future work replicating these experiments would be
needed to verify these conclusions.
Dominance and trial-and-error learning
Dominant social lizards learnt the association task in sig-
nificantly fewer trials compared to subordinate social lizards
based on our learning criterion. Given that we did not detect
the use of social information in either task, this result seems
to suggest differences in the rate of trial-and-error learning
between dominant and subordinate lizards and may reflect
differences in motivation or foraging behavior between
dominant and subordinate lizards. Dominant individuals
have been reported to be superior at trial-and-error learning
in a range of species including meadow voles and European
starlings (Boogert et al. 2006; Spritzer et al. 2004). While
dominant individuals may be intrinsically better than sub-
ordinates at cognitive tasks, learning ability may also be
affected by social context. For example, dominant individ-
uals tend to excel in both group contexts as well as in iso-
lation, whereas subordinate individuals tend to thrive only in
isolated contexts (Drea and Wallen 1999). Stress associated
with learning in the presence of a dominant demonstrator
may have also reduced learning ability of subordinate indi-
viduals. Future studies should consider testing subordinate
and dominant individuals in isolation in order to test for
differences in trial-and-error learning.
Conclusions
We found no support for the hypothesis that relative
dominance affects social information use in E. quoyii.
Lizards that viewed a demonstrator perform a task,
Anim Cogn
123

regardless of whether they were subordinate or dominant,
did not learn faster than the control group. Interestingly,
social dominance predicted associative learning ability:
dominant individuals reached criterion faster than subor-
dinate individuals. Many of the lizards in dominant–sub-
ordinate pairs were matched in size, and therefore, they
may also be similar in age. Using this design, we had more
power to detect an effect for dominance at the expense of
an age effect. It is possible that there may be an effect with
greater disparity in dominance or age. Future studies that
are able to use individuals of known age would be very
valuable in studying the interaction between age and
dominance and its potential role in social learning.
Acknowledgements We would like to thank the two anonymous
reviewers for their constructive feedback on the earlier version of this
manuscript. We are grateful for Christine Wilson for scoring our
video footage, and we would also like to thank the numerous mem-
bers of the Lizard Lab that assisted us with lizard collection, hus-
bandry and experimental setup.
Funding DWAN was supported by an Australian Research Council
(DECRA: DE150101774), and this work was also supported by
Macquarie University and a Discovery Grant (DP130102998) awar-
ded by the Australian Research Council to MJW.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
Ethical approval All protocols for this study were in accordance
with the ethical standards of the Macquarie University Animal Ethics
Committee (ARA 2014/036). A scientific permit for this study was
granted by the New South Wales National Parks and Wildlife Service,
Office of Environment and Heritage (SL100328).
References
Aplin LM, Sheldon BC, Morand-Ferron J (2013) Milk bottles
revisited: social learning and individual variation in the blue
tit, Cyanistes caeruleus. Anim Behav 85:1225–1232. doi:10.
1016/j.anbehav.2013.03.009
Awazu S, Fujita K (2000) Influence of dominance on food transmis-
sion in rats. Jpn J Anim Psychol 50:119–123
Benson-Amram S, Heinen V, Gessner A, Weldele ML, Holekamp KE
(2014) Limited social learning of a novel technical problem by
spotted hyenas. Behav Process 109:111–120
Boogert NJ, Reader SM, Laland KN (2006) The relation between
social rank, neophobia and individual learning in starlings. Anim
Behav 72:1229–1239
Boyd R, Richerson PJ (1995) Why does culture increase human
adaptability? Ethol Sociobiol 16:125–143
Coussi-Korbel S, Fragaszy DM (1995) On the relation between social
dynamics and social learning. Anim Behav 50:1441–1453
Davis KM, Burghardt GM (2011) Turtles (Pseudemys nelsoni) learn
about visual cues indicating food from experienced turtles.
J Comp Psychol 125:404–411
Done BS, Heatwole H (1977) Social behavior of some Australian
skinks. Copeia 1977:419–430
Doody JS, Burghardt GM, Dinets V (2013) Breaking the social–non-
social dichotomy: a role for reptiles in vertebrate social behavior
research? Ethology 119:95–103. doi:10.1111/eth.12047
Drea CM, Wallen K (1999) Low-status monkeys ‘‘play dumb’’ when
learning in mixed social groups. Proc Natl Acad Sci USA
96:12965–12969
Duffy GA, Pike TW, Laland KN (2009) Size-dependent directed
social learning in nine-spined sticklebacks. Anim Behav
78:371–375. doi:10.1016/j.anbehav.20
Galef BG, Laland KN (2005) Social learning in animals—empirical
studies and theoretical models. Bioscience 55:489–499
Galef BG, Marczinksi CA, Murray KA, Whiskin EE (2001) Food
stealing by young Norway rats (Rattus norvegicus). J Comp
Psychol 115:16–21
Halliday TR, Verrell P (1988) Body size and age in amphibians and
reptiles. J Herpetol 22(3):253–265
Hedges LV, Gurevitch J, Curtis PS (1999) The meta-analysis of
response ratios in experimental ecology. Ecology 80:1150–1156.
doi:10.2307/177062
Kar F, Whiting MJ, Noble DW (2016) Influence of prior contest
experience and level of escalation on contest outcome. Behav
Ecol Sociobiol 70:1679–1687. doi:10.1007/s00265-016-2173-4
Kavaliers M, Colwell D, Choleris E (2005) Kinship, familiarity and
social status modulate social learning about ‘‘micropredators’’
(biting flies) in deer mice. Behav Ecol Sociobiol 58:60–71.
doi:10.1007/s00265-004-0896-0
Kendal RL, Coolen I, van Bergen Y, Laland KN (2005) Trade-offs in
the adaptive use of social and asocial Learning. Adv Stud Behav
35:333–379. doi:10.1016/s0065-3454(05)35008-x
Kendal R, Hopper LM, Whiten A, Brosnan SF, Lambeth SP, Schapiro
SJ, Hoppitt W (2014) Chimpanzees copy dominant and knowl-
edgeable individuals: implications for cultural diversity. Evol
Hum Behav 36:65–72
Kis A, Huber L, Wilkinson A (2014) Social learning by imitation in a
reptile (Pogona vitticeps). Anim Cognit 18:325–331
Krueger K, Heinze J (2008) Horse sense: social status of horses
(Equus caballus) affects their likelihood of copying other horses’
behavior. Anim Cognit 11:431–439
Laland KN (2004) Social learning strategies. Learn Behav 32:4–14
Lefebvre L (2010) Taxonomic counts of cognition in the wild. Biol
Lett. doi:10.1098/rsbl.2010.0556
Nicol CJ, Pope SJ (1999) The effects of demonstrator social status
and prior foraging success on social learning in laying hens.
Anim Behav 57:163–171
Noble DW, Wechmann K, Keogh JS, Whiting MJ (2013) Behavioral
and morphological traits interact to promote the evolution of
alternative reproductive tactics in a lizard. Am Nat 182:726–742
Noble DW, Byrne RW, Whiting MJ (2014) Age-dependent social
learning in a lizard. Biol Lett. doi:10.1098/rsbl.2014.0430
Pe
´rez-Cembranos A, Pe
´rez-Mellado V (2015) Local enhancement and
social foraging in a non-social insular lizard. Anim Cognit
18:629–637
Pongracz P, Vida V, Banhegyi P, Miklosi A (2008) How does
dominance rank status affect individual and social learning
performance in the dog (Canis familiaris)? Anim Cognit
11:75–82. doi:10.1007/s10071-007-0090-7
Rieucau G, Giraldeau LA (2011) Exploring the costs and benefits of
social information use: an appraisal of current experimental
evidence. Philos Trans Roy Soc B 366:949–957. doi:10.1098/
rstb.2010.0325
Shepherd SV, Deaner RO, Platt ML (2006) Social status gates social
attention in monkeys. Curr Biol CB 16:R119–R120. doi:10.
1016/j.cub.2006.02.013
Anim Cogn
123

Shettleworth SJ (2010) Cognition, evolution, and behavior. Oxford
University Press, New York
Spritzer MD, Meikle DB, Solomon NG (2004) The relationship
between dominance rank and spatial ability among male
meadow voles (Microtus pennsylvanicus). J Comp Psychol
118:332–339. doi:10.1037/0735-7036.118.3.332
Stahl J, Tolsma PH, Loonen MJ, Drent RH (2001) Subordinates
explore but dominants profit: resource competition in high Arctic
barnacle goose flocks. Anim Behav 61:257–264
Templeton JJ, Giraldeau L-A (1996) Vicarious sampling: the use of
personal and public information by starlings foraging in a simple
patchy environment. Behav Ecol Sociobiol 38:105–114
van Bergen Y, Coolen I, Laland KN (2004) Nine-spined sticklebacks
exploit the most reliable source when public and private
information conflict. Proc R Soc B Biol Sci 271:957–962
Whiting MJ, Greeff JM (1999) Use of heterospecific cues by the
lizard Platysaurus broadleyi for food location. Behav Ecol
Sociobiol 45:420–423
Whiting MJ, Stuart-Fox DM, O’Connor D, Firth D, Bennett NC,
Blomberg SP (2006) Ultraviolet signals ultra-aggression in a
lizard. Anim Behav 72:353–363
Wilkinson A, Kuenstner K, Mueller J, Huber L (2010) Social learning
in a non-social reptile (Geochelone carbonaria). Biol Lett
6:614–616. doi:10.1098/rsbl.2010.0092
Anim Cogn
123