Journal of Ethology (2023) 41:185–193

https://doi.org/10.1007/s10164-023-00784-3

ARTICLE

Bigger male Eurasian nuthatches (Sitta europaea) behave more

aggressively in playback‑simulated territorial intrusion
Mikhail Diatroptov1 · Alexey Opaev1

Received: 16 April 2022 / Accepted: 9 March 2023 / Published online: 30 March 2023
© The Author(s) under exclusive licence to Japan Ethological Society 2023

Abstract
In animals, personality (or temperament) is associated with alternative response patterns in reaction to a potential stressor or
challenging situation. Aggressiveness is one of the basal behavioral axes of variation of personality. Consistent individual
differences in this behavior may arise due to genetic effects or as a result of irreversible developmental plasticity. Here, we
used playback experiments with a songbird to ask whether aggressiveness varies according to body size, body condition and
age of the individuals. In passerine birds, song is very important in territorial defense along with aggressive behavior. We
found that the male acoustic response was not influenced by body size and body condition in playback-simulated territorial
intrusion, while the behavioral response was. Bigger males behaved more aggressively as they performed more flyovers,
came closer to the loudspeaker, and displayed more intensively. Body size did not depend on age in nuthatches and was
individually repeatable between years. Therefore, our study added to accumulating evidence that individuals show consistent
differences along the aggressiveness axis of behavioral variation, and these differences apparently remains constant through
an individual’s life.

Keywords Animal personality · Body size · Playback experiments · Territorial competition · Bird song

Introduction along with direct aggression, e.g. fighting (Catchpole and

Slater 2008). Territorial behavior is important, because ter-
Consistent individual differences in behavior, often labeled ritory can influence greatly reproductive success and sur-
as personality, arise due to genetic effects or as a result of vival. For example, males who occupy territories faster and
irreversible developmental plasticity. Personalities have been of higher quality could be in advantage because the timing
observed along several behavioral axes, such as aggressive- of breeding and nest-site quality are known to have crucial
ness, risk-taking, activity, exploration, and sociality (Réale importance to reproductive success (Bensch and Hasselquist
et al. 2007), and have been documented in a broad diversity 1991; Ens et al. 1992). Territory could be also crucial for
of vertebrate and invertebrate taxa (Gosling 2001). survival because territorial individuals generally have higher
Here, studying a songbird, we focused on aggressiveness survival rates than non-territorial floaters (Lenda et al.
in territorial competition context as it is critical for many 2012).
animals because of their need to defend a territory. Territory In several songbird species, territorial aggression has
is an area occupied by one male of a species which it defends been found to be an individual- specific trait, repeatable
against intrusions of other males of the same species, and within individuals. Examples include blue tit (Parus caer-
territorial behavior is aimed at monopolizing this area. In uleus) (Salazar et al. 2021), collared flycatcher (Ficedula
passerine birds, song is very important in territorial defense albicollis) (Garamszegi et al. 2006; Morinay et al. 2019),
great tit (Parus major) (Araya-Ajoy and Dingemanse 2017),
and Sardinian warbler (Sylvia melanocephala) (Beltrão et al.
* Alexey Opaev 2021). Because repeatability normally sets an upper limit to
[email protected]
heritability (Araya-Ajoy and Dingemanse 2017), one can
1
Laboratory of Comparative Ethology assume a heritable component in this behavior. In accord-
and Biocommunication, A.N. Severtsov Institute ance, aggressiveness can vary independently of age and pre-
of Ecology and Evolution of Russian Academy of Sciences, vious experience but be associated with other more or less
Moscow 119071, Russian Federation

13
Vol.:(0123456789)
186 Journal of Ethology (2023) 41:185–193

heritable traits, such as color polymorphism. For example, Materials and methods
discrete color morphs have been found to differ in levels
of aggression in Gouldian finch (Erythrura gouldi) (Wil- General field methods
liams et al. 2012) and ruff (Philomachus pugnax) (Lank et al.
1995). This study was carried out in the Bittsevsky Park, Mos-
In birds, body size appeared to be partly inheritable trait. cow, Russian Federation, in 2021. The study plot was a
It was found that in some songbird species a part of the total ca. 2 ­k m 2 area of mixed forest located in the northern
variation of body size is caused by genetic effects (Boag part of the park. A total of 37 natural active nest cavities
and Grant 1978; Smith and Zach 1979; Dhondt 1982; van were found in this plot in 2021 (Fig. S1). Mean distance
Noordwijk et al. 1988). In accordance, this characteristic is between cavities was 151 ± 63 m (range 68–320 m, n = 37).
more or less constant through the life as young individuals For each nest cavity, we measured distance to the nearest
(i.e., 1st year) do not differ in body size from the older ones cavity, mean distance to the nearest two cavities, mean dis-
(e.g. Marra 2000). However, this does not hold in some spe- tance to the nearest three cavities, and mean distance to the
cies (e.g. Young et al. 2017). When body size is independent nearest five cavities. All these 4 variables were highly cor-
of age, the association between body size and aggressive- related: spearman rank order correlation, R = 0.84–0.96,
ness will favor a heritable genetic component in aggressive n = 36, and p < 0.01 in all tests.
behavior. Some studies found body size positively associ- We captured 34 out of 37 males in January–March at the
ated with aggression in common waxbill (Estrilda astrild) feeders. Captured individuals were measured (wing length,
(Funghi et al. 2015), crimson finch (Neochmia phaeton) bill length, and tarsus length to the nearest 0.5 mm) and
(Young et al. 2017), and Sardinian warbler (Beltrão et al. ringed. They were categorized as 1 year-old (i.e., hatched
2021). Nonetheless, this pattern is not always supported, last spring) or adult. Individuals hatched last year under-
considering that in some species body size did not influ- goes a partial post-juvenile molt, and there is a difference
ence the contests. For example, body size did not predict in the intensity of the tinge between the juvenile primary
territorial status (either territory holder or floater) in black coverts and the post-juvenile greater coverts (Jenni and
kite (Milvus migrans) (Sergio et al. 2009). Therefore, further Winkler 2011). We used this tinge difference to distinguish
studies are needed to test the relationship between body size between 1 year-old and adult males. As an additional cri-
and aggressiveness. terion, we used the shape of the tail feathers as they tend
Our study species was the European nuthatch (Sitta euro- to be narrower in juveniles than in adults (Svensson 1992).
paea), a passerine bird well suited for testing the relationship Then, in March, we evaluated 36 out of 37 male nut-
between aggression and body size. It is a single-brooded, hatches (including all the 34 marked individuals) for their
cavity-nesting bird that forms a permanent pair bond and aggressive response to playback-simulated territorial
actively defends an all-purpose territory throughout the intrusion during the nest-building phase. We chose this
year. Therefore, this is an extremely territorial species. We phase because we aimed to test the male response not the
used playback experiments to test whether acoustic and joint response of the pair members. Mainly, females alone
aggressive behavior during territorial competition context participate in the nest building in this species (although
depends on body size. Apart from body size per se, males males can carry nest material) and generally does not react
could consistently differ in their musculature characteristics. aggressively to playback in this phase (pers. obs.). By con-
To account for individual differences in the muscle mass, trast, during the chick-feeding phase paired individuals
we additionally estimated body condition at the moment of can give cooperative response to playback-simulated ter-
the lowest weight, i.e., when the fat content was minimal or ritorial intrusion (Naďo et al. 2018). Additionally, dur-
even absent. ing the breeding season, testosterone levels of adult males
The heritability of aggressiveness is not absolute, because exhibit stage-specific differences, higher during the nest-
aggressiveness depends on many other factors, including age building and egg laying phases than during the pre-nesting
and population density. For example, older males were found and incubation (Landys et al. 2010). Testosterone eleva-
to behave more aggressively and dominate younger conspe- tions may upregulate male–male aggression in nuthatches
cifics in several species (Edler and Friedl 2010; Young et al. (Landys et al. 2010), a pattern known in several other pas-
2017). It is also known that male’s territorial behavior can serines (Hau and Goymann 2015). When testosterone is
depend on local density (Shonfield et al. 2012; Araya-Ajoy elevated, males are more stimulable by playback, therefore
and Dingemanse 2017). Therefore, these factors were also individual differences in territorial behavior can be more
taken into account. In this study, we tested a hypothesis that, easily detected.
in nuthatches, aggressiveness depends to the greatest extent Later in the season, 29 out of 34 marked males were
on body size, and to a lesser extent on age and population re-captured at their nest cavities using a small net fixed
density.

13
Journal of Ethology (2023) 41:185–193 187

on a stick. Individuals were weighed to the nearest 0.1 g. 10–15 m away from the nesting tree). Then, the observer(s)
The nuthatch males were captured in the early morning withdrew to a distance of 10–20 m and an experimental trial
(i.e., when the stomach was empty) between 11 and 16 was initiated. Usually, only males responded to playback,
May. At the time of capture, there were 10–15 day old since females were building a nest. Males typically reacted
nestlings in the nests. Earlier, we have found that male to playback by approaching in a few short flights and then
nuthatches have a minimum body mass from the late nest- repeatedly flying over or near the loudspeaker, and by signal-
ling phase until the molt starts because there is no fat and ing their claim to a territory through vocal activity.
testes are reduced (Diatroptov and Diatroptov 2017). More Playback was run for 5 min, during the time that the
importantly, body mass measured in this period was the observer scored or estimated the following behaviors: num-
most repeatable between years (Diatroptov, unpublished ber of flyovers (male flew for more than 1 m within a radius
data). Therefore, nuthatches had the lowest weight of the of 10 m from the loudspeaker), minimum distance to the
year at the moment of capture, and individual differences loudspeaker (m), flyover radius (visual estimation of the
in weight were mainly due to differences in the muscle radius of the area within which the male flew during the
weight of the males. experiment, to the nearest 5 m), and the threat display char-
To evaluate whether body size and body mass are individ- acterized by spread or fanned tail and dropped and spread
ually repeatable between years, we additionally captured 20 wings (four levels were identified: 0—no threat display
marked males in March and May 2022. In March 2022, we was observed during playback presentation, I–1–3 displays
measured both body size (wing length, bill length, and tarsus were observed, II—5–15 displays, and III—more than 15
length) and body mass, while in May we only weighed the displays or displays were observed constantly when the
birds. In 2021, body size measurements were obtained for all male perched). We estimated the minimum distance to the
these males, but only 17 of them were weighted in that year. loudspeaker only for those 31 individuals who approached
the loudspeaker 10 m or closer. The flyover radius was esti-
Playback stimulus mated only for individuals (n = 24) who performed at least
5 flyovers.
The nuthatch recording used to prepare the stimulus was During each experiment, the observer continuously
downloaded from Xeno-canto. We used recording obtained recorded male vocalizations using a Marantz PMD-660 digi-
in Poland by Izabela Dluzyk (XC310154). We selected a sin- tal recorder with a Sennheiser ME66-K6 microphone. When
gle trill from this recording and used it to prepare playback the playback was over, additional 5 min of male vocaliza-
stimulus. The time-and-frequency parameters (duration of tions were recorded.
trill, duration of notes, number of notes in the trill, minimum We performed a single experiment with each 36 males,
and maximum frequencies) of this trill fitted well among the and only data from the ringed 34 males was used in the
range of the TR-1 song (see below for definition of TR-1 analysis.
song, section ‘Acoustic analysis’) frequently produced by the
nuthatches in our study population (Fig. S2). The stimulus Acoustic analysis
was composed of one repeated trill (i.e., TR-1 song) having
17 syllables, had 64 repeats of this trill, and lasted for 5 min. Nuthatches typically sing in short bouts of similar songs
In this study, we aimed to test the between-individual differ- and typically switch song types after several repetitions of a
ences in response. In that case, the usage of different stimuli given song type. The song has a simple structure, consisting
could complicate the interpretation of the data. Therefore, of short series of identical notes (strophes). The five basic
we used only one stimulus to avoid an increase in within- song types have been described from European populations.
stimulus variance (Naďo et al. 2018). They have been descriptively named ‘slow-ascending’ (SA),
‘fast-ascending’ (FA), ‘descending’ (DE), ‘trill’ (TR), and
Playback procedure ‘up–down’ (UD) (Matthysen 1997). Both SA and FA songs
have notes with an upward inflection. SA songs are defined
Experiments were conducted between 24 and 29 March as having up to four notes per second, while FA has more
during morning hours (8–12 a.m.) in good weather condi- than four. DE is a slow song with a downward inflection. UD
tions. Testing for territorial aggression was conducted in is another slow song with notes that have an audible upward
the vicinity of an active nest cavity. We simulated territorial and downward inflection. TR is a fast trill (not less than ten
intrusion by broadcasting the conspecific TR-1 song (see notes per second). In our sample, we identified two ‘vari-
below for definition of this song) at a sound pressure level ants’ in 3 out of 5 basic song types, while 2 basic song types
of 83–84 dB(A). Before each experiment, a loudspeaker had no ‘variants’ (Fig. 1). A single male could produce both
JBL Flip 4 was placed on a branch at the height of 1–1.5 m ‘variants’ of e.g. song type(s) SA, UD, and/or TR. Because
from the ground, in the vicinity of an active nest cavity (ca. of this, we treated each ‘variant’ of a basic song type as a

13
188 Journal of Ethology (2023) 41:185–193

Fig. 1  Population repertoire of the song and call types. See text for abbreviations

true ‘song type’. Thus, in our study population, there were call rate, the presence/absence of each out of 16 song and
8 song types (Fig. 1). call types (i.e., 16 variables in total).
In addition to song, males produced two basic types of
call in the experiments: the excitement call (EX) and alarm Statistical analysis
call (AL) (Matthysen 1998). There were six types of the
excitement call in our sample, and two alarm call types. All statistical analyses were conducted in R (R Development
Therefore, the population repertoire consisted from 8 call Core Team 2020).
types (Fig. 1). In order to assess body size, we performed a principal
For sound visualization and analysis, we used Syrinx component analysis (PCA) using the stats package (func-
PC version 2.6 with a fast Fourier transform size = 256, tion ‘prcomp’) and the factoextra package (Kassambara and
and a Hanning window type. To analyze calling activity, Mundt 2017). In the analysis, we used measurements of
male vocalizations recorded during two experimental stages wing length, bill length and tarsus length from 34 individu-
(during and just after playback presentation) were processed als. Before the PCA was run, we used the REdaS package in
separately. First of all, we estimated the song- and call-type R (Hatzinger et al. 2014; Maier 2015) to evaluate whether
diversity, that is, the number of different song and call types, our data were suitable for this analysis. We found our data
respectively. For each recording, we noted whether a given suitable because Bartlett’s test of sphericity showed that the
song and call type was either present or absent. correlation matrix was significantly different from an iden-
There was considerable variation in the number of notes tity matrix (Chi-square test: X2 = 29.3, df = 3, p < 0.001), and
per second between different song and call types. On the the Kaiser–Meyer–Olkin measure of sampling adequacy was
one hand, song types SA, UD, and DE, and alarm calls were 0.69 (above the minimum acceptable value of 0.5; Kaiser
relatively slow, usually delivering at a rate of no more than 1974) indicating that a large proportion of variance in the
two or three notes per second. Hence, many strophes of these variables can be explained by the components. Using PCA,
sounds consisted of just a single note. Therefore, we counted we extracted the first PC (eigenvalue 2.1) that explained
the number of notes for SA, UD, DE, and AL. On the other 70.6% of the total variation. We used PC1 in the analysis.
hand, songs FA and TR, and call EX were faster, with more Other PCs had eigenvalues smaller than 1 and thus were not
notes per second or up to 10–20 notes in a single strophe. used in the analysis. Hereafter, we refer to the term ‘body
The number of strophes per recording was used in the analy- size’ as PC1 (Sergio et al. 2009). It should be noted that
sis of FA, TR, and EX. We calculated song and call rates as body size did not depend on male’s age (Mann–Whitney
the number of strophes or notes divided by the record length. test: W = 111, p = 0.282).
In total, we used the following 20 acoustic variables in the Body condition index was calculated for 29 males as
analysis: song-type diversity, call-type diversity, song rate, residuals from a least squared linear regression between

13
Journal of Ethology (2023) 41:185–193 189

body mass (dependent variable) and body size (Scham- per 5 min playback interval. The corresponding values for
ber et al. 2009; Labocha and Hayes 2012). Body condition 5 min post-playback interval were 0.7 song types (range 0–2,
index was not dependent on morphometric variables (wing n = 36) and 0.5 call types (range 0–2, n = 36). Each male
length, tarsus length, and bill length: spearman rank order could produce both songs and calls at any interval. Different
correlation: p > 0.05 in all tests), but depended on body mass song/call types did not occur with equal probability. During
(Spearman rank order correlation, R = 0.68, p < 0.001). playback presentation, the most frequently heard vocaliza-
In order to assess local density, we performed PCA as tions were SA-2, FA, TR-1, TR-2, EX-3, and AL-1. Just
described above. In the analysis, we used the four distance after playback, FA, EX-2, and EX-3 were emitted by many
measurements for 37 individuals: distance to the nearest cav- males (Fig. 2).
ity, mean distance to the nearest two cavities, mean distance Song-type diversity, call-type diversity, song rate and
to the nearest three cavities, and mean distance to the nearest call rate were not associated with body size and body con-
five cavities. Bartlett’s test of sphericity showed that the cor- dition either during or after playback presentation (Supp.
relation matrix was significantly different from an identity Tables 1–2). Also, there were no effects of age and local
matrix (Chi-square test: X2 = 282.2, df = 6, p < 0.001), and density on acoustic variables (Supp. Tables 1–2).
the Kaiser–Meyer–Olkin measure of sampling adequacy was The presence/absence of any song and call type did not
0.74 (above the minimum acceptable value of 0.5). There- significantly depend on body size, body condition, age and
fore, this data was suitable for the PCA. Using PCA, we local density. All but one model had p values greater than
extracted the first PC that explained 95.1% of the total vari- 0.05 (GLM). The only exception was that males living in
ation. We used this PC in the analysis. Hereafter, we refer to close proximity to other individuals were less likely to use
the term ‘local density’ as PC1. SA-1 songs during playback (estimate = –0.02 ± 0.01, conf.
Generalized linear models (GLM: function ‘glm’ in R) int. = − 0.04– − 0.005). However, this result seems to be
were used to test the effect of body size, body condition weak as only 3 out of 36 males produced this call.
index, age, and local density on the acoustic and behavio-
ral variables. The analysis of acoustic variables was per- Behavioral response
formed separately for two intervals: during and just after
playback presentation. The models including song rate, call The behavioral response clearly depended on body size
rate, and minimum distance to the loudspeaker as a response (Supp. Table 1). Bigger males performed more flyovers
variable were fitted with a Gaussian error distribution, while (Fig. 3a), had a larger flyover radius, came closer to the loud-
those having threat display and flyover radius were fitted speaker, and displayed more intensively (Fig. 3c). Among
with a Poisson error distribution. Variables “song-type males who did not show the threat display at all, the smaller
diversity” and “call-type diversity” had an excess of zero ones predominated. This group differed significantly from all
counts. Therefore, the zero-inflated Poisson regression in
the R package “pscl” (Zeileis et al 2008) was used to model
these variables. This model has two parts, a Poisson count
model and the logit model for predicting excess zeros. For
the variable ‘threat display’, we additionally used ANOVA
in R and then Tukey's post hoc test (function ‘TukeyHSD’ in
R) to test for differences among groups. Generalized linear
models fitted with a binomial error distribution were used to
model the presence/absence data for each of 16 song and call
types. GLMs were also used to test the repeatability of wing
length, bill length, tarsus length and body mass between
years. These models were fitted with a Gaussian error dis-
tribution. We used Mann–Whitney test to analyze whether
body size and body condition depended on male age.

Results

Acoustic response

Males responded vocally with an average of 1.2 song types Fig. 2  Percentage of males produced song and call types during play-
(range 0–3, n = 36) and 0.6 call types (range 0–3, n = 36) back (a) and after playback (b)

13
190 Journal of Ethology (2023) 41:185–193

Fig. 3  Behavioral response of males observed during presentation regression lines and their 95%-confidence intervals. Boxes (c–d) con-
of song stimulus in relation to body size and body condition index. sist of medians, 25th and 75th percentiles, and min–max whiskers.
Scatterplots (a–b) show the number of flyovers in each male (dots), The significance level of Tukey’s post hoc test is shown: * p < 0.05

other groups (Tukey’s post hoc test, p < 0.05 in all pair-wise
comparison: Fig. 3c).
For body condition index, generally the same patterns
were observed. However, the effect of body condition index
was weaker than that of body size as it was significant only
for the number of flyovers (Supp. Table 1, Fig. 3).
Body size and body condition did not depend on male
age (Mann–Whitney U test, p > 0.05). We found that 1 year-
old males tended to perform more flyovers during playback
presentation (Supp. Table 1, Fig. 4).

Repeatability of male’s size and mass between years

The GLMs showed that wing length, bill length, tarsus

length and body mass were highly repeatable between years
(Supp. Table 3). Then, we analyzed whether body mass
obtained in either March or May 2022 was associated with
body mass measured in May 2021. We found that the most
parsimonious model was the one, that included May 2022
Fig. 4  Number of flyovers observed during presentation of song
data set rather than March 2022 data set among predictors stimulus in relation to male’s age. Boxes consist of medians, 25th and
(Supp. Table 3). Therefore, measurements obtained in May 75th percentiles, and min–max whiskers

13
Journal of Ethology (2023) 41:185–193 191

appeared to be more repeatable between years than March evidence that individuals show consistent inherited differ-
measurements. ences along this axis of behavioral variation.
The heritability of aggressiveness is not necessarily abso-
lute, including in our study species. First of all, aggression is
Discussion subject to high within-individual variation as a consequence
of plastic responses to previous fight outcomes and oppo-
The main finding of this study was that bigger male nut- nent traits (Briffa et al. 2015). Further, many other factors
hatches reacted more aggressively to playback-simulated may influence aggressiveness. The male age could be one
territorial intrusion in terms of physical displays. They per- of such factors. In agreement, we found that 1-year-old male
formed more flyovers, came closer to the loudspeaker, and nuthatches tended to perform more flyovers during playback
gave the threat display more frequently than smaller indi- presentation, i.e., they behaved more aggressively than older
viduals. The same patterns were revealed in the analysis of males.
body condition. However, the effect of this factor was much Aggressive interactions have been found to be age-related,
weaker. We also found that body size and body condition with older individuals being more aggressive than younger
did not depend on male age. More importantly, body size ones (Bosè and Sarrazin 2007; Edler and Friedl 2010; Young
and body mass appeared to be an individual-specific traits et al. 2017). These results are congruent with the hypotheses
as they were repeatable between years. Taking together, that the relationship between age and aggressiveness could
we assumed that both aggressiveness and body size, being be due to life experience or due to older individuals hav-
related to each other, arise due to genetic effects or as a result ing a lower reproductive value, taking higher risks (Wolf
of irreversible developmental plasticity. et al. 2007). However, the opposite pattern was discovered
There are three kinds of evidence that aggressiveness in this study, as well as in some other species where younger
is inheritable and/or influenced by the condition early in individuals were found to be more aggressive than older
life (irreversible developmental plasticity). First, one can conspecifics. Subadult collared flycatcher males behave
directly test whether there is an association between the more aggressively than adult males when confronted with
aggressiveness levels of parents and their descendants. We a conspecific intruder (Szász et al. 2019). In the great tit,
are not aware of any studies that directly measured the herit- males become less aggressive with increasing age (Araya-
ability of aggressive behavior in birds. However, observa- Ajoy and Dingemanse 2017). It was suggested that younger
tions of the social status of daughters and mothers within individuals might require higher levels of aggressive inter-
six strains of domestic chicken (Gallus gallus) revealed that actions to have access to resources, since older and more
differences in aggressiveness among the strains were largely experienced individuals may monopolize them. Therefore,
determined by hereditary differences (Komai et al. 1959). subadult males should be more aggressive to be successful
Secondly, within-individual repeatability suggests the because they are socially inexperienced (Garamszegi et al.
heritability and/or irreversible developmental plasticity of 2006; Szász et al. 2019).
the corresponding behavior (Araya-Ajoy and Dingemanse We revealed no significant association between acoustic
2017). The repeatability of aggressiveness in various con- response and male age and/or body size. It is known that bird
texts has been documented in a wide range of taxa, including acoustic behaviors depend on genetic effects and are subject
passerine birds (reviewed by Bell et al. 2009; Garamszegi to developmental plasticity. First, learned bird song is influ-
et al. 2013; Briffa et al. 2015). enced by inherited predispositions (Beecher and Brenowitz
Thirdly, a positive association between a heritable trait 2005; Mundinger and Lahti 2014). Secondly, talking about
and aggressiveness suggests the heritability of the latter. development plasticity, one should consider the ‘develop-
Body size is interesting from a genetic as well as from an ment stress hypothesis’. The studies showed that the song
ecological point of view. It was found that a considerable control pathway developed during several early weeks of
part of the total variation of body size is caused by genetic life. If the conditions during this period were not optimal,
differences in the great tit (van Noordwijk et al. 1988), blue the vocal equipment did not develop appropriately (Nowicki
tit (Dhondt 1982), Darwin’s medium ground finch (Geospiza et al. 2002). This, in turn, could affect the song in adults; for
fortis) (Boag and Grant 1978), and song sparrow (Melospiza example, a male exhibiting high song-type diversity could
melodia) (Smith and Zach 1979). In this study, we found be of high quality (Buchanan et al. 2003). It is also known,
body size did not depend on age in nuthatches and individu- that repertoire size increases with age in several songbird
ally repeatable between years. Based on this, we assumed species (Kipper and Kiefer 2010). The nuthatch vocalization
that this is at least a moderately inheritable trait in our study is quite simple. This could be one of the reasons why we did
species. We found a positive association between body size not find any dependencies between male acoustic response
and aggressiveness. Hence, our study added to accumulating and body size and/or age. One more reason could be that

13
192 Journal of Ethology (2023) 41:185–193

behavioral rather than acoustic response plays a larger role Ens BJ, Kersten M, Brenninkmeijer A, Hulscher JB (1992) Territory
in male nuthatches’ territorial behavior. quality, parental effort and reproductive success of oystercatchers
(Haematopus ostralegus). J Anim Ecol 61:703–715
Supplementary Information The online version contains supplemen- Funghi C, Leitão AV, Ferreira AC, Mota PG, Cardoso GC (2015)
tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 10164-0​ 23-0​ 0784-3. Social dominance in a gregarious bird is related to body size but
not to standard personality assays. Ethology 121:84–93
Acknowledgements The study was supported by the Russian Science Garamszegi LZ, Rosivall B, Hegyi G, Szöllősi E, Török J, Eens M
Foundation (grant number 22-24-00001). (2006) Determinants of male territorial behavior in a Hungarian
collared flycatcher population: plumage traits of residents and
Author contributions Both authors contributed to the study concep- challengers. Behav Ecol Sociobiol 60:663–671
tion and design. Data collection was performed by Mikhail Diatroptov. Garamszegi LZ, Markó G, Herczeg G (2013) A meta-analysis of cor-
Alexey Opaev analyzed the data and wrote manuscript. Both authors related behaviors with implications for behavioral syndromes:
read and approved the final manuscript. relationships between particular behavioral traits. Behav Ecol
24:1068–1080
Data availability Dataset and audio-recordings are available upon Gosling SD (2001) From mice to men: what can we learn about per-
request. sonality from animal research? Psych Bull 1:45–86
Hau M, Goymann W (2015) Endocrine mechanisms, behavioral phe-
notypes and plasticity: known relationships and open questions.
Declarations Front Zool 12:S7
Hatzinger R, Hornik K, Nagel H, Maier MJ (2014) R: Einführung
Conflict of interest None. durch angewandte Statistik, 2nd edn. Pearson Studium, München
Jenni L, Winkler R (2011) Moult and ageing of European passerines.
Ethical approval All applicable international, national, and/ institu- Christopher Help Publishers, London
tional guidelines for the care and use of animals were followed. All Kassambara A, Mundt F (2017) Factorextra: extract and visualize the
work complied in accordance with the Federal law of Russian Federa- results of multivariate data analyses. R package version 1.0.5.
tion No. 498-Ф3 ‘On responsible treatment of animals’. https://​CRAN.R-​proje ct.org/packa ge=facto extra
Kaiser HF (1974) An index of factorial simplicity. Psychometrika
39:31–36
Kipper S, Kiefer S (2010) Age-related changes in birds’ singing styles:
on fresh tunes and fading voices? Adv Stud Behav 41:77–118
References Komai T, Craig JV, Wearden S (1959) Heritability and repeatability
of social aggressiveness in the domestic chicken. Poultry Sci
Araya-Ajoy YG, Dingemanse NJ (2017) Repeatability, heritability, and 38:356–359
age- dependence of seasonal plasticity in aggressiveness in a wild Labocha MK, Hayes JP (2012) Morphometric indices of body condi-
passerine bird. J Anim Ecol 86:227–238 tion in birds: a review. J Ornithol 153:1–22
Beecher MD, Brenowitz EA (2005) Functional aspects of song learning Landys MM, Goymann W, Schwabl I, Trapschuh M, Slagsvold T
in songbirds. Trends Ecol Evol 20:143–149 (2010) Impact of season and social challenge on testosterone and
Bell AM, Hankison SJ, Laskowski KL (2009) The repeatability of corticosterone levels in a year-round territorial bird. Horm Behav
behaviour: a meta-analysis. Anim Behav 77:771–783 58:317–325
Beltrão P, Godinho C, Lourenço R, Pereira PF (2021) Behavioural Lank DB, Smith CM, Hanotte O, Burke T, Cooke F (1995) Genetic
repeatability in Sardinian warblers (Sylvia melanocephala): larger polymorphism for alternative mating behaviour in lekking male
individuals are more aggressive. Acta Ethol 24:31–40 ruff Philomachus pugnax. Nature 378:59–62
Bensch S, Hasselquist D (1991) Territory infidelity in the polygynous Lenda M, Maciusik B, Skórka P (2012) The evolutionary, ecological
great reed warbler Acrocephalus arundinaceus: the effect of vari- and behavioural consequences of the presence of floaters in bird
ation in territory attractiveness. J Anim Ecol 60:857–871 populations. North-West J Zool 8:394–408
Bosè M, Sarrazin F (2007) Competitive behaviour and feeding rate in Maier MJ (2015) Companion package to the Book ‘R: Einfuhrung
a reintroduced population of Griffon Vultures Gyps fulvus. Ibis durch angewandte Statistik’. R package version 0.9.3. http://​
149:490–501 CRAN.R-​proje ct.org/package=REdaS
Briffa M, Sneddon LU, Wilson AJ (2015) Animal personality as a cause Marra PP (2000) The role of behavioral dominance in structuring pat-
and consequence of contest behaviour. Biol Lett 11:20141007 terns of habitat occupancy in a migrant bird during the nonbreed-
Boag PT, Grant PR (1978) Heritability of external morphology in Dar- ing season. Behav Ecol 11:299–308
win’s finches. Nature 274:793–794 Matthysen E (1997) Geographic variation in the occurrence of song
Buchanan KL, Spencer KA, Goldmith AR, Catchpole CK (2003) Song types in Nuthatch Sitta europaea populations. Ibis 139:102–106
as an honest signal of past developmental stress in the European Matthysen E (1998) The Nuthatches. T and A. D, Poyser, London
starling (Sturnus vulgaris). Proc R Soc Lond B 270:1149–1156 Morinay J, Daniel G, Gustafsson L, Doligez B (2019) No evidence for
Catchpole CK, Slater PJB (2008) Bird song: biological themes and behavioural syndrome and genetic basis for three personality traits
variations, 2nd edn. Cambridge University Press, Cambridge in a wild bird population. Anim Behav 153:69–82
Diatroptov EV, Diatroptov ME (2017) The annual cycle of change Mundinger PC, Lahti DC (2014) Quantitative integration of genetic
of body weight in the nuthatch Sitta europaea. Rus J Ornithol factors in the learning and production of canary song. Proc R Soc
26:713–725 ((in Russian)) B 281:20132631
Dhondt AA (1982) Heritability of blue tit tarsus length from normal Naďo L, Kašová M, Krištín A, Kaňuch P (2018) Cooperative nest-
and cross-fostered broods. Evolution 36:418–419 defence behaviour and territory quality in a resident and socially
Edler AU, Friedl TWP (2010) Plumage colouration, age, testosterone monogamous passerine. Ethology 124:514–526
and dominance in male red bishops (Euplectes orix): a laboratory Nowicki S, Searcy WA, Peters S (2002) Brain development, song learn-
experiment. Ethology 116:806–820 ing and mate choice in birds: a review and experimental test of the
“nutritional stress hypothesis.” J Comp Physiol A 188:1003–1014

13
Journal of Ethology (2023) 41:185–193 193

R Development Core Team (2020) R: A language and environment van Noordwijk AJ, van Balen JH, Scharloo W (1988) Heritability of
for statistical computing. R Foundation for Statistical Comput- body size in a natural population of the Great Tit (Parus major)
ing, Vienna and its relation to age and environmental conditions during
Réale D, Reader SM, Sol D, McDougall PT, Dingemanse NJ (2007) growth. Genet Res Camb 51:149–162
Integrating animal temperament within ecology and evolution. Williams LJ, King AJ, Mettke-Hofmann C (2012) Colourful charac-
Biol Rev 82:291–318 ters: head colour reflects personality in a social bird, the Gouldian
Salazar SM, Camacho García JI, Kuspiel M, Fokkema RW, Komdeur finch. Erythrura Gouldiae Anim Behav 84:159–165
J, Korsten P (2021) Male aggressiveness and risk-taking during Wolf M, van Doorn GS, Leimar O, Weissing FJ (2007) Life-history
reproduction are repeatable but not correlated in a wild bird popu- trade-offs favour the evolution of animal personalities. Nature
lation //. Behav Ecol Sociobiol 75:108 447:581–584
Schamber JL, Esler D, Flint PL (2009) Evaluating the validity of using Young CM, Cain KE, Svedin N, Backwell PRY, Pryke SR (2017)
unverified indices of body condition. J Avian Biol 40:49–56 Predictors of aggressive response towards simulated intruders
Sergio F, Blas J, Hiraldo F (2009) Predictors of floater status in a long- depend on context and sex in Crimson Finches (Neochmia phae-
lived bird: a cross-sectional and longitudinal test of hypotheses. J ton). Behav Proc 138:41–48
Anim Ecol 78:109–118 Zeileis A, Kleiber C, Jackman S (2008) Regression models for count
Shonfield J, Taylor RW, Boutin S, Humphries MH, McAdam AG data in R. J Stat Soft 27(8). http://​www.​jstat​soft.​org/​v27/​i08
(2012) Territorial defence behaviour in red squirrels is influenced
by local density. Behaviour 149:369–390 Publisher's Note Springer Nature remains neutral with regard to
Smith JNM, Zach R (1979) Heritability of some morphological char- jurisdictional claims in published maps and institutional affiliations.
acters in the Song Sparrow. Evolution 33:460–467
Szász E, Jablonszky M, Krenhardt K, Markó G, Hegyi G, Herényi M, Springer Nature or its licensor (e.g. a society or other partner) holds
Laczi M, Nagy G, Rosivall B, Szöllősi E, Török J, Garamszegi LZ exclusive rights to this article under a publishing agreement with the
(2019) Male territorial aggression and fitness in collared flycatch- author(s) or other rightsholder(s); author self-archiving of the accepted
ers: a long-term study. Sci Nature 106:11 manuscript version of this article is solely governed by the terms of
Svensson L (1992) Identification guide to European passerines, 4th such publishing agreement and applicable law.
edn. Fingraf AB, Stockholm

13