Content uploaded by Raine Kortet
Author content
All content in this area was uploaded by Raine Kortet
Content may be subject to copyright.
ORIGINAL PAPER
Boldness as a consistent personality trait in the noble
crayfish, Astacus astacus
Anssi Vainikka &Markus J. Rantala &Petri Niemelä &
Heikki Hirvonen &Raine Kortet
Received: 20 January 2010 /Revised: 30 September 2010 /Accepted: 25 October 2010 /Published online: 13 November 2010
#Springer-Verlag and ISPA 2010
Abstract Consistent individual differences in behavioural
responses to perceived predation risk may have extensive
ecological and evolutionary implications. We studied the
repeatability of boldness across time and its relation to
resource holding potential in the noble crayfish, Astacus
astacus L., using predator-naïve immature individuals. We
followed individual’s shelter use both with and without
exposure to the chemical and physical cues of predators,
and with and without the presence of a conspecific. In
addition, we examined if armament, i.e. relative chelae size,
would be correlated with individual differences in behav-
iour. Individuals showed repeatable behaviours across time
and context. Individuals that occupied the shelter in
competitive dyadic tests also spent more time in the shelter
during individual control observations, suggesting that
boldness is a personality trait that does not necessarily
relate positively to high resource holding potential in the
noble crayfish. The relative size of chelae did not correlate
with any of the measured behavioural variables. Our results
suggest that boldness can be considered as individually
consistent and ecologically important personality trait in the
noble crayfish.
Keywords Antipredatory behaviour .Behavioural
syndrome .Decapoda .Domestication effect .Life history .
Repeatability
Introduction
Perceived predation risk is considered as important as direct
predation mortality for the function of ecological systems
(Preisser et al. 2005). Perceived risk of predation affects
foraging decisions and, therefore, alters the form and
strength of trophic interactions in many species (e.g. Lima
and Dill 1990; Lima and Bednekoff 1999; Carvalho and
Del-Claro 2004;Nyström2005). However, individual
animals show repeatable differences in their sensitivity to
predation risk. This suggests the existence of individual
coping styles, animal personalities or behavioural syn-
dromes among individuals (Dall et al. 2004; Bell 2007; Sih
and Bell 2008; Stamps and Groothuis 2010). By the
definition of Stamps and Groothuis (2010), the term
behavioural syndrome refers to behaviours that are corre-
lated across time or contexts. Here we use term personality
trait to refer to one type of behaviour that is temporally
repeatable and consistent across contexts. Animal person-
alities, which consist of several personality traits, are
potentially important in the function of food webs by
inducing behaviourally mediated cascade effects (e.g.
Ioannou et al. 2008). Also, individual variation in person-
ality traits is potentially important in behavioural adaptation
to predation (Dall et al. 2004; Sih et al. 2004; Réale et al.
2007). For example, consistently bold individuals may gain
fitness advantages in low predation risk environments,
A. Vainikka (*):P. Niemelä
Department of Biology, University of Oulu,
P.O. Box 3000, 90014, Oulu, Finland
e-mail: anssi.vainikka@oulu.fi
M. J. Rantala
Department of Biology, University of Turku,
20024, Turku, Finland
H. Hirvonen
Department of Biosciences, University of Helsinki,
P. O. Box 65, 00014, Helsinki, Finland
R. Kortet
Department of Biology, University of Eastern Finland,
P.O. Box 111, 80101, Joensuu, Finland
acta ethol (2011) 14:17–25
DOI 10.1007/s10211-010-0086-1
whereas consistently shy individuals may have superior
fitness in high predation risk environments (Sih et al.
2004). Boldness traits have been found to be heritable in
some (Sinn et al. 2006; Brown et al. 2007; Dingemanse et
al. 2009) but not all species (e.g. Riesch et al. 2009).
Questions pertaining to the development of behavioural
syndromes become particularly interesting when intraspe-
cific behavioural interactions, such as dominance relation-
ships, are considered. The most well-known behavioural
syndrome is the aggressiveness–boldness syndrome in
which individuals that behave boldly under the risk of
predation are also more aggressive towards conspecifics
(Huntingford 1976; Riechert and Hedrick 1993; Kortet and
Hedrick 2007). Such syndromes can be especially impor-
tant in socially hierarchical species such as crayfishes. For
example, dominance hierarchies affect food uptake levels of
individual crayfish (Ahvenharju and Ruohonen 2006). In
addition, crayfishes are dependent on the availability of
burrows, which provide shelter, and often compete aggres-
sively for them (Ranta and Lindström 1993; Garvey et al.
1994; Figler et al. 1999). When several behaviours in
different contexts promote high food uptake rates, behav-
ioural syndromes may become heritable through pleiotropy
or genetic correlations (Stamps 2007; Biro and Stamps
2008). Therefore, high food uptake rates in crayfish, if
promoted by both boldness and aggressiveness, would be
expected to result in the development of an aggressiveness–
boldness–behavioural syndrome. Supporting results have
been obtained in the signal crayfish Pacifastacus leniuscu-
lus Dana, in which aggression, voracity and boldness were
all positively correlated (Pintor et al. 2008). However, it is
not known if shelter possession, potentially achieved
through aggressive behaviour (e.g. Figler et al. 1999),
would be coupled with boldness in the noble crayfish,
Astacus astacus L.
Many aquatic organisms, including crayfishes, can detect
the presence of predators through chemical cues (Appelberg
et al. 1993; Blake and Hart 1993; Kats and Dill 1998). The
noble crayfish is a species native to Finland and has a long
co-evolutionary history with fish predators such as the
Eurasian perch, Perca fluviatilis L. Predator-naïve juvenile
noble crayfish increase shelter use and reduce feeding
activity when exposed to the chemical cues from predatory
fishes such as the perch, the pike, Esox lucius L., the burbot
Lota lota L. or the European eel, Anguilla anguilla L.
(Appelberg et al. 1993; Hirvonen et al. 2007). In this study,
perch odour and physical disturbance were used to induce/
mimic predation risk.
Large armaments may facilitate the establishment of a
strong position in dominance hierarchies (Berglund et al.
1996). Relative size of the chelae is known to affect the
success of crayfish in aggressive encounters (Rutherford et
al. 1995; Bywater et al. 2008), and a large difference in
chelae size between males helps to resolve dominance
relationships without engaging in fights (Schroeder and
Huber 2001). However, little is known about how person-
ality traits, such as boldness or aggressiveness, relate to the
size of the chelae. For example, large armaments could
increase resource acquisition of an individual and favour
bold behaviour and consequently fast growth by reducing
the frequency of costly fights (cf. Stamps 2007). This could
in turn maintain a positive correlation between armament
size and bold behaviour via a positive feedback loop on
resource acquisition (Luttbeg and Sih 2010).
In this study, we focused on the temporal repeatability
and across-context consistency of individual shelter use that
we considered as a measure of boldness or resource holding
potential, depending on the context. We assumed that in the
absence of conspecifics but in the presence of food outside
the shelter, low shelter use reflects high boldness. On the
contrary, we assumed that in the presence of predation risk
and a conspecific but in the absence of food, high shelter
use reflects high resource holding potential and involves
interference competition. However, we did not study
aggressiveness or explicit dominance relationships between
the individuals. Therefore, even in the latter case, low
shelter use may reflect high boldness. We performed
experiments in two contexts in order to understand the role
of boldness as a potentially evolvable personality trait. In
addition, to evaluate if the size of armaments relates to
behaviour, we explored correlations between the relative
size of the chelae and individual’s behaviour. Based on
comparative data on other crayfish species (Pintor et al.
2008), we predicted that bold behaviour should be
consistent across time, and with and without predator
stress, i.e. form a behavioural syndrome according to the
definition of Stamps and Groothuis (2010). We predicted
that individuals with relatively larger chelae would be
bolder but also have a better ability to possess the shelter
under competition compared to individuals with relatively
smaller chelae.
Materials and methods
Study animals
Two- to three-year-old immature noble crayfish, raised in a
predator-free environment, were obtained from a commer-
cial crayfish producer in Southern Finland in August 2008.
The farmed population had been cultured for several
generations and originated from a large number of wild
noble crayfish specimens. Upon arrival in the laboratory,
the crayfish were individually numbered using a white
marker pen (Textmark 250) and then moved to individual
containers measuring 105 mm (width) × 145 mm (length) ×
18 acta ethol (2011) 14:17–25
230 mm (height of water) built in 300-l tanks. Crayfish
were fed ad libitum with carrot, alder leaves and periodi-
cally fresh fish and shrimps. The light–dark rhythm (lights
on 07:00–17:00) and water temperature (10 ± 1
○
C) were
kept constant. Carapace length (to the tip of the rostrum, to
the nearest 0.1 mm), mass (to the nearest 0.1 g), and length
and width of the chelae (to the nearest 0.01 mm) of the
study animals were measured on the 10th of February 2009.
No moults occurred during the course of experiments,
which were conducted between March and June of 2009.
Overview of tests
First, each crayfish was individually observed for both
latency to emerge from a shelter and total activity levels.
This was done for a period of 3 days in an environment
with no predators or conspecifics (Fig. 1a). Second,
crayfish were stratified by sex and then randomly allocated
between three treatments: (1) control, (2) predator odour
and (3) predator odour + general stress. Individuals were
exposed to their respective treatments for a period of 3 days,
during which behavioural observations were made. Third,
the crayfish were divided to sex- and size-matched pairs,
and observed for shelter possession for 3 days (Fig. 1b).
Between the boldness tests and shelter possession trials, the
crayfish were treated according to the group assignments
for additional 18 days. At the end of this period, the
crayfish were let to encapsulate a 6-mm-long, 0.20 mm in
diameter nylon implant placed through a small puncture in
the first joint of the right cheliped for 7 days. However, the
results of the encapsulation test, assumed to have no
importance for later behavioural tests, are not reported here.
Boldness tests
In order to examine both the effects of perceived predation
risk on the behaviour of crayfish and the consistency of
individual behaviour, 46 females [body mass 7.2±3.7 g
(mean±SD), carapace length 30.3±4.6 mm] and 35 males
(body mass 8.6±3.5 g, carapace length 32.2±4.0 mm) were
used. Each trial period lasted for a total of 6 days. On day 1,
nine crayfish were randomly placed into individual housing
compartments (105×145 mm, water depth 210 mm),
including a gray plastic tube 75 mm in length and 36 mm
in inner diameter and a piece of carrot within the control
tank (Fig. 1a). The control tank had a separate closed water
circulation. After 3 days of individual housing, the crayfish
were randomly allocated in groups of three (two females
and one male or two males and one female) and then
transferred to their respective treatment tanks (Fig. 1a). The
two tanks used for the predator stress treatments were
connected to a tank with 23 wild-caught perch (mean total
length 177 mm, SD 26.6 mm, fed regularly with frozen
shrimps), which functioned as the source of predator odour.
Constant water flow between the tanks was 60 lmin
−1
. The
control crayfish were transferred to other individual
housing compartments within the separate control tank to
guarantee equal levels of stress induced by the handling
procedure in all treatments. Physical disturbance treatments
involved catching and holding the crayfish individuals
above the water level for 10 s every 2 h (8:00–16:00) over
five successive days (Monday–Friday). Crayfish had access
to an excess of food (carrot) during the trials (Fig. 1a).
The crayfish were videotaped using infrared cameras for
3.5 h starting at 16:30, i.e. a half an hour before the lights
went off. During daytime, crayfish most often burrow
themselves in the shelter and emerge after darkness falls.
Therefore, video recordings were analysed for the latency
to emerge from the shelter (time until emergence) and for
the total time the crayfish spent outside the shelter during
the dark period (total exposure time). Both of these
measures reflect boldness but are not interchangeable since,
on average, individuals emerged (and correspondingly went
Fig. 1 Schematic presentation of the boldness tests (a) and shelter
possession tests (b). First, nine crayfish were observed for 3 days
without a predator odour and then distributed to new tanks and
compartments corresponding the treatments (control, predator odour,
odour+ disturbance). All individuals had access to a shelter (rectangle)
and a piece of carrot (circle). In shelter possession tests (b), there was
a shelter for only one crayfish at time, and the two individuals placed
in one arena competed for the shelter when exposed to the predator
odour
acta ethol (2011) 14:17–25 19
back to the tube) 2.7 times during the observation period
(an individual does not stay out once emerged). Due to a
bug in the recording software (InterVideo WinDVR5,
InterVideo Inc.), we lost data for six observation periods
(see ‘Results’section). However, individuals with missing
data were still used for analyses deploying data from
successfully observed periods (in correlation analyses).
Shelter possession trials
In order to study if boldness was related to shelter
possession, the remaining crayfish were assigned to
weight-matched pairs and then subjected to a resource
possession trial. There were 14 female pairs and nine male
pairs, with a mean weight difference of 5.6% between
paired individuals. Past exposure treatment was not taken
into account when assigning pairs (11 pairs had the same
past exposure history). Either the left or right chela was
painted white with a water-insoluble ink for identification.
Each pair was placed in 290 mm× 105 mm × 210 mm (water
level) testing arena for 4 days (Fig. 1b). As shelter, the
arena contained one grey plastic tube 75 mm in length and
36 mm in inner diameter in one end of the arena, placed in
a way so that only one of the crayfish could use it at a time
(Fig. 1b). A similar kind of arena has been used previously
to study the resource holding potential of signal crayfish
(Ranta and Lindström 1993). The arenas were build in one
300-l tank that received water (60 l/min) from another tank
containing 26 perch (the same as in boldness tests,
complemented with three small individuals, length 172.2 ±
28.4 mm, mean±SD). The crayfish were not fed during the
trials. Each arena was videotaped from 7:00 to 9:00 (the
time crayfish naturally return to a shelter) at low light
conditions for 3 days using digital infrared cameras. The
crayfish inside the tube and outside the tube were later
identified using the recordings. At day 4, prior to the
removal of crayfish, we noted identities of the resource
holder and the individual outside of the tube via visual
inspection. The individual found outside of the tube more
often (>50% of occasions) than the other was judged to be
the potentially subordinate individual within each pair
(Ranta and Lindström 1993).
Statistical analyses
The three tanks containing the individual compartments
were equivalent to the three treatments that were applied
during the boldness test. Common holding conditions
(potential chemical cues between individuals) may induce
statistical dependence between individuals. We controlled
for this by entering week (corresponding to the group
effect, as groups were changed every week) as a nesting
factor in the repeated measures ANOVA (RM-ANOVA),
which was used to analyse the effects of predation risk
treatments on the behaviour of crayfish. Period (control
observation vs treatment period, each lasting for 3 days)
and day within the period were included as repeated
measures (within-subject factors). Interactions between
these were omitted in order to avoid overparameterisation
of the model. Treatment, nested by week, sex and the
interaction sex×treatment were entered as between-subject
factors. Two models were run: in the first model, time until
emergence was the dependent variable; in the second
model, the total time spent out of the tube was the
dependent variable. Test statistics were based on Wilk’s
Λ, and Bonferroni post hoc comparisons were used to
determine the source of variation using an all-pairwise
comparison approach.
In order to study the consistency of individual behaviour,
an analysis of repeatability was used (Krebs 1999; Bell et
al. 2009). Statistically significant repeatability of individual
behaviour implies that it is a property of an individual and
not entirely dependent on the situation (Bell et al. 2009).
For a correlation analysis between trait expression in
morphological and behavioural characters, we calculated
individual means for the three behavioural measurements
during the control and the treatment period. These means
were used as estimates of individual boldness and were
examined for correlations with the relative (linear to
carapace length) size of the chelae. Correlations were
explored using Pearson’s correlation analysis (correlation
coefficient as r) and partial Pearson’s correlation analysis
controlling for the effect of carapace length (correlation
coefficient as r
c
). RM-ANOVAs were used to test for sex-
specific differences between the shelter possessor and the
loser. Pearson’sχ
2
test was used to test if shelter possession
was related to the past exposure to the predation risk
treatments. Normality of the studied parameters was
checked using Kolmogorov–Smirnov test and the require-
ment of homoscedasticity was confirmed using Levene’s
test. All GLM, χ
2
and correlation analyses were performed
in SPSS 16.0.1 (SPSS Inc., USA).
Results
Boldness tests
According to multivariate RM-ANOVA results, only the
day within the treatment period explained variation in the
time until emergence (Table 1). Univariate within-subject
effects did not differ from multivariate effects. However,
between-subjects effects of sex, treatment group (nested by
week) and sex×treatment group (nested by week) were
significant (Table 1, Fig. 2). There was a distinct difference
between the sexes in emergence time across the control and
20 acta ethol (2011) 14:17–25
treatment period: males emerged significantly later than
females over all treatments, with the largest difference
occurring in the predator+disturbance group (Fig. 2). With
the exception of the period between days 1 and 2, there was
a general and significant increase in the mean time until
emergence after each transfer of individuals [mean time at
emergence—12.9±6.3 min (mean±SE), 23.3±5.7 min and
39.4±6.6 min for days 1, 2 and 3, respectively].
Similarly, variation between days was the only factor
contributing to the total exposure time (Table 1). Again,
univariate within-subject effect results were in line with the
multivariate results. None of the between-subjects effects
was significant (Table 1). There was a general decrease in
the exposure time after each transfer to a new location,
indicating reduced activity, although only the difference
between days 2 and 3 was significant [102.9 ± 6.5 min
(mean± SE), 102.6±7.7 min and 82.4±6.5 min, respectively
for days 1, 2 and 3]. Mean exposure time decreased slightly
from the control period to the treatment period in all treatment
groups, although this difference was not significant (Fig. 3).
Consistency (repeatability) of individual behaviour
Since the treatments did not influence individual behaviour
relative to individual behaviour during control observa-
tions, the treatment groups were pooled for subsequent
analyses. Time until emergence was repeatable between the
days of the control period (R= 0.21, N=72, P= 0.001), the
treatment period (R=0.17, N=70, P= 0.009) and across the
whole experiment (R=0.16, N=62, P<0.001), indicating
that there are consistent individual differences in this
behaviour. Also, the total exposure time was repeatable
during the control period (R= 0.28, N=72, P<0.001) but
not during the treatment period (R=0.00, N=70, P=0.467).
However, across the whole period, the total exposure time
was weakly repeatable indicating individual consistency in
activity (R=0.10, N=62, P=0.004).
Table 1 Results of repeated measures ANOVA for the time until
emergence from the shelter and the total exposure time
Source of variation Fdf Pvalue η
2
Time until emergence from the shelter
MV Period 1.85 1, 21 0.188 0.081
MV Period×sex 1.85 1, 21 0.188 0.081
MV Period×treatment 1.23 20, 21 0.320 0.540
MV Day 4.14 2, 20 0.031 0.293
MV Day×sex 0.88 2, 20 0.432 0.081
MV Day× treatment 0.86 40, 40 0.686 0.461
MV Day× sex × treatment 0.61 38, 40 0.937 0.366
B-S Sex 9.23 1, 21 0.006 0.305
B-S Treatment by week 3.06 20, 21 0.007 0.744
B-S Sex×treatment by week 2.27 19, 21 0.036 0.672
Total exposure time
MV Period 3.53 1, 21 0.074 0.144
MV Period×sex 0.46 1, 21 0.505 0.021
MV Period×treatment 1.23 20, 21 0.319 0.540
MV Day 4.38 2, 20 0.025 0.307
MV Day×sex 0.12 2, 20 0.887 0.012
MV Day× treatment 0.84 40, 40 0.713 0.455
MV Day× sex × treatment 0.99 38, 40 0.511 0.485
B-S Sex 0.20 1, 21 0.659 0.009
B-S Treatment by week 1.10 1, 20 0.412 0.512
B-S Sex×treatment by week 0.98 1, 19 0.518 0.469
Test statistics are based on Wilk’sΛ.MV refers to a multivariate effect
and B-S refers to a between-subject effect. Partial η
2
refers to the
proportion of the total variance that is attributable to an effect. Period
(including 3 days) refers to the effect of treatments compared to control
observations (first three control days, then 3 days under treatments). If
treatments have an effect on behaviour, interaction between period and
treatment is expected to be significant
0
10
20
30
40
50
60
70
disturbance
Time until emergence ± S.E. (min)
Females
Males
Predator odour
Control Odour +
Fig. 2 Time until emergence from the shelter in female and male
noble crayfish in different treatment groups (control, predator odour,
predator odour and disturbance) across control and treatment periods
(estimated marginal means from the RM-ANOVA)
0
20
40
60
80
100
120
140
160
180
Treatment
Total exposure time ± S.E. (min)
Before
After
Predator odour Odour + disturbance Control
Fig. 3 Total exposure times during the control and treatment period
(before/after the introduction of predator risk) in each treatment group
(estimated marginal means from the RM-ANOVA). An observation
period lasted 180 min
acta ethol (2011) 14:17–25 21
Correlations among behaviours and relative chelae size
The longer it took for the crayfish to come out from the
shelter, the less time it spent outside of the shelter during
the control period (r=−0.72, N=72, P<0.001) and during
the treatment period (r=−0.65, N=79, P<0.001). However,
the mean times until emergence between the control and
treatment periods were not correlated across treatments (r=
0.20, N=71, P=0.090). The mean exposure times between
the control and the treatment periods were also uncorrelated
across treatments (r=0.21, N=71, P= 0.076). Relative
length or width of the chelae were not related to the
behavioural variables measured during the control or the
treatment period (r
c
=−0.073–0.066, N=68, P≥0.546, effect
of carapace length partialled out).
Shelter possession trials
In 16 trials out of the 23 total trials, the same individual
of the two crayfish was outside the tube at every
observation. The individual that was out of the tube
during a greater proportion of the observations was
denoted as the loser of the resource holding competition.
In two of the trials, an individual was found dead outside
of the tube and was subsequently classified as a loser. On
three occasions, both of the crayfish had burrowed
themselves in the tube, and the winner and loser could
not be identified. Only the behaviour during the control
observations was related to the outcome of the shelter
possession trial (Table 2). The individual that monopo-
lised the shelter had a shorter exposure time and emerged
later from the shelter when observed during the control
period. There were no sex-related effects in shelter
possession (Table 2). χ
2
test did not indicate any effects
arising from the exposure to the different predation risk
treatments (χ
2
=1.90, df=4, P=0.754).
Discussion
Despite large between-days variation, individual crayfish
showed repeatable behaviours within and across periods of
varying risk of predation, which suggests that boldness can
be considered as a personality trait in the noble crayfish. In
addition, behaviour observed during the control period was
consistent with the behaviour observed in the shelter
possession trials: individuals behaving shyly in the absence
of predation risk possessed the shelter more often than their
rivals. This suggests that resource holding potential, defined
as shelter possession under interference competition (Ranta
and Lindström 1993), and bold behaviour are negatively
associated in the studied population of the noble crayfish.
Since we did not observe the formation of shelter
possession, it is unclear if shelter possession required
dominance or was associated with aggressiveness. Conse-
quently, we cannot either assess if aggressiveness or
dominance would form a behavioural syndrome with
boldness in the noble crayfish (cf. Huntingford 1976;
Riechert and Hedrick 1993; Kortet and Hedrick 2007).
Behavioural syndromes and animal personalities have
been observed in a wide variety of taxa (e.g. Huntingford
1976; Dall et al. 2004; Wilson and McLaughlin 2007; Bell
et al. 2009). One potential mechanism promoting the
evolution of behavioural syndromes is the coupling
between growth rate and mortality (Stamps 2007). This
mechanism implies that behaviours that promote high food
intake rates may become genetically coupled over evolu-
tionary time, and that the temporal consistency in person-
ality traits is maintained by the cost of deviating from
intrinsic growth trajectories (Stamps 2007; Adriaenssens
and Johnsson 2009; Biro and Stamps 2008). This mecha-
nism should favour the development of an aggressive–bold
behavioural syndrome in crayfishes since both aggression
and boldness are likely to increase resource acquisition
Table 2 Comparison of trait means between the shelter possessor (In) and the crayfish that stayed outside the shelter (Out) according to
multivariate results of RM-ANOVA, where sex was entered as a between-subject factor
Trait In Out Shelter possession Sh. possession× sex
FdfPFdfP
Time until emergence (min) 37 6.3 8.74 1, 21 0.008 0.62 1, 21 0.439
Total exposure time (min) 79 110 4.77 1, 21 0.041 0.00 1, 21 0.990
Tr.p. time until emergence 21.4 36.6 0.76 1, 20 0.393 0.013 1, 20 0.911
Tr.p. total exposure time 79.0 90.5 0.67 1, 20 0.421 0.31 1, 20 0.582
Relative length of chelae 0.28 0.28 0.60 1, 21 0.448 0.21 1, 21 0.650
Relative width of chelae 0.67 0.66 0.52 1, 21 0.478 0.001 1, 21 0.972
The pairs were size matched by weight. Length and width of the chelae are relative to carapace length. Tr.p. refers to treatment period. Ftests
based on Wilk’sΛ
22 acta ethol (2011) 14:17–25
rates in these socially hierarchic animals (Ahvenharju and
Ruohonen 2006). In accordance with the mechanism of
state-dependent safety generated by high foraging rate
(Luttbeg and Sih 2010), a positive correlation would have
been expected between boldness and the size of morpho-
logical armaments signalling fighting ability (safety), i.e.
size of the chelae. Against expectations, the relative size of
the chelae did not relate to behavioural measures despite
having been repeatedly reported to relate to dominance rank
in crayfishes (Snedden 1990; Usio et al. 2001).
Resource value is a subjective concept. Therefore, it
might be possible that shy individuals perceived the value
of shelter higher than bold individuals did. Therefore, the
coupling of shelter possession and boldness could indeed
be negative, and conflict with the idea that aggressiveness
and boldness should be positively correlated (see above).
Since the mean difference in the relative chelae size was
rather small and non-significant between the rivals (7.2% in
length and 7.6% in width), it is unlikely that shelter
possession would have been readily resolved by armament
size. However, we did not study aggressiveness or
dominance over a food resource. A recent study in the
signal crayfish suggests that the aggressiveness–boldness
behavioural syndrome could be a result of resource
acquisition dynamics (Pintor et al. 2008). Therefore, more
direct tests of aggressiveness, in relation to competition for
food, would be needed to evaluate the coupling between
resource intake rates, facilitated by bold behaviour in the
presence of predators, and aggressive behaviour in the
presence of intraspecific competition for food. Consequently,
the presence of an aggressiveness/dominance–boldness–
behavioural syndrome in the noble crayfish cannot be
excluded in the context of food acquisition.
However, an alternative mechanism may be proposed to
explain why shelter possession, which in general is thought
to indicate dominance in crayfishes (Figler et al. 1999), was
inversely related to boldness. In our study, predator odour
did not induce changes in the behaviour of crayfish, and
was thus not likely experienced as a threat either in the
shelter possession trials. Therefore, bold individuals could
have been allocating their time to exploring the arena,
potentially in the search of food, rather than engaged in
fights for the shelter, which in the absence of perceived
predation risk could have had low resource value. Anyhow,
shelter use during predation risk treatments did not relate to
the shelter possession, suggesting that the coupling of shy
behaviour and shelter possession cannot be generalised
over all contexts.
The lack of response to predator odour contrasts earlier
studies in the noble crayfish (Appelberg et al. 1993).
Further, in contrast to an earlier study on Ocronectes
propinquus (Stein and Magnuson 1976), females emerged
earlier than males in our study. In the wild, females are
predicted to be more vulnerable to fish predation due to
their smaller chelae (Stein and Magnuson 1976). In our
study, crayfish were clearly older than the crayfish in
previous experiments (Appelberg et al. 1993). Therefore,
they could have habituated to a predator-free environment
and lost their responsiveness to the predator odour by
learning. In addition, the previously documented responses
were rather weak and slightly stronger only when perch
were starved (Appelberg et al. 1993). We fed the perch that
functioned as the source of predatory fish odour regularly
with shrimps, which might have contributed to the absence
of a detectable effect. Also, the perch used in this study
were probably too small to impose a real threat on the size
of intermoult crayfish used in this study. Whether larger
perch would induce a stronger response only by odours
remains to be examined. Another potential explanation for
the lack of clear response relates to the normally nocturnal
activity of the noble crayfish: usually shelter use increases
in daytime and the most pronounced predator-induced
shifts in behaviour are observed in darkness (Appelberg et
al. 1993; Hirvonen et al. 2007). In this study, several
crayfish stayed outside the shelter also during daytime,
which reduced variation in the behavioural variables
studied after switching off the lights. This made it more
difficult to detect a significant effect. However, handling
associated with transfer of crayfish to their treatment tanks
induced an increase in exposure times, indicating that
human-induced disturbances may increase predation risk
for disturbed individuals. This can occur, for example,
during and after release of undersized individuals in
crayfish harvesting.
Response of prey to predator cues may also be
dependent on the temporal presence of predator cues (Lima
and Bednekoff 1999; Sih and McCarthy 2002). In this
study, the predator odour was constantly present for several
days. Presence of perch led to a considerable predation
mortality and shifts in the behaviour of A. astacus in the
experiment of Söderbäck (1992). However, the behavioural
responses did not differ from those of non-native Pacifas-
tacus leniusculus, indicating that the role of learning might
also be considerable. Thus, it is possible that the crayfish in
our study could have rapidly learned that the predators
caused no danger even though their odour was continuously
present.
In conclusion, the noble crayfish did not clearly respond
to predator odour, but instead showed individually repeat-
able hiding behaviour both with and without predator odour
and a size-matched competitor. Therefore, we propose that
boldness can be considered as a personality trait in the
noble crayfish and has the potential for having extensive
ecological and evolutionary implications depending on the
predation risk and intensity of intraspecific competition.
Finally, we would like to promote studies that couple
acta ethol (2011) 14:17–25 23
behaviours and behavioural responses more tightly with
morphological and life-historical traits.
Acknowledgements This research has been supported by the
Academy of Finland (project #127398) and by the Emil Aaltonen
Foundation. We thank the staff of the Experimental Unit of the
Department of Biology, University of Oulu for the maintenance of the
crayfish and help in the practical execution of the experimental
procedures. We thank Aki Puhka for his help in the experiments and
two anonymous referees for the very valuable comments on the
manuscript. We also gratefully acknowledge Nick DiRienzo for his
valuable linguistic revision of the manuscript. The perch in this
experiment were held in laboratory conditions with the permission
ESLH-2009-06035/Ym-23 from ELLA (Finnish board for the use of
animals in experiments).
References
Adriaenssens B, Johnsson JI (2009) Personality and life-history
productivity: consistent or variable association? TREE 24:179–
180. doi:10.1016/j.tree.2008.12.003
Ahvenharju T, Ruohonen K (2006) Unequal division of food
resources suggests feeding hierarchy of signal crayfish (Pacifas-
tacus leniusculus) juveniles. Aquaculture 259:181–189
Appelberg M, Söderbäck B, Odelström T (1993) Predator detection
and perception of predation risk in the crayfish Astacus astacus
L. Nord J Freshw Res 68:55–62
Bell AM (2007) Future directions in behavioural syndromes research.
Proc R Soc Lond B 274:755–761. doi:10.1098/rspb.2006.0199
Bell AM, Hankison SJ, Laskowski KL (2009) The repeatability of
behaviour: a meta-analysis. Anim Behav 77:771–783.
doi:10.1016/j.anbehav.2008.12.022
Berglund A, Bisazza A, Pilastro A (1996) Armaments and ornaments:
an evolutionary explanation of traits of dual utility. Biol J Linn
Soc 58:385–399
Biro PA, Stamps JA (2008) Are animal personality traits linked to life-
history productivity. TREE 23:361–368. doi:10.1016/j.
tree.2008.04.003
Blake MA, Hart PJB (1993) The behavioural response of juvenile
signal crayfish Pacifastacus leniusculus to stimuli from perch
and eels. Freshw Biol 29:89–97
Brown C, Burgess F, Braithwaite VA (2007) Heritable and experien-
tial effects on boldness in a tropical poeciliid. Behav Ecol
Sociobiol 62:237–243
Bywater CL, Angilletta MJ, Wilson RS (2008) Weapon size is a
reliable indicator of strength and social dominance in female
slender crayfish (Cherax dispar). Funct Ecol 22:311–316.
doi:10.1111/j.1365-2435.2007.01379.x
Carvalho LN, Del-Claro K (2004) Effects of predation pressure on the
feeding behaviour of the serpa tetra Hyphessobrycon eques
(Ostrariophysi, Characidae). Acta Ethol 7:89–93. doi:10.1007/
s10211-004-0093-1
Dall SRX, Houston AI, McNamara JM (2004) The behavioural
ecology of personality: consistent individual differences from an
adaptive perspective. Ecol Lett 7:734–739. doi:10.1111/j.1461-
0248.2004.00618.x
Dingemanse NJ, Van der Plas F, Wright J, Reale D, Schrama M, Roff
DA, Van der Zee E, Barber I (2009) Individual experience and
evolutionary history of predation affect expression of heritable
variation in fish personality and morphology. Proc R Soc Lond B
Biol Sci 276:1285–1293
Figler MH, Cheverton HM, Blank GS (1999) Shelter competition in
juvenile red swamp crayfish Procambarus clarkii: the influences
of sex differences, relative size, and prior residence. Aquaculture
178:63–75
Garvey JE, Stein RA, Thomas HM (1994) Assessing how fish
predation and interspecific prey competition influence a crayfish
assemblage. Ecology 75:532–547
Hirvonen H, Holopainen S, Lempiäinen N, Selin M, Tulonen J (2007)
Sniffing the trade-off: effects of eel odours on nocturnal foraging
activity of native and introduced crayfish juveniles. Mar Freshw
Behav Physiol 40:213–218
Huntingford FA (1976) The relationship between anti-predator behaviour
and aggression among conspecifics in the three-spined stickleback,
Gasterosteus aculeatus. Anim Behav 24:245–260
Ioannou CC, Payne M, Krause J (2008) Ecological consequences of
the bold–shy continuum: the effect of predator boldness on prey
risk. Oecologia 157:177–182
Kats LB, Dill LM (1998) The scent of death: chemosensory assessment
of predation risk by prey animals. Ecoscience 5:361–394
Kortet R, Hedrick A (2007) A behavioural syndrome in the field
cricket Gryllus integer: intrasexual aggression is correlated with
activity in a novel environment. Biol J Linn Soc 91:475–482.
doi:10.1111/j.1095-8312.2007.00812.x
Krebs CJ (1999) Ecological methodology, 2nd edn. Benjamin/
Cummings, Menlo Park
Lima SL, Bednekoff PA (1999) Temporal variation in danger drives
antipredatory behavior: the predation risk allocation hypothesis.
Am Nat 153:649–659
Lima SL, Dill LM (1990) Behavioral decisions made under the risk of
predation: a review and prospectus. Can J Zool 68:619–640
Luttbeg B, Sih A (2010) Risk, resources and state-dependent adaptive
behavioral syndromes. Phil Trans R Soc Lond B (in press).
doi:10.1098/rstb.2010.0207
Nyström P (2005) Non-lethal predator effects on the performance of a
native and an exotic crayfish species. Freshw Biol 50:1938–1949
Pintor LM, Sih A, Bauer ML (2008) Differences in aggression,
activity and boldness between native and introduced populations
of an invasive crayfish. Oikos 117:1629–1636. doi:10.1111/
j.1600-0706.2008.16578.x
Preisser EL, Bolnick DI, Benard MF (2005) Scared to death? The
effects of intimidation and consumption in predator–prey
interactions. Ecology 86:501–509
Ranta E, Lindström K (1993) Body size and shelter possession in
mature signal crayfish, Pacifastacus leniusculus. Ann Zool Fenn
30:125–132
Réale D, Reader SM, Sol D, McDougall PT, Dingemanse NJ (2007)
Integrating animal temperament within ecology and evolution.
Biol Rev 82:291–318. doi:10.1111/j.1469-185X.2007.00010.x
Riechert SE, Hedrick AV (1993) A test for correlations among fitness-
linked behavioural traits in the spider Agelenopsis aperta
(Araneae, Agelenidae). Anim Behav 46:669–675
Riesch R, Duwe V, Herrmann N, Padur L, Ramm A, Scharnweber K,
Schulte M, Schulz-Mirbach T, Ziege M, Plath M (2009) Variation
along the shy–bold continuum in extremophile fishes (Poecilia
mexicana,Poecilia sulphuraria). Behav Ecol Sociobiol 63:1515–
1526
Rutherford PL, Dunham DW, Allison V (1995) Winning agonistic
encounters by male crayfish Orconectes rusticus (Girard)
(Decapoda, Cambaridae): chela size matters but chela symmetry
does not. Crustaceana 68:526–529
Schroeder L, Huber R (2001) Fight strategies differ with size and
allometric growth of claws in crayfish, Orconectes rusticus.
Behaviour 138:1437–1449
Sih A, Bell AM (2008) Insights for behavioral ecology from
behavioral syndromes. Adv Study Behav 38:227–281
Sih A, McCarthy TM (2002) Prey responses to pulses of risk and
safety: testing the risk allocation hypothesis. Anim Behav
63:437–443. doi:10.1006/anbe.2001.1921
24 acta ethol (2011) 14:17–25
Sih A, Bell AM, Johnson JC, Ziemba RE (2004) Behavioral
syndromes: an integrative overview. Quart Rev Biol 79:241–277
Sinn DL, Apiolaza LA, Moltschaniwskyj NA (2006) Heritability and
fitness-related consequences of squid personality traits. J Evol
Biol 19:1437–1447
Snedden WA (1990) Determinants of male mating success in the
temperate crayfish Orconectes rusticus: chela size and sperm
competition. Behaviour 115:100–113
Söderbäck B (1992) Predator avoidance and vulnerability of two co-
occurring crayfish species, Astacus astacus (L.) and Pacifastacus
leniusculus (Dana). Ann Zool Fenn 29:253–259
Stamps JA (2007) Growth–mortality tradeoffs and ‘personality traits’
in animals. Ecol Lett 10:355–363. doi:10.1111/j.1461-
0248.2007.01034.x
Stamps J, Groothuis TGG (2010) The development of animal
personality: relevance, concepts and perspectives. Biol Rev
85:301–325. doi:10.1111/j.1469-185X.2009.00103.x
Stein RA, Magnuson JJ (1976) Behavioral response of crayfish to a
fish predator. Ecology 57:751–761
Usio N, Konish M, Nakano S (2001) Species displacement
between an introduced and a ‘vulnerable’crayfish: the role
of aggressive interactions and shelter competition. Biol
Invasions 3:179–185
Wilson ADM, McLaughlin RL (2007) Behavioural syndromes
in brook charr, Salvelinus fontinalis: prey-search in the field
corresponds with space use in novel laboratory situations.
Anim Behav 74:689–698. doi:10.1016/j.anbehav.
2007.01.009
acta ethol (2011) 14:17–25 25
