1. No category

INDIVIDUAL DIFFERENCES IN ACTIVITY AND EXPLORATION
INFLUENCE LEADERSHIP IN PAIRS OF FORAGING
ZEBRA FINCHES
by
GUY BEAUCHAMP 1)
(Faculty of Veterinary Medicine, University of Montréal, CP 5000, St-Hyacinthe, Québec,
Canada J2S 7C6)
(Acc. 14-X-1999)
Summary
This study investigated the role of dominance and level of activity and exploration on
leadership in zebra Ž nches (Taenopygia guttata) searching for food. In pairs of zebra Ž nches
fairly matched in size and that experienced the same level of food deprivation, the same bird
consistently reached Ž rst one foraging patch over several trials. The same pattern of arrival
to food occurred when resources were provided in two distant patches available concurrently,
a situation that would potentially allow subordinates a greater access to resources. In further
testing, the formation of new pairs with the same birds led to several changes in leadership,
indicating that leadership is not an absolute feature. The member of a pair that proved to be
the most active and exploratory during independent, solitary trials became the leader in nearly
all pairs tested. The same pattern held true in newly rearranged pairs where individuals often
experienced changes in dominance status. Dominance failed to be associated with leadership
in all tests. The results suggest that in a relatively egalitarian species, level of activity and
exploration may be a stronger predictor of leadership than dominance.
Introduction
Leadership in the movement order of individuals searching for food in
groups has been observed in several species of Ž sh (Bumann & Krause,
1993; Krause et al., 1998), birds (Katzir, 1982; Beauchamp, 1992; Lamprecht, 1992; Zanette & Ratcliffe, 1994; Tremblay & Cherel, 1999) and
1) E-mail address: [email protected]
c Koninklijke Brill NV, Leiden, 2000
®
Behaviour 137, 301-314
302
GUY BEAUCHAMP
mammals (Meese & Ewbank, 1973; Arnold, 1977; Dunbar, 1983). Leaders and followers in a group often experience different Ž tness consequences
(Krause, 1994). For instance, individuals leading a group can obtain a greater
share of resources but may also face a greater risk of predation (Krause,
1993; Bumann et al., 1997). Similarly, followers can beneŽ t from the food
discoveries of leaders but may show impaired learning about foraging contingencies (Beauchamp & Kacelnik, 1991; Burt de Perera & Guilford, 1999).
Although the Ž tness consequences associated with leadership are fairly
well understood, the proximate determinants are unclear. Individual differences in physical attributes, such as speed of movement and hunger levels,
may allow some individuals to lead and force others to follow. In roach (Rutilus rutilus), for instance, faster and hungrier individuals were more likely
to be found at the front of a shoal and enjoyed a greater access to resources
(Krause et al., 1998). Individual differences in dominance status have often
been linked to leadership. In some species, dominant individuals consistently
provide the lead in group movement (Dunbar, 1983; Norton, 1986). In other
species, less dominant group members, who often experience restricted access to resources, have been known to take the lead (Katzir, 1982; Waite,
1989; McLeod & Huntingford, 1994; Zanette & Ratcliffe, 1994) perhaps
simply as a result of greater hunger levels.
Individual differences in level of activity and exploration is a factor that
has received comparatively less attention. Level of activity and exploration,
as measured by individual responses in a novel environment, has been found
to correlate with several behavioural traits (Wilson, 1998; Verbeek et al.,
1999). In particular, less active and exploratory individuals tend to be found
close to other individuals rather than alone (Wilson et al., 1993; Budaev,
1997a). More active and exploratory individuals are thus probably more
likely to wander away from a group in search of foraging opportunities.
Remaining group members would then simply follow the lead. I surmise that
leadership in a group can re ect individual differences in level of activity
and exploration.
I examined the role of dominance and level of activity and exploration on
leadership in pairs of captive zebra Ž nches (Taenopygia guttata) searching
for food. Zebra Ž nches are small estrildid Ž nches that forage in groups
with little aggression (Zann, 1996). Independent estimates of dominance
and level of activity and exploration were used to predict the order of
arrival to a single food source and, subsequently, to two distant food
LEADERSHIP IN FORAGING ZEBRA FINCHES
303
sources available concurrently. An increase in the number of food sources
available concurrently was expected to manipulate the potential impact of
dominance on arrival order by allowing subordinates a greater access to
resources (Grant, 1993). I also examined the order of arrival to food in
newly rearranged pairs. If leadership is a trait that changes as a function
of group composition, then variation in dominance and in the relative level
of activity and exploration between birds could be used to predict changes
in leadership. In all trials, individuals experienced the same level of food
deprivation as a control for hunger levels. However, since small and large
birds may experience the same level of food deprivation differently, I also
tested the effect of body mass on leadership.
Methods
Subjects
Seven adult male and three adult female zebra Ž nches, obtained from a commercial supplier,
were used for the experiment. I kept pairs of birds in 60 ´ 30 ´ 45 cm cages under a
13 : 11 L : D photoperiod. Water and a commercial seed mixture were available ad libitum
during non-testing periods. Bath water was provided on a weekly basis and greenery twice
a week. Prior to the experiment, I formed three pairs of males, one pair of females and one
heterosexual pair. Distinctive plumage characteristicsallowed individual recognition of birds.
Prior to the start of the experiment, birds were weighed to the nearest 0.1 g after a 90 min
food deprivation period.
Experimental procedure
The experiment proceeded sequentially in Ž ve parts. The Ž rst two parts documented leadership in each pair under two different food regimes. The third part established the level of
activity and exploration of each bird. Dominance status in each pair was established during
the fourth part. Leadership in four newly rearranged pairs was investigated during the last
part of the experiment.
Establishment of leadership
Determination of leadership in each pair took place over a period of 14 days. During the
whole period, trials were conducted in the evening at the same time each day. In the Ž rst
week, leadership status was evaluated when food was provided from a single source. During
this period, birds received one trial daily after a 80 min food-deprivation period. For each trial,
a total of 30 white millet seeds were spread evenly in a circular  at dish (8 cm in diameter).
The dish was introduced through a sliding door at the bottom of each cage and placed in
the middle of the left or right side of the cage, the side being selected randomly each day.
The observer then sat on a chair located 2 m away for the remainder of the trial. The order
of arrival to the food dish was noted. Leadership was deŽ ned by which bird crossed Ž rst a
304
GUY BEAUCHAMP
circling line 3 cm away from the edge of the food dish. Birds crossing the line always aimed
at the food dish. After all seeds were eaten, the food dish was removed and free access to
food was then provided. Pairs were tested sequentially and the order of testing across pairs
was varied randomly across days.
The following week, the same protocol was used to investigate leadership when two
distant food sources were available concurrently. Prior to each trial, I spread 15 white millet
seeds in two of the aforementioned food dishes. Food dishes were placed simultaneously on
the left and right side of the cage. The two food sources were separated by a distance of 20 cm,
thus reducing to a minimum the number of potential interactions between birds feeding from
different sources. The order of arrival to the food dishes was noted as before.
Establishment of individual differences in level of activity and exploration
Level of activity and exploration for each bird in each pair was recorded over two days. Birds
received one trial daily. At the beginning of a trial with one pair, food and water were removed
and an opaque partition was introduced in the middle of the cage to separate the two birds.
In each compartment, birds could move between two perches and also land on the bottom
of the cage for exploration. Traditionally, level of activity and exploration is examined in a
novel environment (e.g. Jones & Waddington, 1992; Budaev, 1997b; Reboucas & Schmidek,
1997). In the current protocol, even though the familiar home cage was used, the introduction
of the partition and the sudden visual isolation created at least two elements of novelty. The
two birds were given 5 min to habituate to the new conditions. After the habituation period,
I recorded the behaviour of both birds simultaneously every 15 s during a 5-min period.
During each scan, I noted whether each bird was on a perch or on the bottom of the cage and
whether each bird was moving or standing still. After the completion of the test, the partition
was removed and food and water were made available. The remaining pairs were tested in
a similar fashion. The following day, the procedure was repeated using a different order of
testing across pairs.
From the number of scans in the data, I calculated: (1) the proportion of time that each
bird spent on perches (Location), (2) the proportion of time that each bird spent in movement
(Movement), (3) the time each bird needed to reach the bottom of the cage for the Ž rst time
(Bottom) and, Ž nally, (4) the number of transitions from a perch to the bottom of the cage
and vice versa (Transition). Location and Bottom are considered measures of exploration
tendencies as individuals must move away from perches to explore the cage. Movement and
Transition are considered activity measures re ecting horizontal and vertical movements.
I subjected the four variables to principal components analysis. The average score over
the two replicates was used along with the correlation matrix of the four variables. Principal
components analysis allows a representation of data along a reduced number of axes relative
to the original data set (Legendre & Legendre, 1998). Each axis is a linear combination of
independent variables that corresponds to the successive directions of maximum variance in
the data scatter. Not all possible principal components axes are usually interpreted. When
the principal components analysis is based on the correlation matrix, only the components
whose eigenvalue (a value associated with the total variance accounted for by the axis) is
greater than one are usually considered (Legendre & Legendre, 1998). Once the new system
of axes was deŽ ned, co-ordinates of data points for each bird were recalculated. I used the
new co-ordinates along the signiŽ cant axes as a measure of level of activity and exploration.
As determined by the new relative co-ordinates, birds with the highest score in each pair were
categorised as less active and exploratory.
LEADERSHIP IN FORAGING ZEBRA FINCHES
305
Establishment of dominance status
Dominance status in each pair was documented on two occasions over a period of Ž ve days.
During the Ž rst two days, birds received one trial daily. Food was removed 80 min prior to a
trial. At the beginning of a trial, the food dispenser was put back in place. The food dispenser
consisted of an inverted test tube with an opening at the bottom that allowed a seed mixture to
fall in a narrow plastic container. The plastic container protruded in the cage next to a perch.
Due to space limitation, only one bird at a time could have access to the food dispenser.
During each trial, I measured the time needed by one focal bird to accumulate 60 s of feeding
time. The trial started when the Ž rst bird arrived at the food dispenser. A stop-watch was
used to measure the duration of feeding bouts for the focal bird. The stop-watch was started
when the focal bird obtained access to the feeder and stopped when the focal bird left the
food dispenser. One bird in each pair was observed the Ž rst day. The following day, the same
protocol was used for the remaining birds. The whole sequence of observations was repeated
one day later using a different order of testing to estimate the reliability of dominance scores.
The member of a pair that accumulated 60 s of feeding time the fastest was deemed to be
dominant.
Rearrangement of pairs
The Ž nal part of the experiment examined leadership in four newly rearranged pairs. In two
cages, I exchanged the more active and exploratory member of one pair for the less active
and exploratory companion of another pair to form four new pairs. The newly formed pairs
thus consisted of birds found previously to be either active and exploratory or not. After
one day of habituation, I established leadership and dominance status in each pair using the
aforementioned procedures. Leadership status was only examined when two food sources
were available concurrently.
Statistical analyses
Data from the Ž ve original pairs are referred to as subset 1 and those from the four newly
rearranged pairs as subset 2.
Reliabilities of dominance scores over the two dominance tests in subset 1 and subset 2
of the data and of the four time-budget parameters over the activity and exploration tests in
subset 1 of the data were investigated with Spearman’s correlation.
To determine leadership in each pair, I assigned for each trial a score of 1 to the bird that
arrived Ž rst at the food source and a score of 0 otherwise. When two birds arrived at the same
time, a score of 0.5 was given to each individual. Scores for each bird were then added over
the seven trials. Under the null hypothesis that each bird is equally likely to arrive Ž rst at the
food source, the frequency of Ž rst arrivals for one bird over the seven trials should follow
the binomial distribution. I used a unilateral binomial test on the frequencies of Ž rst time
arrivals over the seven trials to determine departures from the null hypothesis. In the withinpair analysis, I set the level of signiŽ cance at 0.06 so that a score of six or seven was required
to provide a signiŽ cant departure from the null hypothesis. In the subset of pairs where one
member led signiŽ cantly more often than the companion, the number of pairs led by an active
and exploratory bird or by a dominant individual should follow the binomial distribution. In
the between-pair analysis, I set the level of signiŽ cance at 0.05. The relationship between
leadership and level of activity and exploration was assessed across pairs with a unilateral
306
GUY BEAUCHAMP
binomial test. Because the effect of dominance status on leadership could not be predicted
a priori, I used a bilateral binomial test in this case.
I used a bilateral Wilcoxon signed ranks test to examine the relationship between body
mass and leadership in each subset of pairs.
Results
After food was introduced, birds in each pair usually waited on a perch and
gradually moved to the bottom of the cage within 30 s. One bird clearly
approached the food source Ž rst in all but two trials. When feeding from
one source, birds in each pair collected seeds with little aggression. When
two food sources were available concurrently, birds in Ž ve out of the nine
pairs tested fed simultaneously from the same dish before moving to the
alternative source.
Dominance scores proved reliable over a period of three days in both
subsets of zebra Ž nches (Table 1). Similarly, the four variables extracted
from the activity and exploration tests also tended to be reliable from one
day to the next (Table 1).
The principal components analysis revealed one signiŽ cant axis that
maximised the variance in the original scatter of points (Table 2). The axis
was loaded positively with exploration measures (Location and Bottom)
and negatively with activity measures (Transition and Movement). In each
pair, birds labelled as active and exploratory were more likely than their
counterparts to: (1) venture away from the top of the cage, (2) reach the
TABLE 1. Reliabilities of the measures scored in the dominance and activity
and exploration tests in two subsets of zebra Ž nches
Variable
Spearman’s rho
Subset 1 (N = 10)
Dominance
Movement
Location
Number of transitions
Time to reach bottom
0.86
0.55
0.94
0.65
0.74
Subset 2 (N = 8)
Dominance
0.79
p
<
0.05 <
<
<
<
0.01
p < 0.10
0.01
0.03
0.03
< 0.03
307
LEADERSHIP IN FORAGING ZEBRA FINCHES
TABLE 2. Principal components analysis loadings for the activity and
exploration test in zebra Ž nches
Variable
Activity/Exploration
Not interpreted
­ 0.44
0.46
­ 0.53
0.56
0.65
0.64
0.30
0.27
2.97
74.2
0.85
21.2
Movement
Location
Number of transitions
Time to reach bottom
Eigenvalue
Variance accounted (%)
TABLE 3. Effects of dominance and level of activity and exploration on the
order of arrival to food in pairs of zebra Ž nches: one available food source
Pairs
S-NS
F-M
C-NC
L-NL
G-W
Dominance status
S>
F>
C>
NL >
G>
NS
M
NC
L
W
Activity/exploration
level
NS >
F>
NC >
L>
G>
S
M
C
NL
W
Arrival order
NS >
F>
NC >
L>
G>
S (7/ 7)a
M (7/ 7)
C (4/ 7)
NL (6/ 7)
W (7/ 7)
p
< 0.01
< 0.01
0.27
0.06
< 0.01
a Number of Ž rst arrivals on total number of trials.
bottom more quickly, (3) make several transitions between perches and the
bottom of the cage and, Ž nally (4) be in movement in any part of the cage.
When feeding from one source, the same birds arrived Ž rst to the food
dish more often than their companions in four out of the Ž ve pairs tested
(Table 3). When two separate food sources were available concurrently, the
same birds also arrived Ž rst more often than their companions (Table 4). A
similarly strong pattern of leadership emerged in the newly rearranged pairs
(Table 4). Pooling results from the last two tests with separate food sources,
the bird that arrived Ž rst more often to Ž nd resources proved to be the more
active and exploratory partner in all nine pairs tested (p < 0.01). On the
other hand, leadership and dominance status failed to show any association
(p = 1).
Several changes in dominance and leadership status occurred in the
rearranged pairs. In addition, the ranking of birds in terms of activity and
exploration score also changed frequently in new pairs. In pairs M-W and
NL-C, two birds (M and NL) that arrived consistently second in the original
308
GUY BEAUCHAMP
TABLE 4. Effects of dominance and level of activity and exploration on the
order of arrival to food in pairs of zebra Ž nches: two available food sources
Pairs
Dominance status
Activity/exploration
level
Arrival order
p
Subset 1
S-NS
F-M
C-NC
L-NL
G-W
S>
F>
C>
NL >
G>
NS
M
NC
L
W
NS >
F>
NC >
L>
G>
S
M
C
NL
W
NS >
F>
NC >
L>
G>
S (7/ 7)a
M (7/ 7)
C (6/ 7)
NL (7/ 7)
W (7/ 7)
< 0.01
< 0.01
0.06
< 0.01
< 0.01
Subset 2
M-W
C-NL
L-NC
F-G
W>
NL >
L>
G>
M
C
NC
F
M>
NL >
L>
F>
W
C
NC
G
M>
NL >
L>
F>
W (7/ 7)
C (7/ 7)
NC (6/ 7)
G (6/ 7)
< 0.01
< 0.01
0.06
0.06
a Number of Ž rst arrivals on total number of trials.
pairings arrived Ž rst consistently with their new companion. Similarly,
in pairs L-NC and F-G, two birds (NC and G) that arrived consistently
Ž rst in the original pairings now arrived consistently second with a new
companion. When compared with their original status, birds M, NL, NC and
G experienced a shift in the ranking based on activity and exploration scores
but experienced no concomitant shift in dominance status. Only changes
in relative activity and exploration score thus correlated with changes in
leadership for these four birds. The remaining four birds (W, C, L, and F)
experienced a shift in dominance status without a concomitant change in
leadership.
Leadership status failed to be associated with body mass in both subsets
of birds (Subset 1: T = 12, p = 0.16, N = 5; Subset 2: T = 6, p = 0.44,
N = 4).
Discussion
When two zebra Ž nches approach a foraging patch, one bird consistently
leads the way and reaches resources Ž rst. Because leaders and followers
were fairly matched in terms of body mass and experienced a similar level
of food deprivation, birds leading the way were unlikely to be hungrier than
their more reticent companions. Differences in speed of movement are also
LEADERSHIP IN FORAGING ZEBRA FINCHES
309
unlikely to be involved due to the short distance between a perch and the
bottom of a cage.
Leadership was poorly predicted by dominance status within a pair.
Subordinate and dominant pair members arrived Ž rst at food sources in
a nearly equal number of pairs. In addition, the same qualitative pattern
of arrival to food occurred when pair members fed from the same food
source or had access to two distant food sources available concurrently.
Therefore, variation in the potential for direct dominance interactions at
one food source failed to hamper leadership. Finally, in newly rearranged
pairs, where individuals often acquired new dominance and leadership status,
changes in leadership failed to correlate with changes in dominance status.
Level of activity and exploration appears a stronger predictor of leadership
in zebra Ž nches. Indeed, the more active and exploratory member of a pair
arrived Ž rst at a food source consistently more often than the less active and
exploratory companion in well-established and newly rearranged pairs.
Level of activity and exploration emerges as a strong determinant of leadership in a relatively egalitarian species when hunger levels are controlled.
Few other studies have investigated the role of behavioural attributes on leadership. Sheep (Ovis aries) that spent the most time alone or that called less
when isolated became movement leaders more often (Arnold, 1977; Syme,
1981). Spending time alone was found to be more frequent among bold
wrasse (Symphodus ocellatus) which would support the hypothesis that more
exploratory individuals tend to lead more often (Budaev, 1997a). Similarly,
low levels of vocalisation while isolated could be regarded as a sign of low
fearfulness, a feature that could be associated with a more active and exploratory predisposition. More attention to behavioural attributes may reveal
a greater role of level of activity and exploration on leadership and other
behavioural patterns as well (Wilson, 1998). In addition, more research will
help understand the development and maintenance of such individual differences in behaviour within populations.
The effect of level of activity and exploration on leadership has potential consequences for producing and scrounging in social species. Foraging
in groups allows the use of two tactics to obtain resources (Giraldeau &
Beauchamp, 1999). In a society with a strict dominance hierarchy, socially
dominant foragers can use the scrounger tactic and rely on producer subordinates to Ž nd resources. In a more egalitarian society, other behavioural
attributes may be relevant to the relative use of the two foraging tactics. For
310
GUY BEAUCHAMP
instance, more active and exploratory foragers may be expected to use the
producer tactic to a greater extent than less active and exploratory companions who might rely entirely on the scrounger tactic to obtain resources. Although individual differences in the use of alternative foraging tactics have
been documented in zebra Ž nches (Giraldeau et al., 1990; Biondolillo et al.,
1997), the association between level of activity and exploration and scrounging behaviour remains to be examined.
The lack of an effect of dominance on leadership is not totally unexpected
in zebra Ž nches, a relatively egalitarian species in terms of foraging behaviour (Zann, 1996). In previous studies with zebra Ž nches, dominance status
also failed to correlate with the use of scrounging in foraging groups (Giraldeau et al., 1990). In addition, dominance status is not always a strong
predictor of leadership in other species. In domestic mammals, no relationship between movement leadership and dominance status emerged in cows
(Bos taurus: Kilgour & Scott, 1959; Tulloh, 1961; Beilharz & Melrea, 1963;
Leyhausen & Heinemann, 1975), pigs (Sus scrofa: Meese & Ewbank, 1973),
and goats (Capra hircus: Stewart & Scott, 1947) under a wide range of Ž eld
and captive conditions. However, in one study, dominant sheep were found
to initiate movements in a Ž eld more often than subordinate group members
(Squires & Daws, 1975). In wild primates, leaders often tend to be more
dominant group members (Dunbar, 1983; Norton, 1986). In sticklebacks
(Gasterosteus aculeatus), the relationship between leadership in predator inspection visits and dominance was not linear and middle-ranking individuals
were more likely to provide the lead than either top- or bottom-ranking companions (McLeod & Huntingford, 1994).
In birds, the results are also inconclusive. Dominance status in domesticated hens (Gallus gallus) failed to predict the order of emergence from a
shelter to a novel area (Grigor et al., 1995). On the other hand, lower ranking
jackdaws (Corvus monedula) were more likely to enter a novel area Ž rst than
their more dominant companions (Katzir, 1982). In bar-headed geese (Anser
indicus) under semi-captive conditions, strong leadership by certain group
members occurred at different times of the year but no permanent leader of
a pair or family group was found (Lamprecht, 1992). In chickadees (Parus
atricapillus: Zanette & Ratcliffe, 1994) and tufted titmice (Parus bicolor:
Waite, 1989) subordinate group members were more likely to emerge Ž rst
from hiding to search for food. When dominance in uences leadership status, in birds at least, subordinate group members are more likely to lead the
search for food.
LEADERSHIP IN FORAGING ZEBRA FINCHES
311
Three hypotheses can explain the relationship between dominance and
leadership. First, subordinate group members often experience reduced access to resources (Hogstad, 1988; Ficken et al., 1990; Henderson & Hart,
1995; Koivula et al., 1995; Lahti, 1998) and may therefore be hungrier than
more dominant companions. However, a relationship between hunger levels
and dominance was not suspected in some of the above studies on leadership
(Katzir, 1982; Waite, 1989; Zanette & Ratcliffe, 1994). Second, dominance
may be related to level of activity and exploration; subordinate group members could therefore be more active and exploratory than dominant companions. In several species, however, the relationship between dominance and
personality traits in general is not clear (Wilson et al., 1993; Forkman et
al., 1995; Grigor et al., 1995; Thodberg et al., 1999; Verbeek et al., 1999).
Finally, subordinate group members may take the lead to get access to resources Ž rst as a means to lessen, at least temporarily, the negative impact
of competition with more dominant companions. In this case, leadership by
subordinates represents a trade-off between foraging beneŽ ts and exposure
to risk rather than a behavioural attribute (Koivula et al., 1994; Verhulst
& Hogstad, 1996; Lahti et al., 1997). The trade-off hypothesis would predict more risk averse behaviour in the absence of dominant group members
(Desrochers, 1989). Other attributes, such as level of activity and exploration,
must be invoked in species where access to food is less dependent on dominance status.
One avenue for future research in zebra Ž nches is the study of factors that
in uence leadership in groups with more than two individuals. Results from
the current experiment already show that leadership is a relative trait that
changes as a function of group composition. Although the speciŽ c leader
of one group may be expected to change when group size changes, one
would predict that the more active and exploratory group members will
tend to be among the leaders. Nevertheless, simple extensions of Ž ndings
from small to large group sizes are not always straightforward given the
more complex nature of interactions in larger groups (van Oortmerssen
et al., 1985; Verbeek et al., 1999). Another issue is whether leadership
status can extend to several types of contexts or vary as a function of the
situation. Some studies have shown that different individuals lead under
different contexts and at different times of the year (Beilharz & Melrea, 1963;
Lamprecht, 1992), which would support the idea that pervasive leadership
should not always be expected. Group leadership must follow from the
312
GUY BEAUCHAMP
complex interactions between the behaviour of all group members and will
be predicted to vary from one situation to the next when the costs and
beneŽ ts of leading and following vary. What causes individual differences
in behavioural attributes, such as level of activity and exploration, and how
stable these differences are across time are issues also worth considering. In
some species, genetic underpinnings are suspected (van Oortmerssen et al.,
1985; McCune, 1995). Individual differences have been found to be stable
over several years (Figuredo et al., 1995; Capitanio, 1999) but little work is
available in birds.
References
Arnold, G.W. (1977). An analysis of spatial leadership in a small Ž eld in a small  ock of
sheep. — Appl. Anim. Ethol. 3, p. 263-270.
Beauchamp, G. (1992). Diving behavior in surf scoters and Barrow’s goldeneyes. — Auk
109, p. 819-827.
— — & Kacelnik, A. (1991). Effects of the knowledge of partners on learning rates in zebra
Ž nches Taeniopygia guttata. — Anim. Behav. 41, p. 247-253.
Beilharz, R.G. & Melrea, P.J. (1963). Social position and movement orders of dairy heifers.
— Anim. Behav. 11, p. 529-533.
Biondolillo, K., Stamp, C., Woods, J. & Smith, R. (1997). Working and scrounging by zebra
Ž nches in an operant task. — Behav. Proc. 39, p. 263-269.
Budaev, S.V. (1997a). Alternative styles in the European wrasse, Symphodus ocellatus:
boldness-related schooling tendency. — Env. Biol. Fishes 49, p. 71-78.
— — (1997b). “Personality” in the guppy (Poecilia reticulata): A correlational study of
exploratory behavior and social tendency. — J. Comp. Psychol. 111, p. 399-411.
Bumann, D. & Krause, J. (1993). Front individuals lead in shoals of three-spined sticklebacks
(Gasterosteus aculeatus) and juvenile roach (Rutilus rutilus). — Behaviour 125, p. 189198.
— —, — — & Rubenstein, D. (1997). Mortality risk of spatial positions in animal groups:
The danger of being in the front. — Behaviour 134, p. 1063-1076.
Burt de Perera, T. & Guilford, T. (1999). The social transmission of spatial information in
homing pigeons. — Anim. Behav. 57, p. 715-719.
Capitanio, J.P. (1999). Personality dimensions in adult male rhesus macaques. Prediction of
behavior across time and context. — Am. J. Primatol. 47, p. 299-320.
Desrochers, A. (1989). Sex, dominance, and microhabitat use in wintering black-capped
chickadees: A Ž eld experiment. — Ecology 70, p. 636-645.
Dunbar, R.I.M. (1983). Structure of gelada baboon reproductive units. IV. Integration at group
level. — Z. Tierpsychol 63, p. 265-282.
Ficken, M.S., Weise, C.M. & Popp, J.W. (1990). Dominance rank and resource access in
winter  ocks of black-capped chickadees. — Wilson Bull. 102, p. 623-633.
Figueredo, A.J., Cox, R.L. & Rhine, R.J. (1995). A generalizability analysis of subjective
personality assessments in the stumptail macaque and the zebra Ž nch. — Multivar.
Behav. Res. 30, p. 167-197.
LEADERSHIP IN FORAGING ZEBRA FINCHES
313
Forkman, B., Furuhaug, I.L. & Jensen, P. (1995). Personality, coping patterns, and aggression
in piglets. — Appl. Anim. Behav. Sci. 45, p. 31-42.
Giraldeau, L.-A. & Beauchamp, G. (1999). Food exploitation: Searching for the optimal
joining policy. — Trends Ecol. Evol. 14, p. 102-106.
— —, Hogan, J.A. & Clinchy, M.J. (1990). The payoffs to producing and scrounging: what
happens when patches are divisible? — Ethology 85, p. 132-146.
Grant, J.W.A. (1993). Whether or not to defend? The in uence of resource distribution. —
Marine Behav. Physiol. 23, p. 137-153.
Grigor, P.N., Hughes, B.O. & Appleby, M.C. (1995). Emergence and dispersal behaviour in
domestic hens: effects of social rank and novelty of an outdoor area. — Appl. Anim.
Behav. Sci. 45, p. 97-108.
Henderson, I.G. & Hart, P.J.B. (1995). Dominance, food acquisition and reproductive success
in a monogamous passerine — the jackdaw Corvus monedula. — J. Avian Biol. 26,
p. 217-224.
Hogstad, O. (1988). Rank-related resource access in winter  ocks of willow tits Parus
montanus. — Ornis Scand. 19, p. 169-174.
Jones, R.B. & Waddington, D. (1992). ModiŽ cation of fear in domestic chicks, Gallus gallus
domesticus, via regular handling and early environmental enrichment. — Anim. Behav.
43, p. 1021-1033.
Katzir, G. (1982). Relationships between social structure and response to novelty in captive
jackdaws, Corvus monedula L., I. Response to novel space. — Behaviour 81, p. 231263.
Kilgour, R. & Scott, T.H. (1959). Leadership in a herd of dairy cows. — Proc. New Zealand
Soc. Anim. Prod. 19, p. 36.
Koivula, K., Lahti, K., Rytkonen, S. & Orell, M. (1994). Do subordinates expose themselves
to predation — Ž eld experiments on feeding site selection by willow tits. — J. Avian
Biol. 25, p. 178-183.
— —, Orell, M., Rytkonen, S. & Lahti, K. (1995). Fatness, sex and dominance — seasonal
and daily body mass changes in willow tits. — J. Avian Biol. 26, p. 209-216.
Krause, J. (1993). The relationship between foraging and shoal position in a mixed shoal of
roach (Rutilus rutilus) and chub (Leuciscus cephalus): a Ž eld study. — Oecologia 93,
p. 356-359.
— — (1994). Differential Ž tness returns in relation to spatial position in groups. — Biol. Rev.
69, p. 187-206.
— —, Reeves, P. & Hoare, D. (1998). Positioning behaviour in roach shoals: The role of body
length and nutritional state. — Behaviour 135, p. 1031-1039.
Lahti, K. (1998). Social dominance and survival in  ocking passerine birds: a review with an
emphasis on the willow tit Parus montanus. — Ornis Fen. 75, p. 1-17.
— —, Koivula, K. & Orell, M. (1997). Dominance, daily activity and winter survival in
willow tits — detrimental cost of long working hours. — Behaviour 134, p. 921-939.
Lamprecht, J. (1992). Variable leadership in bar-headed geese (Anser indicus): An analysis
of pair and family departures. — Behaviour 122, p. 105-120.
Legendre, P. & Legendre, L. (1998). Numerical ecology. — Elsevier, Amsterdam.
Leyhausen, P. & Heinemann, I. (1975). “Leadership” in a small herd of dairy cows. — Appl.
Anim. Ethol. 1, p. 206.
McCune, S. (1995). The impact of paternity and early socialisation on the development of
cats behaviour to people and novel objects. — Appl. Anim. Behav. Sci. 45, p. 109-124.
314
GUY BEAUCHAMP
McLeod, P.G. & Huntingford, F.A. (1994). Social rank and predator inspection in sticklebacks. — Anim. Behav. 47, p. 1238-1240.
Meese, G.B. & Ewbank, R. (1973). Exploratory behaviour and leadership in the domesticated
pig. — Brit. Vet. J. 129, p. 251-259.
Norton, G.N. (1986). Leadership: decision processes of group movement in yellow baboons.
— In: Primate ecology and conservation (J.G. Else & P.C. Lee, eds). Cambridge
University Press, New York, p. 145-156.
van Oortmerssen, G.A., Benus, I. & Dijk, D.J. (1985). Studies in wild house mice: genotypeenvironment interactions for attack latencies. — Neth. J. Zool. 35, p. 155-169.
Reboucas, R.C.R. & Schmidek, W.R. (1997). Handling and isolation in three strains of rats
affect open Ž eld, exploration, hoarding and predation. — Physiol. Behav. 62, p. 11591164.
Squires, V.R. & Daws, G.T. (1975). Leadership and dominance relationships in Merino and
Border Leicester sheep. — Appl. Anim. Ethol. 1, p. 263-274.
Stewart, J.C. & Scott, J.P. (1947). Lack of correlation between leadership and dominance
relationships in a herd of goats. — J. Comp. Physiol. Psychol. 40, p. 255-264.
Syme, L.A. (1981). Social disruption and forced movement orders in sheep. — Anim. Behav.
29, p. 282-288.
Thodberg, K., Jensen, K.H. & Herskin, M.S. (1999). A general reaction pattern across
situations in prepubertal gilts. — Appl. Anim. Behav. Sci. 63, p. 103-119.
Tremblay, Y. & Cherel, Y. (1999). Synchronous underwater foraging behavior in penguins.
— Condor 101, p. 179-185.
Tulloh, N.M. (1961). Behaviour of cattle in yards. I. Weighing, order and behaviour before
entering scales. — Anim. Behav. 9, p. 20-24.
Verbeek, M.E.M., De Goede, P., Drent, P.J. & Wiepkema, P.R. (1999). Individual behavioural
characteristics and dominance in aviary groups of great tits. — Behaviour 136, p. 23-48.
Verhulst, S. & Hogstad, O. (1996). Social dominance and energy reserves in  ocks of willow
tits. — J. Avian Biol. 27, p. 203-208.
Waite, T.A. (1989). Dominance-speciŽ c vigilance in the tufted titmouse: effects of social
context. — Condor 89, p. 932-935.
Wilson, D.S. (1998). Adaptive individual differences within single populations. — Phil.
Trans. R. Soc. Lond. B 353, p. 199-205.
— —, Coleman, K., Clark, A.B. & Biederman, L. (1993). Shy-bold continuum in pumpkinseed sunŽ sh (Lepomis gibbosus): an ecological study of a psychological trait. —
J. Comp. Psychol. 107, p. 250-260.
Zanette, L. & Ratcliffe, L.M. (1994). Social rank in uences conspicuous behaviour of blackcapped chickadees, Parus atricapillus. — Anim. Behav. 48, p. 119-127.
Zann, R.A. (1996). The zebra Ž nch. — Oxford University Press, Oxford.