You are on page 1of 14

Behav Ecol Sociobiol

DOI 10.1007/s00265-015-1958-1


Workers ‘specialized’ on inactivity: Behavioral consistency

of inactive workers and their role in task allocation
Daniel Charbonneau 1 & Anna Dornhaus 2

Received: 9 April 2015 / Revised: 7 June 2015 / Accepted: 11 June 2015

# Springer-Verlag Berlin Heidelberg 2015

Abstract Social insect colonies are often considered to be results underline the importance of inactivity as a behavioral
highly efficient collective systems, with division of labor at state and the need for further studies on its evolution.
the root of their ecological success. However, in many species,
a large proportion of a colony’s workers appear to spend their Keywords Task allocation . Specialization . Inactivity .
time completely inactive. The role of this inactivity for colony Colony organization . Shift work . Circadian rhythm . Social
function remains unclear. Here, we investigate how inactivity insect . Temnothorax . Division of labor
is distributed among workers and over time in the ant
Temnothorax rugatulus. We show that the level of inactivity
is consistent for individual workers, but differs significantly Introduction
among workers, that is, some workers effectively specialize
on ‘inactivity’. We also show that workers have circadian Social insects are among the most abundant, diverse, and wide-
rhythms, although intra-nest tasks tend to be performed uni- spread taxonomic groups (Wilson 1991; Samways 1993). Their
formly across the whole day. Differences in circadian ecological success is often attributed to division of labor (DOL)
rhythms, or workers taking turns resting (i.e., working in and worker specialization (Oster and Wilson 1978; Hölldobler
shifts), cannot explain the observation that some workers are and Wilson 1990). Social insects have evolved successful strat-
consistently inactive. Using extensive individual-level data to egies for allocating tasks to workers while solving a set of
describe the overall structure of division of labor, we show specific problems and constraints (Charbonneau and Dornhaus
that ‘inactive workers’ form a group distinct from other task In revision). Some of these strategies have inspired solutions to
groups. Hierarchical clustering suggests that inactivity is the analogous problems in diverse fields such as computer sciences
primary variable in differentiating both workers and tasks. Our (Johnson 2012), robotics (Gerkey and Matarić 2004), logistics
(Zhang and Chen 2011), sociology (Durkheim 1997), and eco-
nomics (Becker and Murphy 1992).
Communicated by L. Keller
A key feature of DOL is worker specialization, where sub-
Electronic supplementary material The online version of this article sets of workers tend to do one or a few tasks more than other
(doi:10.1007/s00265-015-1958-1) contains supplementary material,
workers. Most studies of specialization focus on a few prom-
which is available to authorized users.
inent tasks such as foraging, building, and brood care (Beshers
* Daniel Charbonneau
and Fewell 2001; Dornhaus 2008; Duarte et al. 2011), but rarely on more innocuous tasks such as grooming (though
Anna Dornhaus
these can also be specialized, Moore et al. 1995). Much work has been done to show that workers within colonies can often
Graduate Interdisciplinary Program in Entomology and Insect
be separated into discrete task groups, sometimes called be-
Science, University of Arizona, Biological Sciences West, 1041 East havioral castes (Wilson 1976; Mirenda and Vinson 1981;
Lowell, Room 235, Tucson, AZ 85721, USA Herbers 1983; Lenoir and Ataya 1983; Corbara et al. 1989;
Department of Ecology and Evolutionary Biology, University of Retana and Cerdá 1991; Pinter-Wollman et al. 2012; Mersch
Arizona, 1041 E Lowell St, Tucson, AZ 85721, USA et al. 2013). There is also a wealth of empirical, theoretical,
Behav Ecol Sociobiol

and conceptual work on potential task allocation strategies their exposure to risk so that they may lay their own eggs
(e.g., reviewed in: Charbonneau and Dornhaus In revision; (tested and supported, Hillis et al. In prep.; Jandt and
Gordon 1996; Robinson and Huang 1998; Beshers and Fewell Dornhaus 2011; tested, but not supported, Cole 1981; Cole
2001; Duarte et al. 2011). The different proposed task alloca- 1986; Ishii and Hasgeawa 2013), and inactive workers
tion mechanisms are not mutually exclusive, and different performing an as-yet unidentified function in which they ap-
tasks may be governed by different mechanisms (Gordon pear idle but providing a function (behaviorally idle rather
2002). However, we generally still know very little about the than functionally idle) such as playing a role in communica-
adaptive benefits/costs of each strategy (Charbonneau and tion (proposed explanation, not tested, O’Donnell and Bulova
Dornhaus In revision). 2007) or acting as food reserves (evidence of food reserves,
Despite high levels of worker inactivity being prevalent in but link to inactivity not tested, Sendova-Franks et al. 2010).
social insects, inactivity is one of the least well understood Alternatively, inactivity may be a constraint related to age
behaviors that workers engage in. The literature is filled with where young workers may less active due to inexperience/
reports of >50 % of workers in colonies being inactive across physical vulnerability (tested and supported, Corbara et al.
all social insect taxa (bees, Lindauer 1952; Jandt et al. 2009; 1989; Klein et al. 2008) and/or old workers due to senescence
wasps, Gadagkar and Joshi 1984; ants, Mirenda and Vinson (tested and supported immature, Corbara et al. 1989; Klein
1981; Herbers 1983; Cole 1986; Retana and Cerdá 1990; et al. 2008; immature and senescent, Fresneau 1984; evidence
Schmid-Hempel 1990; Dornhaus 2008; Dornhaus et al. against, Johnson 2008; Retana and Cerdá 1990). Ultimately,
2009; and termites, Rosengaus and Traniello 1991), and yet, the question of why colonies would produce so many inactive
the role of inactivity (rest and quiescence) is rarely considered workers, in spite of potentially high production and mainte-
in understanding task allocation strategies or colony organiza- nance costs, is still very much a mystery.
tion (for exceptions, see Herbers 1981; Fresneau 1984; Cole Perhaps one of the simplest explanations for inactivity is
1986; Corbara et al. 1989; Retana and Cerdá 1990; Retana and that, because of physiological need for rest (or sleep) even in
Cerdá 1991). Perhaps this is because inactivity is essentially a insects (Klein et al. 2003; Klein et al. 2010), all workers may
lack of doing anything else and so seems unimportant. How- be spending a certain amount of time ‘inactive’. In such a case,
ever, because so many workers spend so much time doing we would expect all workers to have more or less similar
what appears to be nothing at all, omitting inactivity in task needs and consequently for all workers to have consistent
allocation studies may skew our understanding of social insect and comparable levels of inactivity. We know that workers
DOL. For example, if inactivity does not serve a purpose per can vary in activity levels over the course of the day
se, and is common for all workers, then activity in one task (circadian rhythms; e.g., Klein and Seeley 2011) and between
may not trade off with activity in another task simply because seasons (Fellers 1989). Complex activity patterns can arise
much time is spent being inactive anyway. On the other hand, when these interact (Pol and de Casenave 2004). In the case
if only a few workers are very inactive, then they may fulfill of honey bees and bumble bees, foragers have been shown to
some unrecognized critical function. Finding adaptive expla- have diurnal rhythms, while in-nest workers, such as nurses,
nations for inactivity is particularly relevant because high do not (Moore 2001; Yerushalmi et al. 2006).
levels of inactivity are likely not an artifact of simplified living There is also evidence that honey bee foragers employ ‘shift
conditions in the lab (Charbonneau et al. 2015). work’ during the day, where workers show fidelity to foraging
Potential adaptive functions of high levels of inactivity in either in the morning or in the afternoon (Kraus et al. 2011).
social insects have been proposed but rarely tested (reviewed This may be linked to sleep timing and resource availability
in Charbonneau and Dornhaus In revision). The most com- (Klein and Seeley 2011). However, in-hive tasks, such as nurs-
monly cited explanation is that inactive workers are ‘reserves’ ing and even food-storing behaviors associated with foraging,
that become active when workload increases, and yet, the are performed at all times of the day by the same group of
literature on this topic is surprisingly inconsistent. In fact, individuals, suggesting an absence of circadian rhythm for the-
several studies that explicitly set out to test the reserve worker se internal workers (Moore et al. 1998). This pattern was con-
hypothesis fail to support it (Fewell and Winston 1992; John- sistent across all workers, suggesting an absence of ‘shift work’
son 2002; Jandt et al. 2012). Moreover, there is evidence that among in-hive workers (Moore et al. 1998). In comparison to
when workload increases, workers other than inactive workers honey bees, ant circadian rhythms have rarely been studied
increase their activity (Mirenda and Vinson 1981; Johnson (although see North 1987; North 1993; Sharma et al. 2004;
2002), or that workers are incapable of reallocating workers Lone and Sharma 2011; Lone and Sharma 2011).
to necessary tasks, even at the expense of losing half of their Shift work (individually different circadian rhythms in ac-
brood (Kwapich and Tschinkel 2013). tivity) could be an efficient way to allocate work in cases
Additional ecological functions that have been proposed where tasks require constant attention (e.g., brood requiring
include a worker/colony conflict involving inactive workers constant attention from nurses) or to increase the amount of
acting selfishly by conserving their energy and minimizing work that can be accomplished in a limited workspace (e.g.,
Behav Ecol Sociobiol

nurses caring for brood may need to move around the brood Behavioral data collection
pile and the presence of additional workers would only serve
to slow them down). Furthermore, if workers show consisten- Workers were individually painted 3–7 days before filming
cy in their circadian rhythm with regard to when they are with unique combinations of four paint spots, one on the head,
active and when they are resting, and individuals vary in the one on the thorax, and two on the abdomen, so that they could
timing of activity phases (thus effectively working in stag- be individually identified and tracked (Fig. S 1b). Videos (5-
gered shifts), then monitoring worker behavior only at certain min long) of normal colony activity were taken with an HD
times (e.g., only during the day) will generate the appearance camera (Nikon D7000 with 60 mm lens) at six time points
of consistently more or less active individuals. This would be throughout the day: 8 a.m., 12 p.m., 4 p.m., 8 p.m., 12 a.m.,
the case even if, over a 24 h period, all workers essentially and 4 a.m., repeated on 3 days spread over a 3-week period for
spend the same amount of time active. Because of this, it is each colony (18×5 min videos per colony). During the night
critical to monitor workers on different time scales, including videos, minimal white CFL lighting was used (switched on 5–
at different times of day, to establish whether individual dif- 10 min before the start of recording and switched off 15–20 min
ferences in activity level exist. after) because red light or infrared would have made reading the
Here, we investigate the structure of DOL, the degree of color coded ants impossible, thus preventing measurement of
individual task specialization, and worker variation in the ant individual level activity. All videos were taken within 3 months
Temnothorax rugatulus. We pay special attention to the be- of their collection to limit potential laboratory effects, such as
havioral state/task of inactivity. Specifically, we quantify DOL artificial age structures due to increased forager age.
and specialization, and describe how tasks relate to each other
by determining how the likelihood of performing one task Video analysis
affects the likelihood of performing other tasks, including be-
ing inactive. We investigate temporal variation in worker ac- For each ant, the task it performed was recorded at every
tivity levels, testing for worker circadian rhythms and whether second by an observer analyzing the videos. A complete list
individual consistency in activity level can be explained by of tasks and definitions can be found in Table 1. The tasks
workers taking turns resting or effectively doing shift work. were broadly classified as being either ‘active’ (e.g., brood
care), ‘undifferentiated’ (walking inside the nest with no clear
task), or ‘inactive’ (completely immobile), comparable to the
Methods broad classification used by Cole (1986). If less than 10 s
separated two events of brood care, feeding, foraging, or
Colony collection and housing 20 s for building, the task was considered to be uninterrupted.
Videos were analyzed by seven different observers. Data from
We collected five colonies of T. rugatulus ants in the Santa Cat- each video were spot checked by a single person to ensure
alina Mountains near Tucson, Arizona, USA in pine forest at an uniformity of behavioral observations.
altitude of approximately 2500 m (see Table S 1 for detailed
information about collected colonies). Colonies were housed in Statistical analyses
artificial nests that emulate the small rock crevices they inhabit in
the field (Charbonneau et al. 2015). The artificial nests consist of Statistical analyses were performed in R (Version 3.0.3) and
a 2-mm-thick piece of cardboard sandwiched between two glass consisted of mixed-effects models and Tukey’s post hoc tests
slides (76.2×50.8 mm). The cardboard (38.1×50.8 mm, or half (packages ‘nlme’ v3.1–115 and ‘multcomp’ v1.3–2), principal
of the size of the glass slides) serves as a spacer between the glass component analyses (PCA) (base ‘stats’ package and ‘prcomp’
slides and as a back wall. Additional walls are constructed by the function), hierarchical cluster analysis (base ‘stats’ package and
ants from grains of green ceramic-coated sand (Fig. S 1a). In the ‘hclust function), Pearson’s correlations (base ‘stats’ package
field, colonies create similar walls from surrounding sand and and ‘cor’ function), and repeatability measures (modified pack-
dirt. Artificial nests are kept in open-top plastic containers (11.1× age ‘rptr’ v0.6.405). DOL was quantified using the index
11.1×3.3 cm) whose walls were lined with ‘insect-a-slip’ established in Gorelick et al. (2004); for updated definitions
(BioQuip product #2871A). see also Gorelick and Bertram 2007; Dornhaus et al. 2009).
Colonies were given water and fed ad libitum (water-filled For the PCA, data from all five colonies were aggregated to
plastic test tubes, stoppered with cotton balls, semiweekly, and increase sample size (PCAs on individual colonies show sim-
2 mL Eppendorf tube of honey water and 10 frozen adult ilar patterns to the pooled PCA–Fig. S 2). The aggregated data
Drosophila flies weekly), kept on a 12 h light regimen (lights were centered and scaled by subtracting the mean times spent
on at 8 a.m. and off at 8 p.m.), and at constant temperatures on a task and dividing by the standard deviation. Only daytime
(approximately 21–24 °C) and humidity (approximately 20– observations were used to avoid variation due to circadian
25 % relative humidity). rhythms, and workers with less than three observations (40
Behav Ecol Sociobiol

Table 1 List of possible behaviors observed during video analysis, their broad class of activity, codes, and detailed descriptions. For every second of
analyzed video, each ant has one of these behaviors attributed to it. (Similar to Charbonneau et al. 2015)

Class Task Definition

Active Nest building Manipulating a stone in any way (moving, pushing, and pulling)
Foraging Located in feeding area or on water tube or wandering outside of
the nest and not engaged in building. Also if returning to the colony
from foraging areas and performing trophallaxis or returning with drosophila.
Brood care Manipulating brood (feeding, grooming, and moving)
Self-grooming Grooming itself
Grooming other (giver) Grooming another ant
Grooming other (receiver) Be groomed by another ant
Trophallaxis Receive or give liquid food to/from another adult ant
Eating Feeding on drosophila inside nest (brought back by foragers)
Undifferentiated Wandering inside nest Anytime an ant is mobile inside the nest wall and not engaged in any ‘active’ task
Inactive Inactive Immobile and not engaged in any ‘active’ task

workers out of a total of 265) were excluded from the PCA in on all tasks, except trophallaxis and brood care which are
order to minimize random variation. This data set was also not correlated to inactivity (Fig. 1). Although this may not
used in a hierarchical cluster analysis. Rotations were not nec- be surprising because inactivity is the lack of doing a task, a
essary as only eight variables were used, and the spread be- possible positive correlation might have emerged if highly
tween variable vectors was good. The first three components inactive workers were at the same time highly specialized
of the PCA were retained (Kaiser–Guttman stopping rule, on another task, such as brood care; e.g., it is possible for a
Guttman 1954). These explain 62.2 % of inter-worker varia- narrowly specialized ant to be doing nothing most of the
tion in time spent on tasks. time but if all its remaining time is spent on brood care, that
For balance in analyses of temporal variation in individual- ant could still be doing most of the brood care compared to
level inactivity, we retained only workers identified at least its nestmates, particularly because it is the rank, not the
once at each time period (day and night—across colonies, a absolute fraction of time spent on a task that is analyzed.
mean of 68 % retained, SD 15 %). In all analyses, we used This however was not the case, and highly inactive workers
only worker data because queens are thought to have smaller tended not to spend much time on any other specific tasks
task repertoires (Herbers 1983) and their sample size does not compared to their nestmates.
allow independent analyses. Brood care was negatively correlated to foraging and pos-
Repeatability was calculated according to the methods de- itively correlated to feeding on Drosophila flies brought back
scribed in Nakagawa and Schielzeth (2010) where fixed ef- to the nest by foragers (i.e., ‘eating’). Foraging, building,
fects are used for systematic effects across individuals (e.g., grooming, trophallaxis, and wandering inside are all generally
time of day), while random effect is used for non-systematic positively correlated. Building and foraging were positively
variation (e.g., if the observer identity was not systematically correlated to eating, which is surprising since brood care is
related to repeated measurements taken) (see Nakagawa and also positively correlated to eating, but foraging and brood
Schielzeth 2010). Thus, our model includes individual worker care are negatively correlated.
IDs as the random effect (and grouping variable), while colo- A PCA performed on the amount of time workers spent
ny, which is expected to have a systematic effect on time on each observed task shows that workers vary widely
budgets because colonies differ in their overall level of activ- along an inactivity-wandering inside axis that closely
ity, is a fixed effect variable. tracks PC1 (Fig. 2, left). Because inactivity is nearly or-
thogonal to both PC2 and PC3, the relationships between
task vectors on these axes (PC2 and PC3) highlight rela-
Results tionships between tasks independently of inactivity. There
appear to be three distinct groups of tasks branching out
Overall structure of division of labor in different directions: (1) foraging/building/trophallaxis,
(2) brood care/eating, and (3) grooming/wandering inside
Rank correlations of mean worker time spent on tasks (Fig. 2, right). Vector loadings and eigenvalues for the
(ranks calculated within colonies) showed that across ants, principal components are shown in Table 2 (loadings
time spent inactive was negatively correlated to time spent and eigenvalues for PC1-8 can be found in Table S 2).
Behav Ecol Sociobiol

Fig. 1 Pearson’s ρ for

correlations between worker
ranks of time allocated to tasks.
Bars in dark gray represent
contrasts that were significant
(p<0.05 after Holms correction)
and white bars are non-
significant. Inactivity is
negatively correlated with all
other tasks, except for trophallaxis
and brood care for which the
correlation is non-significant

A subsequent hierarchical cluster analysis (Euclidian dis- This shows a similar pattern of task relationships as the PCA
tance and Ward’s linkage method) classified workers (from all where inactivity and other tasks are first separated, then nurse
colonies, data scaled, and centered) into four separate groups tasks (brood care and eating), foraging tasks (foraging and build-
based on the proportion of time spent on each task (cluster ing), and generalist tasks (trophallaxis, wandering inside, and
number based on the four task vector groups from the PCA): grooming). Interestingly, here trophallaxis was grouped with
nurses (34 workers), extra-nest workers (26 workers), gener- inactive workers rather than with foraging tasks as in the PCA.
alists (62 workers), and inactive workers (103 workers; We found weak worker specialization (workers focusing
Fig. 3). In the analysis, workers were first separated into active on few tasks: median DOLind for all colonies 0.208), but high
and inactive workers (first branch of left dendrogram) and segregation in which workers perform which tasks (median
active workers are subsequently divided into generalists, DOLtask 0.838). Overall, we found a median DOLtotal 0.425, a
nurses, and extra-nest workers. Tasks were also placed into a measure that incorporates specialization and task segregation,
hierarchical dendrogram (Euclidian distance and Ward’s link- for T. rugatulus. Because these indices of DOL used are nor-
age method), but not explicitly clustered (top dendrogram). malized for number of tasks and number of individuals, they

Fig. 2 Principal component analysis on mean worker time allocated to care/eating, and (3) grooming/wandering inside. Hierarchical clustering
tasks shows that (left) workers strongly vary along an inactivity/ analysis shows four separate worker groups: + nurses (n=24), *extra-nest
wandering inside axis, approximately parallel to PC1. Also, as workers (n=27), black circle patrollers/groomers (n=56), black triangle
inactivity is nearly parallel to PC1 and orthogonal to PCs 2 and 3, the inactives (n=118). Data were mean-centered and scaled (subtracted the
figure on the right (PC2 vs PC3) shows the approximate relationships mean activity in a task and divided by the standard deviation) before
between task time allocations, independently of inactivity. Specifically, analysis. Analysis includes only daytime data for workers having a
there appear to be three groups of tasks (tasks parallel to each other and minimum of 3 observations
apart from other groups): (1) foraging/building/trophallaxis, (2) brood
Behav Ecol Sociobiol

Table 2 Eigenvector loadings, eigenvalues, and proportion and the ant Camponotus festinatus (0.15–0.25; Dornhaus A, Duffy
cumulative proportion of variance for the principal components retained
K, unpublished data).
(PCs 1–3)

Eigenvectors PC1 PC2 PC3

Inter-worker variation in worker time budgets
Foraging 0.202 0.428 0.519
Eating 0.251 −0.393 0.473
Linear-mixed effects models show significant variation be-
Brood care 0.276 −0.595 0.146
tween workers for all tasks, including inactivity, but not for
Grooming 0.355 0.189 −0.436
trophallaxis (LMM p<0.0001 for all tasks, except p=0.038
Wandering inside 0.449 −0.011 −0.478
for grooming, and p=0.64 for trophallaxis; Random Effect:
Inactivity 0.237 0.323 0.194
Colony). This indicates that individual workers differ in their
Trophallaxis −0.642 0.037 −0.053 propensity to engage in each task with the exception of troph-
Building 0.150 0.408 0.166 allaxis (which is the way ants receive and give liquid food).
PC1 PC2 PC3 Workers differed significantly in their level of inactiv-
Eigenvalue 2.356 1.554 1.065 ity, and these differences were consistent over the 2-week
Prop. of Variance 0.295 0.194 0.133 period of observation (LMM F=2.14 p<0.0001, Table 3;
Cumul. Prop. Var. 0.295 0.489 0.622 see Fig. 4 for distributions of worker inactivity for each
colony). Overall, 25.1 % of workers (of a total of 265
recorded) were never observed being anything but inac-
can be directly compared with other systems (Gorelick et al. tive over the 18 sample observations of 5 min each, while
2004). Our DOLtotal values were higher than those reported 2.6 % of workers were observed being constantly active
for Temnothorax albipennis (0.38, Dornhaus et al. 2009), sol- over this period (i.e., engaged in ‘active’ tasks according
itary and communal halictine bees (~0.08–0.21, Jeanson et al. to Table 1). Over the observation period, 71.9 % of
2007), Bombus impatiens (0.09–0.12, Jandt et al. 2009), and workers were inactive at least 50 % of the time.

Fig. 3 Hierarchical cluster analysis (Euclidian distance, Ward’s linkage (foraging and building), and other tasks (trophallaxis, wandering inside
method) of workers according to time spent on tasks (left dendrogram) and grooming). Z-scores indicate whether individual workers (rep-
shows that workers are first separated into active and inactive workers resented by thin colored lines in the central graph) spend more time
(first branch) and active workers are subsequently divided into patrollers/ (positive z-score —red), less time (negative z-score—blue) or equal time
groomers, nurses and extra-nest workers. Clustering of tasks shows a (zero z-scorewhite) than the mean amount of time spent on that tasks for
similar pattern (top dendrogram) where inactivity and other tasks are first all workers
separated, then nurse tasks (brood care and eating), extra-nest tasks
Behav Ecol Sociobiol

Table 3 Individual workers (Ind_Ant) consistently differ in the

proportion of time they are inactive, and overall inactivity is lower
during the day compared to night (timeperiod) the lack of interaction
between these factors suggests the absence of ‘shift work’. Results from
a linear mixed-effects model*

Parameter numDF denDF F-value p value

(Intercept) 1 2406 576.02 <0.0001

Ind_Ant 264 2406 2.14 <0.0001
Timeperiod 1 2406 7.24 0.0072
Ind_Ant:Timeperiod 264 2406 1.07 0.2068

*Fixed-effects: Ind_Ant, timeperiod, and Ind_Ant x timeperiod.

Random effects: colony/date.

Typically specialized tasks (foraging, building, and

brood care) are shown to be highly repeatable (i.e., re-
peatability test (Nakagawa and Schielzeth 2010) shows
that intra-worker variation is much lower than inter-
worker variation; Fig. 5), indicating that the amount of Fig. 5 Foraging, building, brood care, and inactivity have comparably
time workers spend on these tasks is highly consistent. high levels of repeatability. All task repeatability estimates are significant.
Tasks such as eating, grooming, and wandering inside Colony was added as a fixed effect to the calculation of adjusted
have low repeatability, suggesting that most workers repeatability to control for inter-colony variation. All tasks were
significantly repeatable (p value <0.05)
perform them to some extent, with little consistent dif-
ferences among workers. Most interestingly, inactivity is
also highly repeatable, on par with tasks typically per- Does individual inactivity depend on the time of day?
formed by specialized workers such as foraging, build- Circadian rhythm
ing, and brood care.
Workers spent significantly less time inactive during ‘day-
time’ (i.e., observations with lights on; 8 a.m., 12 p.m., and

Fig. 4 Frequency distribution of individual inactivity for each colony measured across 3 days and 6 time points each day. The number of workers and the
mean observation time (seconds/ant) is shown for each colony in the upper left corners
Behav Ecol Sociobiol

4 p.m.) than during ‘nighttime’ periods (i.e., with the lights Table 4 Individual workers (Ind_Ant) consistently differ in the
proportion of time they are inactive (Ind_Ant) and overall inactivity
off; 8 p.m., 12 a.m., and 4 a.m.) (LMM F=7.24 p<0.01;
varies over 4-h intervals (timepoint). The lack of interaction between
random effects: colony/date, Table 3). Extra-nest activity these factors suggests the absence of ‘shift work’ at this timescale. Results
(‘foraging’ and ‘building’) and ‘wandering inside’ were each from a linear mixed-effects model*
significantly higher during ‘daytime’ observations (p<0.01,
Parameter numDF denDF F value p value
mean of 5.2 % of time during day vs. 3.6 % at night;
p<0.01, mean of 16.5 % of time during day vs. 14.0 % at (Intercept) 1 1303 488.94 <0.0001
night, respectively), while time spent on in-nest activities Ind_Ant 182 1303 2.11 <0.0001
was not significantly different (p=0.41, mean of 15.7 % of Timepoint 5 1303 7.36 <0.0001
time during day vs. 16.5 % at night; Fig. 6). Ind_Ant:Timeperiod 910 1303 0.97 0.6620
Overall worker inactivity also varies significantly at shorter
timescales over the course of the day (i.e., among 4 h intervals *Fixed-effects: Ind_Ant, timepoint, and Ind_Ant x timeperiod.
at which the videos were recorded; LMM F=7.36 p<0.0001; Random effects: colony/date.
random effects: colony/date, Table 4). However, the only sig-
nificant contrast at this level is between 8 p.m. and 12 a.m. the differences do not depend on the time of day they
observations (Fig. 7). were observed. Thus, workers do not have complementing
or staggered circadian rhythms as would be expected if
they were working in ‘shifts’, and all workers generally
Do workers differ in circadian rhythm, employing ‘shift
increase the proportion of time spent inactive from day-
time to nighttime. This suggests that daytime inactivity is
positively correlated to nighttime inactivity and the least
Above, we have discussed that the proportion of time spent in
inactive workers during the day will also be the least
the inactive state in a sample is significantly affected by the
active at night (further confirmed by a Pearson correlation
identity of the ant and by the time of day (main effects in
on within-colony worker inactivity ranks: Fig. 8, ρ 0.45,
Table 2). If ants were working in ‘shifts’, i.e., taking turns
p<0.0001, n=265); however, there is considerable varia-
resting, we would not necessarily expect these main effects;
tion (see Fig. 9 for individual changes in inactivity).
moreover, we would expect a significant interaction be-
tween worker identity and time, to reflect the fact that
different individuals are inactive at different times. This
is not the case: ant identity and time period (‘daytime’ vs. Discussion
‘nighttime’, and all 4-h interval time points compared to
each other) show no significant interaction in their effect Our results show that workers differ in the tasks or groups of
on level of inactivity (p = 0.21 and 0.66, respectively, tasks they allocate most time to. Workers can be grouped into
Tables 3 and 4). This indicates that, although workers external workers (who forage and build the nest), nurses, pa-
may differ in the proportion of time they are inactive, trollers/groomers, and inactive workers. Perhaps the most sur-
prising result of this study is that inactivity is highly repeatable
and explains a large portion of inter-worker variation, on par
with specialized tasks such as foraging, building, and brood

Fig. 6 Significant difference in extra-nest tasks, in-nest tasks, and

inactivity between day and night (p<0.01, 0.41, and <0.01). Boxplots
show the lower and upper quartiles (box), median (horizontal line
within box), and extremes (whiskers) for worker inactivity at each time Fig. 7 Variation of worker inactivity over 4-h intervals. Boxplot shows
period (day is light grey, night is dark grey). n.s. p>0.05, *p<0.05, the lower and upper quartiles (box), median (horizontal line within box),
**p<0.01, and ***p<0.001 and extremes (whiskers) for worker inactivity at each time point
Behav Ecol Sociobiol

time), involve very little movement and so would register little

activity according to Cole’s measure of activity.
Nonetheless, our observations of short timescale variation
in inactivity are consistent with the idea of colony-wide activ-
ity fluctuations, in that we find strong consistency within 5-
min intervals and apparently stochastic variation across such
intervals. However, our observation timescale (5 min videos at
4 h intervals) was not designed to test for a periodicity over
intervals such as 26 min (as found in Cole 1991).
Observations of the temporal patterns of inactivity show
higher levels of inactivity during the night, suggesting a circa-
dian rhythm. Furthermore, more time is spent on extra-nest
activities such as foraging activity during the day than during
Fig. 8 Within colony inactivity ranks are consistent between ‘daytime’
the night. However, in-nest tasks show no such periodicity.
and ‘nighttime’ observations. Pearson correlation performed on within-
colony worker inactivity ranks (Pearson’s ρ 0.45, p<0.0001) Earlier work on social synchronization in honey bees suggested
the existence of a ‘colony-clock’, where worker activity cycles
were coordinated within colonies (Frisch and Koeniger 1994).
care. This suggests that some workers are much more likely to More recent work suggests that this may not the case. In fact,
be inactive than other workers, effectively ‘specializing’ on honey bee foragers show strong evidence of periodicity in their
inactivity, and these differences appear to be stable over at activity patterns where they are more active during the day, but
least a 2-week period. Whether worker inactivity changes at in-nest workers, including workers involved in food storage, do
longer timescales, such as seasonally or ontogenetically, re- not (see Moore 2001 for a full discussion on this topic). A
mains to be tested. Thus, individual-level differences in activ- similar phenomenon has been shown in bumble bees where
ity are a real phenomenon which cannot be explained by tem- foragers have been shown to have diurnal rhythms while in-
poral variation in inactivity over the course of the day. This nest workers do not (Yerushalmi et al. 2006) but, prior to this
suggests that inactivity is likely not the result of constraints, study, these patterns had not been shown in ants.
such as need for rest, delays between tasks, time for digestion, Why are foragers more active during the day? Honey bee
or other necessary side-effects of other activities, all of which and bumble bee foragers rely heavily on visual cues to nav-
would be expected to affect all workers more or less equally. igate (Dyer and Could 1983) and cannot fly in total dark-
This raises the question of whether high levels of inactivity ness. They also adjust their activity according to daily floral
could be an adaptive element of DOL in social insect colonies. rhythms so as to maximize the amount of nectar and pollen
The level of inactivity of individual workers was shown collected during foraging trips (Moore et al. 1998), while
here to fluctuate over the course of the day at short timescales in-nest workers do not have these constraints. T. rugatulus
(5-min intervals). Cyclical fluctuations in colony-level activity ants have been shown to use visual cues in navigation
patterns had been shown in Temnothorax allardycei, where (Bowens et al. 2013), and so they may also benefit from
bursts of activity would occur approximately every 26 min foraging during the day, even if they do not need light sim-
(Cole 1991). Workers in isolation showed no such activity ply to move. In addition, brood may need tending around
cycles, and as group size increased, the variation in cycle the clock. Although we do not know the feeding frequen-
period decreased. As such, these cycles are thought to be the cies of larvae in T. rugatulus, we do know that in the fire ant
result of inter-individual interactions, rather than a circadian Solenopsis invicta, larvae are fed 2–50 times per hour, de-
pattern of activity. Additionally, there is evidence that in-nest pending on larval size and satiation (Cassill and Tschinkel
workers tend to have regular activity cycles, but that these can 1995) and larva feeding requirements are unlikely to
be disturbed by returning foragers (Boi et al. 1999). However, change overnight. Thus, nurses may need to be equally
both Cole (1991) and Boi et al. (1999) define activity as move- active at night as during the day.
ment rate, measured as pixel change rates. It is difficult to There is a long standing tradition of using multivariate
estimate how such measures of movement rate translate to analyses to study colony organization (Lenoir and Mardon
measure of behavioral activity discussed herein. For instance, 1978). Here, we use PCA to look at the relationships
in our data, time spent on ‘wandering inside’ (7.4 % of mean between tasks in a multivariate space. Our analyses show
worker time) is considered an undifferentiated activity and is that inactivity and wandering inside are almost parallel to
excluded both from active time and inactive time, but likely the first component. Although many active tasks correlate
accounts for much of the activity recorded by Cole (1991). negatively with inactivity, they show a large amount of
Furthermore, tasks which account for large portions of active variation that is independent of it (their vectors in the
time in our data, such as brood care (4.5 % of mean worker PCA are not parallel to the inactivity vector). This
Behav Ecol Sociobiol

suggests that the amount of time workers spend on a par- from active tasks suggests that may be a motivation other
ticular active task is not well predicted by their level of than a lack of available work for workers to spend more
inactivity. Instead, a relative independence of inactivity time on inactivity.
Behav Ecol Sociobiol

ƒFig. 9 Consistent differences are shown by workers (each represented by progression of workers from more central (e.g., nurses) to less
a very narrow box) having variable inactivity levels on each graph. Each central (e.g., foragers) specializations as predicted both by
worker is represented by a (very narrow) box, with the black line
indicating the median inactivity. Individual ants in both daytime and
most interpretations of age/temporal polyethism (e.g., Seeley
nighttime graphs are sorted from left to right in decreasing median time 1982; Gordon 1996) as well as the foraging-for-work task
spent inactive during the daytime. For daytime data (left graphs with light allocation strategy (Franks and Tofts 1994). Our results are
gray boxes), the median for workers on the left end of the graphs tends broadly in agreement with this body of work; however, we
towards 100 % time spent inactive while on the right end it tends towards
0 % time inactive (as indicated by the overall black median line generally
also show the presence of an additional group composed of
starting at 100 % and dropping to 0 %). If workers did not have significant highly inactive workers. There are only a few studies where
differences, we would expect similar medians for all workers (a inactivity is considered in colony organization and a distinct
horizontal black line) and largely overlapping boxes. The comparison ‘inactive’ group is found (Fresneau 1984; Corbara et al. 1989;
of the daytime (left graphs with light gray boxes) and nighttime (right
graphs with dark grey boxes) data are helpful for understanding the lack
Retana and Cerdá 1991). However, the data presented here is
of day/night shift-work. If all workers were either active during the day significantly more comprehensive (continuous tracking of
OR active during the night, we would expect graphs skewed in opposite complete colonies over 5 min spans at multiple time points
directions. If activity level of workers was consistent during day and across the day and over multiple days vs scan sampling of
night, as is predicted by the rank correlation in Fig. 8, we should see a
downward sloping median line on the right set of graphs that parallels the
subsets of workers over multiple days) which allows us to
median line on the left. Although this pattern is not obvious, there is a show that worker inactivity levels are highly consistent at
general trend of decreasing inactivity from workers on the left to those on multiple timescales (daily and over 2–3 weeks). The presence
the right for the nighttime graphs. Boxplots show the lower and upper of a patrolling/grooming groups is also somewhat interesting
quartiles (box), median (horizontal line within box), 1.5 interquartile
ranges (whiskers), and outliers (points) for each ant
as this is also a rarely described group which may serve a
similar function as that suggested by Johnson (2008) in hon-
eybees of global information collecting.
Looking at the organization of task vectors along PC2 and The results from the PCA generally agree with the inter-
PC3, we can look at the relationships between active tasks, task rank correlations, but provide slightly different insights.
independent of inactivity. There appear to be three groups of On the one hand, ranks are relative measures (to a worker’s
tasks (i.e., where task vectors within groups roughly parallel nestmates) of likelihood to engage in tasks, while the PCA
each other and are at an angle from other groups). This group- deals with absolute likelihood to engage in tasks. For example,
ing is further supported by the hierarchical cluster analysis a worker that does the most foraging and brood care in a
between tasks (top dendrogram) that first distinguishes be- colony (thus highly ranked) may nonetheless be doing very
tween active and inactive tasks, and subsequently between little foraging (i.e., spend little time foraging) as compared to
nurse tasks, foraging tasks, trophallaxis, and patrolling/ workers in other colonies, or even to how much total available
grooming. The main difference here is that trophallaxis seems time for work. Because the PCA uses absolute measures and
to be its own group, while in the PCA trophallaxis is grouped pools multiple colonies, observed trends could potentially be
with foraging tasks. the result of inter-colony variation rather than inter-worker
Overall, DOL in colonies of T. rugatulus appears to segre- variation. This is likely not the case because PCAs done on
gate work into task groups. Similarly to task clustering, individual colonies show similar patterns to the pooled PCA.
workers can first be separated into active and inactive Additionally, rank correlations that necessarily account for
workers, and subsequently active workers can be further sep- inter-colony variation generally agree with the PCA further
arated into three main task groups: external workers (building, suggesting that the patterns of worker variation are indeed
foraging, and trophallaxis), nurses (brood care and in-nest real.
feeding), and patrollers/groomers (wandering inside and feed- Highly repeatable tasks are tasks that workers tend to spe-
ing). That colonies might be organized in groups of workers cialize on and thus highly visible and tractable. This might
with similar task profiles is not a new idea. Many studies have explain why they are often the focus of DOL studies (Gordon
shown that monomorphic colonies could be grouped into a 1996; Beshers and Fewell 2001; Duarte et al. 2011). However,
number of behavioral (temporal) castes in ants (Wilson as suggested by the spread of workers along PC1 as well as the
1976; Mirenda and Vinson 1981; Calabi 1988; Corbara et al. task and worker hierarchical cluster analyses, these tasks tend
1989; Retana and Cerdá 1990; Retana and Cerdá 1991). to only be done by active workers. Thus, studies that uniquely
Workers tend to be clustered into either two groups (intra- focus on these tasks are studying a biased subsample of the
and extra-nidal; Lenoir and Ataya 1983; Pamminger et al. colony’s workers and consequently only looking at one di-
2014), or three groups (extra-nidal, nurse, and social/ mension of worker variation. Tasks such as eating, grooming,
generalist; Mirenda and Vinson 1981; Herbers 1983; Mersch wandering inside may not be very repeatable, and thus not
et al. 2013), though some have found many more (e.g., very specialized, but most workers do them to some extent
Fresneau 1984 found five groups; Corbara et al. 1989 found and are therefore very relevant to understanding worker time
six groups). Generally, worker groups seem to represent the budgets (i.e., how workers allocated their time to tasks)
Behav Ecol Sociobiol

because the time spent on these tasks is time that cannot be extra-nest workers, nurses, and patrollers/groomers, which
spent on other, more specialized tasks. Furthermore, the dif- tracks predictions from age/temporal polyethism and/or forag-
ference between specialized and generalized tasks may reflect ing for work. We also showed the occurrence of a less com-
a difference between self-serving and colony-serving tasks. monly described group: the inactive workers. Indeed, inactiv-
Generalized tasks such as eating and grooming tend to be ity was shown to be a highly consistent individual behavior on
actions that most workers will need to do for themselves which a subset of workers effectively ‘specialize’. Incorporat-
(self-serving) and contribute to individual function, while spe- ing inactivity in future studies of task allocation and DOL may
cialized tasks such as foraging and brood care tend to be ac- be important as we show that inactives are a distinct group of
tivities thought to contribute to colony function (colony-serv- workers with their own sets of behaviors and should likely not
ing). Additionally, self-serving tasks have been linked to in- be either ignored for lack of undertaking ‘active’ tasks, or be
activity and worker reproduction (Hillis et al. In prep.). counted as less efficient workers in typically described groups
At first glance, inactivity may be thought of as the lack of such as nurses and foragers.
activity and thus may seem trivial. However, we show that
there is a subset of workers that effectively ‘specialize’ on
inactivity. Studies that uniquely focus on active or specialized Acknowledgments We thank Alex Downs, Andrew Scott, Mary
Levandowski, Matthew Velazquez, Neil Hillis, and Nicole Fischer for
tasks are studying a biased subsample of the colony’s workers; their help with ant painting and maintenance, and data collection. We also
by excluding inactivity, an entire dimension of task allocation thank the entire Dornhaus lab for their ongoing feedback. Research sup-
is eliminated. This would be like using the left-hand graph of ported through the GIDP-EIS and EEB Department at University of Ar-
Fig. 2 to understand colony organization without knowing that izona, as well as NSF grants no. IOS-1045239, IOS-0841756, and DBI-
1262292 (to A.D.).
there is a rich third dimension in which workers vary in inac-
tivity, independently of active tasks. How can we expect to
obtain a comprehensive understanding of the tradeoffs in-
volved in allocating time to tasks when we ignore inactives, References
which compose close to half of the colony, or simply lump
them in with other workers? Becker GS, Murphy KM (1992) The division of labor, coordination costs,
and knowledge. Q J Econ 107:1137–1160
The question of consistent individual differences in behav- Bell AM, Hankison SJ, Laskowski KL (2009) The repeatability of be-
ior is central to the studies of personality and behavioral syn- haviour: a meta-analysis. Anim Behav 77:771–783
dromes (Dingemanse et al. 2002; Bell et al. 2009; Pearish Beshers SN, Fewell JH (2001) Models of division of labor in social
et al. 2013). In recent studies, personality is broadly defined insects. Annu Rev Entomol 46:413–440
as a behavior that is consistent through time and across situa- Boi S, Couzin ID, Del Buono N et al (1999) Coupled oscillators and
activity waves in ant colonies. Proc R Soc Lond B Biol Sci 266:
tions (Bell et al. 2009). This definition can sometimes be 371–378
problematic because the timescale over which behavior needs Bowens SR., Glatt DP., Pratt SC. (2013) Visual navigation during colony
to be consistent is open-ended, and there are no clearly emigration by the ant temnothorax rugatulus.
established criteria on what is a biologically meaningful time- Calabi P (1988) Behavioral flexibility in hymenoptera: a re-examination
scale (Stamps and Groothuis 2010; Dall and Griffith 2014). of the concept of caste. Adv Myrmecol 237–258
Cassill DL, Tschinkel WR (1995) Allocation of liquid food to larvae via
Early personality literature stated that ‘temperament’, the trophallaxis in colonies of the fire ant, solenopsis invicta. Anim
foundation of personality, is reflected in early appearing ten- Behav 50:801–813
dencies that continue through the life of an individual Charbonneau D, Dornhaus A (In revision) When doing nothing is some-
(reviewed in Gosling 2001). Although it may seem that we thing. How task allocation mechanisms compromise between flex-
could use the term ‘personality’ to describe observed consis- ibility, efficiency, and inactive agents.
Charbonneau D, Hillis N, Dornhaus A (2015) BLazy^ in nature: ant
tent levels of worker inactivity, we hesitate to do so because colony time budgets show high’inactivity’ in the field as well as in
we can easily imagine scenarios where the behaviors are con- the lab.
sistent over the observed timescale (~weeks), but not over the Cole BJ (1986) The social behavior of leptothorax allardycei (hymenop-
lifetime of individuals. For example, inactivity could conceiv- tera, formicidae): Time budgets and the evolution of worker repro-
duction. Behav Ecol Sociobiol 18:165–173
ably be the result of inexperience or immaturity in younger
Cole BJ (1981) Dominance hierarchies in leptothorax ants. Sci NY 212:
workers or senescence in older workers (Seid and Traniello 83
2006). The timescale of our experiment (a few weeks) would Cole BJ (1991) Short-term activity cycles in ants: Generation of period-
not have changed the status of Temnothorax workers, who are icity by worker interaction. Am Nat 137:244–259
thought to live for years (which makes a longitudinal study Corbara B, Lachaud J-P, Fresneau D (1989) Individual variability,
difficult in this species), with regard to whether they are young social structure and division of labour in the ponerine ant
ectatomma ruidum roger (hymenoptera, formicidae). Ethology
or senescing workers. 82:89–100
In this study, we showed that colony organization in Dall SR, Griffith SC (2014) An empiricist guide to animal personality
T. rugatulus ants resembles typically observed worker groups: variation in ecology and evolution. Behav Evol Ecol 2:3
Behav Ecol Sociobiol

Dingemanse NJ, Both C, Drent PJ et al (2002) Repeatability and herita- Jandt J, Robins N, Moore R, Dornhaus A (2012) Individual bumblebees
bility of exploratory behaviour in great tits from the wild. Anim vary in response to disturbance: a test of the defensive reserve hy-
Behav 64:929–938 pothesis. Insect Soc 59:313–321
Dornhaus A (2008) Specialization does not predict individual efficiency Jeanson R, Fewell JH, Gorelick R, Bertram SM (2007) Emergence of
in an ant. PLoS Biol 6, e285 increased division of labor as a function of group size. Behav Ecol
Dornhaus A, Holley JA, Franks NR (2009) Larger colonies do not have Sociobiol 62:289–298
more specialized workers in the ant temnothorax albipennis. Behav Johnson BR (2008) Global information sampling in the honey bee.
Ecol 20:922–929 Naturwissenschaften 95:523–530
Duarte A, Weissing FJ, Pen I, Keller L (2011) An evolutionary perspec- Johnson BR (2002) Reallocation of labor in honeybee colonies during
tive on self-organized division of labor in social insects. Annu Rev heat stress: the relative roles of task switching and the activation of
Ecol Evol Syst 42:91–110 reserve labor. Behav Ecol Sociobiol 51:188–196. doi:10.1007/
Durkheim E (1997) The division of labor in society. Simon and Schuster, s00265-001-0419-1
New York, NY Johnson S (2012) Emergence: the connected lives of ants, brains, cities,
Dyer FC, Could JL (1983) Honey Bee navigation: the honey bee’s ability and software. Simon and Schuster, New York, NY
to find its way depends on a hierarchy of sophisticated orientation Klein BA, Gibbs AG, Larsen KMF (2003) Signatures of sleep in the
mechanisms. Am Sci 71:587–597 paper wasp Polistes flavus. The 2003 ESA Annual Meeting and
Fellers JH (1989) Daily and seasonal activity in woodland ants. Exhibition.
Oecologia 78:69–76 Klein BA, Klein A, Wray MK et al (2010) Sleep deprivation impairs
Fewell JH, Winston ML (1992) Colony state and regulation of pollen precision of waggle dance signaling in honey bees. Proc Natl
foraging in the honey bee, apis mellifera L. Behav Ecol Sociobiol Acad Sci 107:22705–22709
30:387–393 Klein BA, Olzsowy KM, Klein A et al (2008) Caste-dependent sleep of
Franks N, Tofts C (1994) Foraging for work—How tasks allocate worker honey bees. J Exp Biol 211:3028–3040
workers. Anim Behav 48:470–472. doi:10.1006/anbe.1994.1261 Klein BA, Seeley TD (2011) Work or sleep? honeybee foragers opportu-
Fresneau D (1984) Développement ovarien et statut social chez une nistically nap during the day when forage is not available. Anim
fourmi primitiveNeoponera obscuricornis emery (Hym. Behav 82:77–83
Formicidae, ponerinae). Insect Soc 31:387–402 Kraus FB, Gerecke E, Moritz RFA (2011) Shift work has a genetic basis
Frisch B, Koeniger N (1994) Social synchronization of the activity in honeybee pollen foragers.
rhythms of honeybees within a colony. Behav Ecol Sociobiol 35:
Kwapich CL, Tschinkel WR (2013) Demography, demand, death, and the
seasonal allocation of labor in the Florida harvester ant
Gadagkar R, Joshi NV (1984) Social organisation in the Indian wasp
(Pogonomyrmex badius). Behav Ecol Sociobiol 67:2011–2027
ropalidia cyathiformis (Fab.) (hymenoptera: vespidae). Z Für
Lenoir A, Ataya H (1983) Polyéthisme et répartition des niveaux
Tierpsychol 64:15–32
d’activité chez la fourmi lasius niger L. Z Für Tierpsychol 63:213–
Gerkey BP, Matarić MJ (2004) A formal analysis and taxonomy of task
allocation in multi-robot systems. Int J Robot Res 23:939–954
Lenoir A, Mardon JC (1978) Note sur l’application de l’analyse des
Gordon DM (2002) The organization of work in social insect colonies.
correspondances a la division du travail chez les fourmis.
Complexity 8:43–46
Lindauer M (1952) Ein beitrag zur frage der arbeitsteilung im bienenstaat.
Gordon DM (1996) The organization of work in social insect colonies.
J Comp Physiol A Neuroethol Sens Neural Behav Physiol 34:299–
Nature 380:121–124
Gorelick R, Bertram SM (2007) Quantifying division of labor: borrowing
tools from sociology, sociobiology, information theory, landscape Lone SR, Sharma VK (2011) Timekeeping through social contacts: social
ecology, and biogeography. Insect Soc 54:105–112 synchronization of circadian locomotor activity rhythm in the car-
Gorelick R, Bertram SM, Killeen PR, Fewell JH (2004) Normalized penter Ant camponotus paria. Chronobiol Int 28:862–872
mutual entropy in biology: quantifying division of labor. Am Nat Mersch DP, Crespi A, Keller L (2013) Tracking individuals shows spatial
164:677–682 fidelity is a Key regulator of Ant social organization. Science 340:
Gosling SD (2001) From mice to men: what can we learn about person- 1090–1093. doi:10.1126/science.1234316
ality from animal research? Psychol Bull 127:45 Mirenda JT, Vinson SB (1981) Division of labour and specification of
Guttman L (1954) Some necessary conditions for common-factor analy- castes in the red imported fire ant solenopsis invicta Buren. Anim
sis. Psychometrika 19:149–161 Behav 29:410–420
Herbers JM (1981) Time resources and laziness in animals. Oecologia 49: Moore D (2001) Honey bee circadian clocks: behavioral control from
252–262 individual workers to whole-colony rhythms. J Insect Physiol 47:
Herbers JM (1983) Social organization in leptothorax ants: within-and 843–857
between-species patterns. Psychol J Entomol 90:361–386 Moore D, Angel JE, Cheeseman IM et al (1995) A highly specialized
Hillis N, Charbonneau D, Nguyen H, et al. (In prep.) Are Blazy^ ants social grooming honey bee (hymenoptera: apidae). J Insect Behav 8:
selfish? Testing whether inactive ant workers invest more in their 855–861
own reproduction. Moore D, Angel JE, Cheeseman IM et al (1998) Timekeeping in the
Hölldobler B, Wilson EO (1990) The ants. Belknap press of harvard honey bee colony: integration of circadian rhythms and division of
university press. MA, Cambridge labor. Behav Ecol Sociobiol 43:147–160
Ishii Y, Hasgeawa E (2013) The mechanism underlying the regulation of Nakagawa S, Schielzeth H (2010) Repeatability for gaussian and non-
work-related behaviors in the monomorphic ant, myrmica kotokui. J gaussian data: a practical guide for biologists. Biol Rev 85:935–956
Ethol 31:61–69 North RD (1987) Circadian rhythm of locomotor activity in individual
Jandt JM, Dornhaus A (2011) Competition and cooperation: bumblebee workers of the wood ant Formica rufa. Physiol Entomol 12:445–454
spatial organization and division of labor may affect worker repro- North RD (1993) Entrainment of the circadian rhythm of locomotor ac-
duction late in life. Behav Ecol Sociobiol 65:2341–2349 tivity in wood ants by temperature. Anim Behav 45:393–397
Jandt JM, Huang E, Dornhaus A (2009) Weak specialization of workers O’Donnell S, Bulova SJ (2007) Worker connectivity: a review of the
inside a bumble bee (bombus impatiens) nest. Behav Ecol Sociobiol design of worker communication systems and their effects on task
63:1829–1836 performance in insect societies. Insect Soc 54:203–210
Behav Ecol Sociobiol

Oster GF, Wilson EO (1978) Caste and ecology in the social insects. Univ Schmid-Hempel P (1990) Reproductive competition and the evolution of
Pr, Princeton work load in social insects. Am Nat 501–526
Pamminger T, Foitzik S, Kaufmann KC et al (2014) Worker personality Seeley TD (1982) Adaptive significance of the age polyethism schedule
and its association with spatially structured division of labor. PLoS in honeybee colonies. Behav Ecol Sociobiol 11:287–293. doi:10.
ONE 9, e79616 1007/BF00299306
Pearish S, Hostert L, Bell AM (2013) Behavioral type–environment cor- Seid MA, Traniello JF (2006) Age-related repertoire expansion and divi-
relations in the field: a study of three-spined stickleback. Behav Ecol sion of labor in pheidole dentata (hymenoptera: formicidae): a new
Sociobiol 1–10 perspective on temporal polyethism and behavioral plasticity in ants.
Pinter-Wollman N, Hubler J, Holley J-A et al (2012) How is activity Behav Ecol Sociobiol 60:631–644
distributed among and within tasks in temnothorax ants? Behav Sendova-Franks AB, Hayward RK, Wulf B et al (2010) Emergency net-
Ecol Sociobiol 66:1407–1420 working: famine relief in ant colonies. Anim Behav 79:473–485
Pol R, de Casenave JL (2004) Activity patterns of harvester ants Sharma VK, Lone SR, Goel A, Chandrashekaran MK (2004) Circadian
pogonomyrmex pronotalis and pogonomyrmex rastratus in the cen- consequences of social organization in the ant species camponotus
tral monte desert, Argentina. J Insect Behav 17:647–661 compressus. Naturwissenschaften 91:386–390
Retana J, Cerdá X (1991) Behavioural variability and development of
Stamps J, Groothuis TG (2010) The development of animal personality:
cataglyphis cursor ant workers (hymenoptera, formicidae) 1).
relevance, concepts and perspectives. Biol Rev 85:301–325
Ethology 89:275–286
Retana J, Cerdá X (1990) Social organization of cataglyphis cursor ant Wilson EO (1991) Ants. Bull Am Acad Arts Sci 45:13–23
colonies (hymenoptera, formicidae): inter-, and intraspecific com- Wilson EO (1976) Behavioral discretization and the number of castes in
parisons. Ethology 84:105–122 an ant species. Behav Ecol Sociobiol 1:141–154
Robinson GE, Huang Z-Y (1998) Colony integration in honey bees: Yerushalmi S, Bodenhaimer S, Bloch G (2006) Developmentally deter-
genetic, endocrine and social control of division of labor. mined attenuation in circadian rhythms links chronobiology to so-
Apidologie 29:159–170 cial organization in bees. J Exp Biol 209:1044–1051
Rosengaus RB, Traniello JF (1991) Biparental care in incipient colonies Zhang J, Chen G (2011) The influence of logistics development on
of the dampwood termiteZootermopsis angusticollis Hagen manufacturing division. Artificial Intelligence, Management
(isoptera: termopsidae). J Insect Behav 4:633–647 Science and Electronic Commerce (AIMSEC), 2011 2nd
Samways MJ (1993) Insects in biodiversity conservation: some perspec- International Conference on. pp 791–794
tives and directives. Biodivers Conserv 2:258–282