View
0
Download
0
Category
Preview:
Citation preview
Pattern choices by bumblebees 1
Food restriction and threat of predation affect visual pattern choices by flower-naïve bumblebees
Service, E. W. & Plowright, C. M. S.*
School of Psychology
University of Ottawa
Ottawa, Ont. K1N 6N5
Canada
* Corresponding author.
Phone: 613-562-5800 x 4849
Fax: 613-562-5147
E-mail address: cplowrit@uottawa.ca
Wordcount: 5,353
1
2
3
4
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Pattern choices by bumblebees 2
ABSTRACT
The aim of this study was to determine whether a preference by flower-naïve bumblebees could
be created or enhanced by manipulating variables relevant to food collection and to defense
against predation. In two experiments, colonies of bumblebees (Bombus impatiens) were
deprived of pollen, exposed to CO2, or neither. Choices of individual workers in a radial arm
maze were monitored. In Experiment 1, both variables lead to a preference for corridors
occupied by a conspecific bee. The effect was specific: no change in preference for corridors
occupied by other objects (a coin and a piece of Styrofoam) was detected. In Experiment 2,
radial and concentric patterns were used, both of which were unoccupied. Only pollen
deprivation increased preference for radial stimuli, while CO2 had no discernible effect.
Preferences for visual patterns by bees leaving their colony for the first time are modulated by
variables that affect the internal state of the bees in problem-specific ways.
Keywords: Bumblebees; floral choice; carbon dioxide; social cognition; pattern recognition;
predation threat
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Pattern choices by bumblebees 3
Highlights
For colonies of bumblebees (Bombus impatiens), predation threat was simulated by
exposure to CO2
Need for food was increased through deprivation of pollen
Both variables affected choice by flower-naïve bees of visual patterns occupied by
another bee
Only food deprivation affected choice of unoccupied patterns with a floral resemblance
Pattern preferences by bumblebees leaving their nest for the first time are context specific
39
40
41
42
43
44
45
46
47
Pattern choices by bumblebees 4
The possibility that foraging bees use the presence of other insects on flowers to govern
their own floral choices has received little if any support in the field. For instance, one survey
of counts of floral visitors found on flower heads of sunflowers (Helianthus annuus L.) and
onion flowers (Allium cepa L.) addressed the question of whether there would be attraction or
avoidance of inflorescences that were already occupied by one or more individuals. The
distribution of insects, including honeybees (Apis mellifera L.) and bumblebees (Bombus spp
Latreille), followed a Poisson distribution (Tepedino & Parker, 1981): there were neither more
nor less flowers with just one forager than would be expected by chance. The notion that insects
forage independently of other insects may, however, have been dismissed prematurely by field
biologists. Recently, an examination of the details of social effects on foraging in the laboratory
has revealed that bees use associative learning to take advantage of cues that predict important
outcomes such as presence of floral reward (Avarguès-Weber & Chittka, 2014a; Dawson,
Avarguès-Weber, Chittka, & Leadbeater, 2013; Leadbeater & Chittka, 2009) and presence of
predators (Dawson & Chittka, 2014)—in general, social information is used by insects
strategically (Grüter & Leadbeater, 2014). Moreover, workers that have just left their colony for
the first time also show a significant preference for flowers that are already occupied by another
forager (Kawaguchi, Ohashi, & Toquenaga, 2006; Leadbeater & Chittka, 2009), though this
effect is most evident when the flowers are rare and the occupiers are large relative to the flowers
(Plowright et al., 2013). Because of the difficulty in obtaining the pre-experimental histories of
bees seen foraging in the field, these sorts of effects may have eluded detection in nature.
In this paper, our focus is on the behaviour of “flower-naïve” bumblebee workers: bees
that leave their colony for the first time, and as such, have had no prior experience with flowers.
One view regarding the preferences for occupied flowers by flower-naïve bees is that they may
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
Pattern choices by bumblebees 5
help inexperienced bees to locate sources of food: a pattern with another bee on it is likely to be
a flower. One difficulty with this interpretation is that it is post hoc. Indeed, the exact opposite
argument can be made a priori: a flower that is occupied by a forager is likely to be empty or
well on its way to being depleted, and so it ought not to be preferred but avoided. In other
words, other foragers may possibly act as competitors rather than as informers (Baude, Danchin,
Mugabo & Dajoz., 2011). Another view is that the presence of an occupier on a flower has little
to do with signalling resource availability. Field observations have suggested that interactions
among bees on a flower are in fact aggressive (Kikuchi, 1963). Indeed, in the course of a prior
study on pattern preferences of flower-naïve bees (Orbán & Plowright, 2013), we captured a few
such interactions on film (Videos 1 and 2 in the Supplementary Materials): the behaviours of the
bee landing on the flower seemed more directed at the occupier than at the flower, which
suggests that the occupier might have been perceived as a predator or a competitor.
In two experiments, we manipulated internal states of flower-naïve bees. To promote
food finding behaviours, we manipulated the availability of pollen, which is needed for feeding
to larvae. To promote aggressive or defensive behaviours, we exposed the bees to CO2 to
simulate the presence of a mammalian predator. Predators of bumblebee colonies (Goulson,
2010) include mice (Mus domesticus), badgers (Meles meles L.) in Europe, and skunks (Mephitis
mephitis Schreber) North-America. Bumblebees perceive CO2: they respond to mammalian
breath and to currents of air containing 5 or 10% CO2 by hissing, which serves as an inter-
specific defence signal (Kirchner & Röschard, 1999).
If a preference for occupied stimuli is engaged when bees are food searching, then pollen
deprivation should create or increase the preference. If the preference is engaged in situations
that trigger aggressive behaviours, then exposure to CO2 should create or increase the preference.
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
Pattern choices by bumblebees 6
Experiment 1 tested these two predictions. In addition, by comparing the preference of stimuli
occupied by another bee with the preference of stimuli occupied by non-organic objects (a coin,
as in Dawson and Chittka (2012), and a piece of Styrofoam), we could begin to address the
question of whether any preference for occupied stimuli was indeed social.
Aggressive tendencies ought to be directed at other individuals and not at flowers
themselves. Heightening an aggressive tendency ought not to increase a preference for a pattern
that is more “floral” than another. To determine whether the effects of our variables were
specific to social preferences, Experiment 2 examined their effects on preference of floral
patterns with no occupiers. Given that radial patterns (illustrated in the legend of Figure 3) are,
by and large, preferred over concentric patterns (Orbán & Plowright, 2013) as they are thought to
resemble flowers in nature (Lehrer, Horridge, Zhang & Gadagkar, 1995), we examined the effect
of food deprivation and CO2 exposure on relative choice of these patterns. Food deprivation
ought, if anything, to increase preference for radial patterns, while CO2 should not. Given that
larger bumblebees tend to invest themselves in foraging duties while smaller bees tend to the
nest (Goulson, 2010), though task specialization is not as marked as in honeybees, we reasoned
that the effect of food deprivation might interact with body mass.
Methods
Subjects
Three commercial colonies of Bombus impatiens Cresson in plastic nest boxes (19.5 cm
X 17.5 cm X 12 cm high) were supplied by Koppert Canada. The colonies were covered with
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
Pattern choices by bumblebees 7
an opaque lid and so they received little light, as in nature. In Experiment 1, 48 bees from each
of two colonies were used, for a total of 96. In Experiment 2, 60 bees from the third colony were
used. All bees were tested the first time they left the colony: they had no pre-experimental
experience outside of the nest box. In all conditions, to motivate the bees to exit the colony, the
wick that absorbed sugar solution from a plastic bag beneath the nest box was capped for one or
two days prior to testing.
Apparatus
The 12-corridor maze that we used, modelled on that of Lehrer et al. (1995), is
diagrammed in Figure 1. Photographs are shown in Plowright et al. (2011, Fig.1). It was
constructed of grey Plexiglas® with a clear cover. From the central area (22 cm wide, 15 cm
high), 12 corridors radiated outwards. The bees entered the maze via a screen tube that
connected their colony exit hole to an entrance hole in the center of the floor of the maze. The
exit was stoppered when the colony was not in use during the experiment: bees were not allowed
to travel to and from the apparatus prior to testing. The entrance to the maze was gated during
testing sessions: bees were let into the maze individually. The corridors were 6 cm wide at their
entrance from the center of the maze and 15 cm long. The back walls were 13 cm wide.
The maze was illuminated from above by high-frequency (> 40 KHz) lighting equipment
(three Sylvania Quicktronic T8 QHE4x32T8/112 light ballasts, each with four Sylvania model
FO32/841/XP/SS/EC03 fluorescent light bulbs).
Stimuli
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
Pattern choices by bumblebees 8
In both experiments, the stimuli were mounted with Velcro® on the back walls of the
corridors in the maze, as shown in Figure 1.
Experiment 1
Nine corridors in the maze were empty, i.e. “Unoccupied”. The remaining three corridors
enclosed a stimulus, so they were “Occupied”. In one of these three corridors, a dead pinned
queen B. impatiens was attached to the back wall (“Occupied-Conspecific”). Bumblebees are
attracted to artificial flowers where a model bee or a dead bee have been pinned in much the
same way they are attracted to artificial flowers where other bees are already foraging
(Leadbeater & Chittka, 2007), though there is no claim here that motion cues are irrelevant in
general (indeed, they are important in defensive behaviour of honeybees (Free, 1961) and in
social learning of foraging behaviour by bumblebees (Avarguès-Weber & Chittka, 2014 a&b)).
Accordingly, we used a dead pinned bee: a dead queen bee as in Plowright et al. (2013) to
maximize the chances that it would be detected by virtue of its larger size. In nature, the times at
which workers and queens forage within a season overlap and so workers could well encounter
other queens. In another corridor was a Canadian dime (“Occupied-Coin”) and in the third, a
piece of white Styrofoam® (“Occupied-Styrofoam”) of the same diameter as the coin (18 mm),
which was the colour of steel.
Experiment 2
Two opposing corridors were blocked off and the remaining ten contained circular (6 cm
diameter) stimuli: three radial patterns consisting of four black and four white wedges, and
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
Pattern choices by bumblebees 9
seven concentric patterns consisting of two black and two white circles. The patterns were
printed on paper in black and white and then laminated.
Design
Experiment 1
For each of the following four conditions, 12 bees from each of the two colonies were
used: (1) “Baseline”: the colony had ad libitum access to a ball of pollen paste placed beside the
cocoons (pollen purchased from an apiarist, crushed and mixed with honey and water), though
access to sugar solution was prevented for two days prior to testing so that the bees would be
motivated to exit the colony (2) “Pollen deprivation”: To increase the level of motivation to
forage over and above the level in baseline, not only was there deprivation of sugar solution, but
pollen deprivation was superimposed, with the pollen ball being removed for 24 h prior to testing
(3) “Strong CO2”: as with baseline, but with the addition of a steady stream into the maze of a
mixture of CO2 and O2 (1:3 by volume) for the duration of the time spent in the maze by each bee
and (4) “Pure CO2”: pure carbon dioxide gas (though the maze was not gas tight so there was
some dilution). In the last two conditions, the gas was released directly into the maze through an
aperture in the lid so as to ensure that individual bees were under the influence of CO2 as they
made their pattern choices. The rate of gas flow was 1.1 - 1.3 L/min. For the first colony used,
the conditions were run in order 1-4, and for the second colony, the reverse order was used, so
that any overall effect of experimental condition could not be attributed to colony aging.
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
Pattern choices by bumblebees 10
Three unoccupied corridors separated each of the three occupied corridors. The positions
of the three types of occupied corridors (conspecific, coin and Styrofoam) were counterbalanced
so that across 12 bees, they appeared equally often in each of the 12 corridors in the maze.
Experiment 2
For each of the following three conditions, 20 bees were used: (1) Baseline (2) Pure CO2
and (3) Pollen deprivation. These conditions, which were run in order, are described above for
Experiment 1. The positions of the patterns were counterbalanced so that across 20 bees in each
condition, the three radial patterns appeared six times and the seven concentric patterns appeared
fourteen times in each of the ten corridors used.
Procedure
The procedure was common to both experiments. Once a worker exited the colony and
entered the maze for the first time (where the first time was also the last time), its first 20 choices
were recorded as in Plowright et al. (2013). A choice was defined as making contact with the
back wall of the corridor or the stimulus attached to it. A new choice was only recorded once
the bee had crossed the threshold of the corridor and returned to the center of the maze (in other
words, ‘dithering’ within a corridor did not count as multiple choices). Once 20 choices were
made, the bee was removed from the apparatus and killed, after which the bees in Experiment 2
were weighted on a Mettler Toledo balance model PB 303-S. In the Pure CO2 condition, some
bees (n = 7 in Experiment 1 and n = 4 in Experiment 2) became inactive for over 15 min. with
exposure to CO2 and did not complete 20 choices—they were excluded from the data set.
Testing for each condition took place for 4-8 h per day, over the course of 1-3 days.
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
Pattern choices by bumblebees 11
The maze was cleaned with a damp cloth between subjects to minimize any visual
evidence of a corridor having been visited. Though bumblebees can learn whether chemical
“footprints” (i.e. scent-marks) predict reward or non-reward (Saleh & Chittka, 2006), in our case
all patterns were non-rewarding and so there was no possibility of cued discriminative learning.
Statistical analyses
The data were treated as binomial proportions. Two analyses were undertaken in
Experiment 1. First, the proportion of choices for the corridor occupied by the conspecific was
compared across the 4 conditions. A logistic model, which specifies a binomial error term, was
fit to the data using SPSS 21. Subsequently, each of the choice proportions was compared to a
theoretical value of chance (1/12) using the replicated goodness-of-fit test (Sokal & Rohlf,
2012). The replicated goodness-of-fit test yields two G values, which are compared to χ2 values
in the tests of significance: GP (P for Pooled) compares group choice frequency to that expected
by chance, and GH (H for Heterogeneity) tests for individual differences. In Experiment 2, a
logistic model was used, with bee weights entered as a covariate. The G-test was used to
determine whether the proportion of choices for the radial pattern differed from a theoretical
value of chance (3/10) for each of the 3 conditions.
Results
Experiment 1: Social preferences
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
Pattern choices by bumblebees 12
Comparisons across conditions
Between-colony differences in the overall preference for the Occupied-Conspecific
corridor were observed. Individuals from the second colony chose it 3.5 times out of 20 on
average, which was significantly higher than the mean of 2.5 for the first colony (χ2 = 8.98, df =
1, p = .003). No interaction between colony and experimental condition, however, was detected
(χ2 = 1.17, df = 3, p = .76), and so the data from both colonies were pooled. The effect of
condition was significant (χ2 = 32.57, df = 3, p <.0001). Given that behaviour in the baseline
condition was almost indistinguishable from chance (Fig. 2 a & b), the question as to whether
both the food restriction and the CO2 manipulations were effective in changing preference is
addressed below in a comparison of those conditions with chance.
Comparisons with theoretical value of chance
The choice proportion for the Occupied-Conspecific corridor that would be expected by
chance (Fig. 2a) was 8% (1/12). The group choice proportion in the Baseline condition was no
different from chance (GP = 1.66, 1 df, p = .20), though individual differences were significant
(GH = 37.90, df = 23, p = 0.03) with one outlier making 7 out of 20 choices of the Occupied-
Conspecific corridor while the rest made 0-4 such choices.
In the Pollen deprivation condition (Fig. 2c), the preference of 14% was significantly
higher than chance (Gp = 18.01, 1 df, p < .0001). The same was true of the Strong CO2
condition (Gp =14.59, 1df, p < .0001) shown in Figure 2d. The strongest effect was observed in
the Pure CO2 condition (Gp =94.34, 1 df, p < .0001): Figure 2e shows that the proportion of
choices for the Occupied-Conspecific corridor was approximately twice as high as chance and
there was a commensurate decrease in the proportion of choices for the Unoccupied corridors.
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
Pattern choices by bumblebees 13
The increase in preference for occupied corridors was specific to the corridor occupied by the
conspecific: little if any change can be seen in the proportion of choices for the corridors
occupied by the coin or the Styrofoam. In all three experimental conditions, there was
significant individual variation (GH ≥ 38.81, df = 23, p < .05). In the Pure CO2 condition, two
bees showed a preference for the Occupied-Conspecific corridors as high as 50:50.
Experiment 2: Preferences for floral patterns
Comparisons across conditions
A significant effect of condition was found on preference for the radial patterns, (χ2 =
7.93, df = 2, p = .019). Again, because the choice proportions in Baseline were at chance levels,
the question of which treatments were effective in modifying behaviour are addressed below in
comparisons between group choice proportions and chance levels. Body mass (Mean = .178 g,
SD = .03 g), which was entered in the logistic model as a covariate, was not an important
explanatory variable (χ2 = 2.20, df = 1, p = .14). The interaction between body mass and
condition was not significant (χ2 = .41, df = 2, p = .81).
Comparisons with theoretical value of chance
In the Baseline condition, the choice proportion for the radial patterns was not
significantly different from a chance value of 30% (GP = 0.295, df = 1, p = .59). Figure 3 shows
that a preference emerged in the condition of Pollen deprivation (GP = 13.14, df = 1, p = .0003):
the choice proportion rose to 38%. No such preference, however, was detected in the Pure CO2
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
Pattern choices by bumblebees 14
condition (GP = .05, df = 1, p = .82). No individual differences were detected in any of the three
conditions (GH ≤ 18.65, df = 19).
Discussion
In general, the results of our experiments show that pattern preferences by bumblebees
leaving their nest for the first time are context specific: they are tied to variables relevant to
different functions. As such this work fits well within an ecological approach to learning and
motivation (Shettleworth, 2010) whereby specific situations trigger appropriate goal-specific
sequences of unlearned and learned behaviour patterns. Preferences for unrewarding radial
patterns over concentric patterns, which have been reported for flower-naïve bumblebees
(Simonds & Plowright, 2004; Orbán & Plowright, 2013; Plowright, Simonds & Butler, 2006) are
thought to reflect preferences for floral shapes (Lehrer et al., 1995). Such preferences might help
to guide inexperienced bees toward their first floral contact. If so, the preferences should depend
on conditions of food deprivation. Here, we found no detectable preference when the bees had
unlimited access to pollen, but a preference was created when pollen was taken away from the
colony (Experiment 2). Administering CO2, however, led to no discernible change in
preference, though the manipulation of CO2 was anything but subtle: the concentration of CO2
could not have been higher. One caveat is that while there was ample room on the scale of
measurement for an increase in preference, there was less room for a decrease, though one might
have been envisaged given that exposure to CO2 in honeybees (A. mellifera) leads to a decrease
in pollen collection (Ribbands, 1950).
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
Pattern choices by bumblebees 15
In colonies of bumblebees, division of labour is related to size variation, with larger
workers carrying out a larger share of the foraging duties than smaller workers (Jandt &
Dornhaus, 2009) and being more adept these tasks (Spaethe & Weidenmuller, 2002). Current
research on size variation in bumblebees has considered both its proximal causes (Couvillon et
al., 2010) and functional significance (Peat, Tucker & Goulson, 2005; Couvillon & Dornhaus,
2010). Our argument that the preference for radial patterns is linked to food gathering would
have been bolstered had larger bees been more affected by the manipulation of food availability
than smaller bees, but no interaction of our experimental manipulation with body mass was
obtained—with self-selected groups of bees that exited the colony rather than stayed in, the
range of weights was presumably restricted.
If a preference for a pattern occupied by another forager also helps to guide bees toward
their first floral contact, then it too should depend on food deprivation. It did. Pollen
deprivation, however, was not the only variable that affected preference. CO2 also had an effect,
in both concentrations that we used (Experiment 1). Perhaps the preference for corridors
occupied by a conspecific is a general stress response (see Muth, Scampini & Leonard, this
issue) to threatening situations that include both lack of food and the presence of predators. The
function of the response to CO2, however, is less clear than the function of a response to food
deprivation. Here we suggested three possibilities: (1) The increase in attraction to an occupied
stimulus may be the beginnings of an aggressive response, similar to that of individuals within
colonies that hiss and buzz in response to the vibrations and sudden increase in CO2 levels
characteristic of disruption by a predator (Kirchner & Röschard, 1999). One untested
assumption underlying this interpretation is that the aggressive motivation triggered by the CO2
was not predator specific and would lead to an increased tendency to attack any organism in the
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
Pattern choices by bumblebees 16
vicinity. (2) Alternatively, it may be a response, in the context of danger, to what may be
perceived as a signal of safety: an unlearned response analogous to the learned response
documented by Dawson and Chittka (2014). (3) We can not rule out the possibility that it may
even be protective behaviour whereby one bee joins another to defend herself and her sister from
a common threat. What does seem clear is that the response is selective: it was specific to
corridors occupied by another insect, and not corridors occupied by a similarly sized coin or
piece of Styrofoam. The specific physical characteristics of the insect that triggered approach
behaviour remain to be determined. Perhaps dark colour is important, since dark cotton balls
elicit more stinging in honeybees than white ones (Free, 1961). Indeed, black ping pong balls
are more likely to be attacked by bumblebees (Bombus sonorus Say) than white ping pong balls
(Visscher & Vetter, 1995). Rough texture may also be important (Free, 1961).
The mechanisms by which our variables exerted their effects on choice remain to be
determined. Pure CO2 is an anesthetic, and indeed a few bees in the pure CO2 condition in
Experiment 2 did gradually became immobile and failed to complete 20 choices, perhaps just as
much because of the decrease in O2 as the presence of CO2. The effects in our study, however,
were not simply decreases in motor activity, since choice of any corridor required the same
approach toward the back wall and we obtained groups of equal size all completing 20 choices.
The effects may be the product of increased attention to the relevant stimuli. They may reflect
memory and decision making—indeed CO2 produces retrograde amnesia in honeybees
(Beckmann, 1974) and individual decision-making in honeybees is influenced by variables, such
as being shaken, that mimic predation (Bateson, Desire, Gartside & Wright, 2011).
Having established both the effect of food deprivation on preference for radial stimuli,
and the effects of CO2 exposure and food deprivation on preference for patterns occupied by
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
Pattern choices by bumblebees 17
other bees, future research should harness these variables in more ecologically realistic ways. In
the future, gradations of pollen availability could be used instead of all-or-nothing. In addition,
colonies at various stages of development could be used: the effect of pollen restriction is almost
certainly specific to young colonies during periods of rapid colony growth when protein is in
high demand for the feeding of larvae (Free, 1955). The CO2 concentration could also be
brought down to levels that would be encountered in nature. We used concentrations of 25%
CO2 as well as 100%, which vastly exceed the concentration in the colony [0.2 - 6% in honeybee
hives (Nicolas & Sillans, 1989)], in the atmosphere [385 ppm in 2008 (Hansen et al., 2008)--up
from 350 ppm in 1989 (Nicolas & Sillans, 1989)] or in mammalian breath [3.5% for humans
(Kirchner & Röschard, 1999)]. In our study, the workers were also exposed to CO2 individually
in the maze whereas in nature, the influx of CO2 would not likely be experienced at the foraging
site but rather at the site of the colony. In future research, the two locations could be compared,
though prolonged CO2 exposure in the colony can lead to aberrant behaviours such as expelling
larvae (Pomeroy & Plowright, 1979) and also can affect the reproductive behaviour of the queen
(Mackensen, 1947).
Following the lead of Gould (1990), today bees are used as a model for understanding
cognition (Menzel, 2012; Menzel & Brembs, 2007). Contemporary research topics in bee-
cognition include visual categorization (Benard, Stach & Giurfa, 2006; Dyer, 2012), timing
(Boisvert & Sherry, 2006; Craig et al., 2014), spatial cognition (Brown & Demas, 1994; Cheng,
2000), numerical cognition (Pahl, Si & Zhang, 2013), concept learning (Brown & Sayde, 2013;
Giurfa et al., 2001; Gould 2002), picture-object correspondence (Thompson & Plowright, 2014)
and problem solving (Mirwan & Kevan, 2014). Particularly relevant here is research on social
learning and cognition (reviewed by Sherry & Strang, 2014) which have been neglected until
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
Pattern choices by bumblebees 18
recently (Dukas, 2008). Much of this cognition, however, is built upon some “pre-functional”
behaviour (Hogan, 1994): the building blocks of learning (Tinbergen, 1951). The behaviour that
occurs prior to experience with the goals it is designed to attain remains poorly understood for
bees. Prior research has documented some of the variables regarding the visual appearance of
flowers that influence this pre-functional behaviour (reviewed by Orbán & Plowright, 2014), and
this paper documents some of the variables that affect the internal state of the bees. Now that
the ‘cognitive revolution’ has become mainstream in bee behaviour, there may be something to
be gained by stepping back to consider the details of how bees do what they do when they are
met with the world outside their colony for the first time.
366
367
368
369
370
371
372
373
374
375
Pattern choices by bumblebees 19
References
Avarguès-Weber, A., & Chittka, L. (2014a). Observational conditioning in flower choice
copying by bumblebees (Bombus terrestris): Influence of observer distance and
demonstrator movement. PLOS ONE, 9: e88415.
Avarguès-Weber, A., & Chittka, L. (2014b). Local enhancement or stimulus enhancement?
Bumblebee social learning results in a specific pattern of flower preference. Animal
Behaviour 97, 185-191.
Bateson, M., Desire, S., Gartside, S.E., & Wright, G.A. (2011). Agitated honeybees exhibit
pessimistic cognitive biases. Current Biology, 21, 1070-1073.
Baude M., Danchin É., Mugabo M. & Dajoz I. (2011). Conspecifics as informers and
competitors: An experimental study in foraging bumble-bees. Proceedings of the Royal
Society B, 278, 2806-2813.
Beckmann, H. E. (1974). The damaging effect of supercooling, narcosis, and stress on memory
of the honey bee. Journal of Comparative Physiology, 94, 249-266.
Benard, J., Stach, S., & Giurfa, M. (2006). Categorization of visual stimuli in the honeybee Apis
mellifera. Animal Cognition, 9, 257-270.
Boisvert, M. J., & Sherry, D. F. (2006). Interval timing by an invertebrate, the bumble bee
Bombus impatiens. Current Biology, 16, 1636-1640.
Brown, M. F., & Demas, G. E. (1994). Evidence for spatial working memory in honeybees Apis
mellifera. Journal of Comparative Psychology, 108, 344-352.
Brown, M. F., & Sayde, J. M. (2013). Same/different discrimination by bumblebee colonies.
Animal Cognition, 16, 117-125.
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
Pattern choices by bumblebees 20
Cheng, K. (2000). How honeybees find a place: Lessons from a simple mind. Animal Learning
& Behavior, 28, 1-15.
Couvillon, M. J., & Dornhaus, A. (2010). Small worker bumble bees (Bombus impatiens) are
hardier against starvation than their larger sisters. Insectes Sociaux, 57, 193-197.
Couvillon, M. J., Jandt, J. M., Duong, N. H. I., & Dornhaus, A. (2010). Ontogeny of worker
body size distribution in bumble bee (Bombus impatiens) colonies. Ecological
Entomology, 35, 424-435.
Craig, D. P. A., Varnon, C. A., Sokolowski, M. B., Wells, H., & Abramson, C. I. (2014). An
assessment of fixed interval timing in free-flying honey bees (Apis mellifera ligustica):
An analysis of individual performance. PLOS ONE, 9, e101262.
Dawson, E. H., Avarguès-Weber, A., Chittka, L., & Leadbeater, E. (2013). Learning by
observation emerges from simple associations in an insect model. Current Biology, 23,
727-730.
Dawson, E. H., & Chittka, L. (2012). Conspecific and heterospecific information use in
bumblebees. PLOS ONE, 7, e31444.
Dawson, E. H., & Chittka, L. (2014). Bumblebees (Bombus terrestris) use social information as
an indicator of safety in dangerous environments. Proceedings of the Royal Society B,
281, 1-5.
Dukas, R. 2008. Evolutionary biology of insect learning. Annual Review of Entomology, 53,
145-160
Dyer, A. G. (2012). The mysterious cognitive abilities of bees: Why models of visual processing
need to consider experience and individual differences in animal performance. Journal
of Experimental Biology, 215, 387-395.
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
Pattern choices by bumblebees 21
Free, J.B. (1955). The division of labour within bumblebee colonies. Insectes Sociaux, 2, 195-
212.
Free, J. B. (1961). The stimuli releasing the stinging response of honeybees. .Animal
Behaviour, 9, 193-196.
Giurfa, M., Zhang, S., Jenett , A., Menzel, R., & Srinivasan, M. V. (2001). The concepts of
‘sameness’ and ‘difference’ in an insect. Nature, 410, 930-933.
Gould, J. L. (1990). Honey bee cognition. Cognition, 37, 83-103.
Gould, J. L. (2002). Can honey bees create cognitive maps? In Bekoff, M., Allen, C. &
Burghardt, G. M. (Eds). The cognitive animal (pp. 41-45). Cambridge, Massachussets:
MIT Press.
Goulson D. (2010). Bumblebees: Behaviour, ecology, and conservation. (2nd ed). New York:
Oxford University Press. 317 pp
Grüter, C., & Leadbeater, E. (2014). Insights from insects about adaptive social information use.
Trends in Ecology & Evolution, 29, 177-184.
Hansen, J., Sato, M., Kharecha, P., Beerling, D., Berner, R., Masson-Delmotte, V., Pagani, M., .
Raymo, M., Royer, D. L., & Zachos, J. C. (2008). Target atmospheric CO2: Where
should humanity aim? Open Atmospheric Science Journal, 2, 217-231.
Hogan, J. A. (1994). Structure and development of behavior systems. Psychonomic Bulletin and
Review, 1, 439-450.
Jandt, J. M., & Dornhaus, A. (2009). Spatial organization and division of labour in the
bumblebee Bombus impatiens. Animal Behaviour, 77, 641-651.
Kawaguchi, L. G., Ohashi, K., & Toquenaga, Y. (2006). Do bumble bees save time when
choosing novel flowers by following conspecifics? Functional Ecology, 20, 239-244.
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
Pattern choices by bumblebees 22
Kikuchi, T. (1963). Studies on the coaction among insects visiting flowers. III Dominance
relationship among flower-visiting flies, bees and butterflies. Science Reports of Tohoku
University - Fourth Series: Biology, 29, 1-8.
Kirchner, W. H., & Röschard, J. (1999). Hissing in bumblebees: an interspecies defence signal.
Insectes Sociaux, 46, 239-243.
Leadbeater, E., & Chittka, L. (2007). The dynamics of social learning in an insect model, the
bumblebee (Bombus terrestris). Behavioral Ecology and Sociobiology, 61, 1789-1796.
Leadbeater, E., & Chittka, L. (2009). Bumble-bees learn the value of social cues through
experience. Biology Letters, 5, 310-312.
Lehrer, M., Horridge, G. A., Zhang, S. W. & Gadagkar. R (1995). Shape vision in bees: Innate
preference for flower-like patterns. Philosophical Transactions of the Royal Society B,
347, 123-137.
Mackensen, O. (1947). Effect of carbon dioxide on initial oviposition of artificially inseminated
and virgin queen bees. Journal of Economic Entomology, 40, 344-349.
Menzel, R. (2012). The honeybee as a model for understanding the basis of cognition. Nature
Reviews Neuroscience, 13, 758-768.
Menzel, R., & Brembs, B. (2007). Cognition in invertebrates. In J. H. Kaas (Ed.) Evolution of
nervous systems, Vol. II: Evolution of nervous systems in invertebrates (pp. 403–422).
Oxford: Academic Press.
Mirwan, H. B., & Kevan, P. G. (2014). Problem solving by worker bumblebees Bombus
impatiens (Hymenoptera: Apoidea). Animal Cognition, doi: 10.1007/s10071-014-0737-0
Muth, F., Scampini, A. V., & Leonard, A. S. (this issue.). The effects of acute stress on learning
and memory in bumble bees.
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
Pattern choices by bumblebees 23
Nicolas, G., & Sillans, D. (1989). Immediate and latent effects of carbon dioxide on insects.
Annual Review of Entomology, 34, 97-116.
Orbán, L. L., & Plowright, C. M. S. (2013). The effect of flower-like and non-flower-like visual
properties on choice of unrewarding patterns by bumblebees. Naturwissenschaften, 100,
621-631.
Orbán, L. L., & Plowright, C. M. S. (2014). Getting to the start line: how bumblebees and
honeybees are visually guided towards their first floral contact. Insectes Sociaux doi:
10.1007/s00040-014-0366-2
Pahl, M., Si, A., & Zhang, S. (2013). Numerical cognition in bees and other insects. Frontiers
in Psychology, 4, 162.
Peat, J., Tucker, J., & Goulson, D. (2005). Does intraspecific size variation in bumblebees allow
colonies to efficiently exploit different flowers? Ecological Entomology, 30, 176-181.
Plowright, C. M. S., Evans, S. A., Chew Leung, J., & Collin, C. A. (2011). The preference for
symmetry in flower naïve and not-so-naïve bumblebees. Learning and Motivation, 42,
76-83.
Plowright, C. M. S., Ferguson, K. A., Jellen, S. L., Xu, V., Service, E. W., & Dookie, A. L.
(2013). Bumblebees occupy: When foragers do and do not use the presence of others to
first find food. Insectes Sociaux, 60, 517-524.
Plowright, C. M. S., Simonds, V. M. & Butler, M. A. (2006). How bumblebees first
find flowers: Habituation of visual pattern preferences, spontaneous recovery
and dishabituation. Learning and Motivation, 37, 66-78.
Pomeroy, N., & Plowright, R. C. (1979). Larval ejection following narcosis of bumble bees
(Hymenoptera: Apidae). Journal of the Kansas Entomological Society, 52, 215-217.
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
Pattern choices by bumblebees 24
Ribbands, C. R. (1950). Changes in the behaviour of honey-bees following their recovery from
anaesthesia. Journal of Experimental Biology, 27, 302-310.
Saleh, N., & Chittka, L. (2006). The importance of experience in the interpretation of conspecific
chemical signals. Behavioral Ecology and Sociobiology, 61, 215-220.
Sherry, D. F., & Strang, C. G. (2014). Contrasting styles in cognition and behaviour in
bumblebees and honeybees. Behavioural processes,
http://dx.doi.org/10.1016/j.beproc.2014.09.005
Shettleworth, S.J. (2010). Cognition, evolution, and behavior. (2nd ed.) New York: Oxford
University Press. 720 pp.
Simonds V. & Plowright C. M. S. (2004). How do bumblebees first find flowers? Unlearned
approach responses and habituation. Animal Behaviour, 67, 379-386
Sokal, R. R., & Rohlf, J .F. (2012). Biometry. (4th ed.). New York: W.H. Freeman.
Spaethe, J., & Weidenmüller, A. (2002). Size variation and foraging rate in bumblebees (Bombus
terrestris). Insectes Sociaux, 49, 142-146.
Tepedino, V. J., & Parker, F. D. (1981). The effect of floral occupancy on the foraging of
flower-visiting insects. Psyche, 88, 321-330.
Thompson, E. L., & Plowright, C. M. S. (2014). How images may or may not represent flowers:
Picture-object correspondence in bumblebees (Bombus impatiens)? Animal Cognition.
doi: 10.1007/s10071-014-0733-4
Tinbergen, N. (1951). The study of instinct. London: Oxford University Press.
Visscher, K. P., & Vetter, R. S. (1995). Smoke and target color effects on defensive behavior in
yellow jacket wasps and bumble bees (Hymenoptera: Vespidae, Apidae) with a
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
Pattern choices by bumblebees 25
description of an electronic attack monitor. Journal of Economic Entomology, 88, 579-
583.
512
513
Pattern choices by bumblebees 26
Acknowledgments
This research was supported by a research grant to C.M.S.P. from the Natural Sciences
and Engineering Research Council of Canada. We thank Daniel Dostie for his help with the
graphics and Levente Orbán for his assistance with the videos. Vicki Xu and Emma Thompson
provided constructive comments on the manuscript. We also thank Koppert Canada for their
generosity in donating the bumblebee colonies for our research.
514
515
516
517
518
519
520
Pattern choices by bumblebees 27
Supplementary materials
Video 1. Film clip of one bumblebee landing on an unrewarding artificial flower that is already
occupied by another bee. The grappling of the two bees together is characteristic of aggressive
interactions.
Video 2. Film clip of one bumblebee landing on an unrewarding artificial flower that is already
occupied by another bee. The raising of the leg by the bee that was on the flower first is
characteristic of defensive behaviour.
521
522
523
524
525
526
527
528
529
530
531
532
533
Pattern choices by bumblebees 28
Figure captions
Fig. 1. Radial maze and connected hive box. Patterns were mounted vertically on the walls of
the corridors in the positions marked with a red circle. Reproduced from Plowright et al. (2013),
with kind permission from Springer Science and Business Media.
Fig. 2. Choice proportions, pooled for two colonies in Experiment 1. There were 9 Unoccupied
corridors and 3 Occupied (with Conspecific, Styrofoam or Coin). In each condition, 12 bees per
colony were given 20 choices each. (a) Theoretical proportions of chance (b-e) Observed
proportions under four conditions (baseline, pollen deprivation, Strong CO2 and Pure CO2.
* The choice proportion for the Occupied-Conspecific corridor was significantly greater than a
chance value of 8%.
Fig. 3. Mean choice frequencies of radial and concentric patterns in Experiment 2 with standard
error bars. In each condition, 20 bees were given 20 choices in a maze where 3 corridors
contained a radial pattern and 7 contained a concentric pattern.
* Choice proportion significantly different from a chance value of 30%.
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
Pattern choices by bumblebees 29
551
552
Pattern choices by bumblebees 30
553
554
555
556
Pattern choices by bumblebees 31
557
Recommended