Independence and interdependence in collective decision making: an agent-based model of nest-site choice by honeybee swarms

Condorcet''s jury theorem shows that when the members of a group have noisy but independent information about what is best for the group as a whole, majority decisions tend to outperform dictatorial ones. When voting is supplemented by communication, however, the resulting interdependencies between decision makers can strengthen or undermine this effect: they can facilitate information pooling, but also amplify errors. We consider an intriguing non-human case of independent information pooling combined with communication: the case of nest-site choice by honeybee (Apis mellifera) swarms. It is empirically well documented that when there are different nest sites that vary in quality, the bees usually choose the best one. We develop a new agent-based model of the bees'' decision process and show that its remarkable reliability stems from a particular interplay of independence and interdependence between the bees.     

Collective Behaviour without Collective Order in Wild Swarms of Midges

Collective behaviour is a widespread phenomenon in biology, cutting through a huge span of scales, from cell colonies up to bird flocks and fish schools. The most prominent trait of collective behaviour is the emergence of global order: individuals synchronize their states, giving the stunning impression that the group behaves as one. In many biological systems, though, it is unclear whether global order is present. A paradigmatic case is that of insect swarms, whose erratic movements seem to suggest that group formation is a mere epiphenomenon of the independent interaction of each individual with an external landmark. In these cases, whether or not the group behaves truly collectively is debated. Here, we experimentally study swarms of midges in the field and measure how much the change of direction of one midge affects that of other individuals. We discover that, despite the lack of collective order, swarms display very strong correlations, totally incompatible with models of non-interacting particles. We find that correlation increases sharply with the swarm''s density, indicating that the interaction between midges is based on a metric perception mechanism. By means of numerical simulations we demonstrate that such growing correlation is typical of a system close to an ordering transition. Our findings suggest that correlation, rather than order, is the true hallmark of collective behaviour in biological systems.     

Stop-Signaling Reduces Split Decisions without Impairing Accuracy in the Honeybee Nest-Site Selection Process

In the honeybee swarm nest-site selection process, individual bees gather information about available candidate sites and communicate the information to other bees. The swarm makes an agreement for a candidate site when the number of bees that supports the site reaches a threshold. This threshold is usually referred to as the quorum threshold and it is shown by many studies as a key parameter that is a compromise between the accuracy and speed of decisions. In the present work, we use a model of the honeybee Apis mellifera nest-site selection process to study how the quorum threshold and discovery time of candidate sites have major impact on two unfavorable situations in selecting a nest site: decision deadlock and decision split. We show that cross-inhibitory stop-signaling, delivered among bees supporting different sites, enables swarms to avoid the decision split problem in addition to avoiding the decision deadlock problem that has been previously proposed. We also show that stop-signaling improves decision speed, but compromises decision accuracy in swarms using high quorum thresholds by causing the swarms to be trapped in local optima (e.g., choosing a sub-optimal option that is encountered first). On the other hand, we demonstrate that stop-signaling can reduce split decisions without compromising decision accuracy in swarms using low quorum thresholds when it is compared to the accuracy of swarms using the same threshold values but not exhibiting stop-signaling. Based on our simulations, we suggest that swarms using low quorum thresholds (as well as swarms with large population sizes) would benefit more from exhibiting the stop-signaling activity than not exhibiting it.     

Dynamical aspects of animal grouping: Swarms,schools, flocks,and herds

Grouping of animals is a natural phenomenon in which a number of animal individuals are involved in movement as forming a group. Examples are insect swarms and fish schools. In this article an attempt is made to describe the motion of grouping individuals kinematically as distinct from simple diffusion or random walk, to model the grouping on the basis of dynamics of animal motion, and to interpret the grouping from the standpoint of advection-diffusion processes. Also presented is dynamical modeling for the group size distribution as a result of amalgamation and splitting processes of groups.Examples of animal grouping are described in detail. They are insect swarms, zooplankton swarms, fish schools, bird flocks, and mammal herds. The presented mathematical models are compared with data of these animal groupings.     

Behavioural phase polyphenism in the Australian plague locust (Chortoicetes terminifera)

Swarming and the expression of phase polyphenism are defining characteristics of locust species. Increases in local population density mediate morphological, physiological and behavioural changes within individuals, which correlate with mass marching of juveniles in migratory bands and flying swarms of adults. The Australian plague locust (Chortoicetes terminifera) regularly forms migratory bands and swarms, but is claimed not to express phase polyphenism and has accordingly been used to argue against a central role for phase change in locust swarming. We demonstrate that juvenile C. terminifera express extreme density-dependent behavioural phase polyphenism. Isolated-reared juveniles are sedentary and repelled by conspecifics, whereas crowd-reared individuals are highly active and are attracted to conspecifics. In contrast to other major locust species, however, behavioural phase change does not accumulate across generations, but shifts completely within an individual''s lifetime in response to a change in population density.     

Birds and army ants in a fragment of the Atlantic Forest of Brazil

ABSTRACT Little is known about the birds associated with army‐ant swarms in the Brazilian Atlantic forest. Our objectives were to locate and monitor army‐ant swarms in the Atlantic rainforest of Brazil and to identify the species of birds that attended the swarms and exhibited bivouac‐checking behavior. From July 2004 to August 2005, we located 49 swarms of army ants, including 28 Eciton burchelli, 19 Labidus praedator, and 2 Eciton vagans swarms. No birds were present at 17 (35%) swarms. At 32 swarms where birds were present, 22 (69%) were E. burchelli swarms and 10 (31%) were L. praedator swarms. No birds were observed at the two E. vagans swarms. We identified 66 species of birds attending the swarms, but only 43 species were observed foraging on prey flushed by the ants. Eighteen of these species had not been previously reported to forage in association with army‐ant swarms. Most birds observed during our study attended army‐ant swarms opportunistically, with White‐shouldered Fire‐eyes (Pyriglena leucoptera) the only obligate ant follower. Our observations suggest that the arthropods and other organisms flushed by army ants represent an important food resource for several species of birds in the Atlantic forest ecosystem.     

Pelagic swarms and beach strandings of the squat lobster Munida gregaria (Anomura: Munididae) in the Beagle Channel, Tierra del Fuego

The present article is the first formal record of pelagic swarms and beach strandings of the squat lobster Munida gregaria in the Beagle Channel, southern South America. To describe size composition and natural diet of beach strandings and swarms, samples from strandings were taken in April 2007, March 2008 and March 2010. Samples from swarms were taken in May 2007 and April 2010. Also, during November 2008 and January 2009, two acoustic surveys were carried out to describe both, horizontal distribution and school metrics of swarms. Pelagic swarms and beach strandings were composed of 100% individuals of the morph gregaria of M. gregaria. Mean size of males and females in both, beach strandings and swarms, were similar. The diet of pelagic M. gregaria was composed mainly of crustaceans, unicellular algae and small macroalgae, sediment and particulate organic matter. Swarms of pelagic Munida took the shape of extended layers of varying density and height. Their positions in the water column were also variable: swarms were found at different depths, from the subsurface layer to near the sea bottom. Occasionally, some large swarms occupied most of the water column. In the Beagle Channel, SW winds during spring tides could be a forcing factor for M. gregaria strandings. The absence of shoaling animals in the Beagle Channel between 1997 and 2002 and the recent occurrence of swarms are coincident with their appearance in other locations in Patagonia. In consequence, we hypothesize that the factor influencing the formation of swarms must be acting at a regional scale.     

Hierarchy of Attractants for Honey Bee Swarms

Chemical signals influence the selection of potential nest cavities by honey bee reproductive swarms. Attractants for swarms include the odors of old dark honey bee brood combs, odors from noncomb hive materials and propolis, and Nasonov pheromone, the odor released from the Nasonov glands of worker bees. Based on crossover and choice test experiments, swarms were shown to prefer, among otherwise identical cavities, those cavities containing Nasonov pheromone over cavities with only comb or other hive odors, cavities containing old comb over those with only noncomb odors or propolis, and cavities containing noncomb odors or propolis over those without bee or hive odor. Synergy between odors was not observed; that is, comb and/or noncomb hive odors did not enhance the attractiveness of Nasonov pheromone. The data support a model based on a hierarchy of olfactory attractants used by honey bee swarms, in order of highest to lowest: Nasonov pheromone, comb odor, noncomb and propolis odors, and, finally, absence of bee- or hive-produced odor.     

The impact of Lake Victoria's lakefly abundance on Palearctic passerines

Wanink, J.H. & Goudswaard, K. 2000. The impact of Lake Victoria's lakefly abundance on Palearctic passerines. Ostrich 71 (1 & 2): 194–197.

In spite of an increase in lakeflies emerging from Lake Victoria, these periodically swarming insects remained an erratic food source for birds. However, even the relatively poor south-eastern shores were exploited by some Palearctic warblers on spring passage. Numbers and weights of Willow and Garden Warbler were correlated with lakefly abundance. The occurrence of lakefly swarms may trigger the birds' departure to the breeding areas, as the superabundance of food allows for rapid premigratory fattening.     

A review of the reproductive behavior of Polyphemus pediculus (L.) Müller (Crustacea: Branchiopoda)

The reproductive behavior of Polyphemus pediculus (L.) includes mating, laying of resting eggs and giving birth to juveniles. All these forms of behavior are performed in a population daily and simultaneously by means of individual and social actions. Individual behavior is manifested by certain sets of stereotyped movements which can be observed at any time of the day and at any location in a water body. Individual behavior has an episodic, stable character and low intensity. Social behavior is accomplished by specimens of similar physiological state who form transitory swarms. They form within the main continually existing main swarms of the population. Social behavior is rhythmic and of high intensity. Reproductive parthenogenetic swarms first form around dawn and later during the day: about noon in May–July and before sunset in Autumn. As swarms of parthenogenetic females disintegrate, swarms of old gamogenetic females form and mass laying of resting eggs begins. Young gamogenetic females and males swarm only when resting-egg-laying swarms disintegrate. Parthenogenetic reproduction peaks in northern Russia in May and August and persists until late September. Swarms of resting-egg-laying females form once a day around dawn, from May to September. Mating swarms form in summer in midmorning, at twilight and on moonlit nights. In Autumn, there is only one peak of mating, in midmorning. Intensity, duration and time of manifestation of different forms of social behavior change from spring to autumn. Location and character of their manifestation are predictable.     

<Emphasis Type="Italic">Hilara</Emphasis> sp. (Diptera: Empididae; Empidinae): Mating System,Swarm Movements,and Inbreeding Avoidance

In a study spanning parts of nine years, an undescribed species of Hilara Meigen was observed to form mating swarms displaying complex behaviors. Typically, swarms were shaped like a flattened torus rotating rapidly about a horizontal axis. Many swarms also moved up and down and turned slowly back-and-forth about a vertical axis. Both up-and-down and turning movements were random in extent and direction, suggesting that they might arise as random, asymmetric density fluctuations within the swarms themselves. A rotating secondary swarm appeared intermittently inside one end of some primary swarms. Swarm membership changed continually as flies left one swarm to join another and as entire swarms coalesced. At one site the set of all swarms displayed properties not found in the swarms individually: spatial extension, daily dissipation and reconstitution over a period of weeks or months, reproductive potential, and gene flow. Such emergent properties qualify the set as a multicomponent swarm, an object heretofore known only in computer models. Hilara sp. appears to be protandrous, univoltine, and promiscuous. Generally, males paired preferentially with somewhat smaller females, but some small and medium-sized males paired with much larger females. Although males of nearly all known Hilara species present nuptial gifts of prey or other items to females, nuptial gifts were not observed at any time during the present study. Many characteristics of swarms of Hilara sp. can be understood as adaptations that reduce inbreeding.     

The swarming behaviour of the Scottish biting midge, Culicoides impunctatus (Diptera: Ceratopogonidae)

Abstract.

• 1 Swarms of Culicoides impunctatus males were examined in the field in Scotland. Most swarms were close to midge emergence/breeding grounds over a variety of vegetation, some of which clearly acted as swarm markers. Low light (0–1000 lux) and still/humid conditions favoured swarming.
• 2 Swarm size ranged between less than 10 and more than one thousand midges. The modal size was 200 individuals. The smallest swarms were usually columnar in shape and the larger swarms ovoid.
• 3 Midges behaved individually within swarms, moving in an elliptical manner characteristic of other dipterans in swarms.
• 4 Swarms were classified as either ‘persistent’ or ‘transient’ in terms of their shape, size and stability. Wind was the most influential factor in disrupting swarms.
• 5 Swarms were confirmed as mating sites for C.impunctatus.

     

Non-random Mating in the Dance Fly Empis borealis: The Importance of Male Choice

In the dance-fly Empis borealis (Diptera, Empididae), females form swarms to which males, carrying a nuptial gift, come for mating. We examined whether males or females were choosy and/or competed for mates. First, measurements of the size relationships between copulating males and females, nuptial gifts and the swarming females from different swarms were assessed. Second, male visiting time in differently sized female swarms was recorded. Larger (wing-length) females participated disproportionately in copulations in each swarm, but not for the population at large. Female mating status (virgin/non-virgin) or proximity to oviposition (egg size) did not influence the likelihood of copulation. No assortative mating pattern was found: male size and size of nuptial gift did not correlate with size of the mating female. The time spent by males in swarms increased with the number of females present and it took longer when males left a swarm without copulation than when doing so. Male visiting time per female was negatively correlated with number of females in swarms. Males more often left smaller than larger swarms without mating. We conclude that E. borealis males discriminate among females but find no evidence for male competition or for female choice. It is still a question to what degree females compete for males.     

A predator's costs of overcoming the confusion-effect of swarming prey

Swarms of prey are known to have a confusion effect on predators that increases with swarm density. Hungry sticklebacks (Gasterosteus aculeatus) overcome the confusion (Heller & Milinski 1979) except when frightened by a predator of their own (Milinski & Heller 1978). This study investigates whether the costs of overcoming the confusion consist of decreased attention for predators. Hungry sticklebacks detected a predator of their own less often when they fed upon a prey swarm of high density than when they attacked a low density swarm. They also overlooked the predator more frequently when they fed with a high rate than with a low one in the same high density. Thus, there is a cost of attacking high density swarms. These results confirm two necessary conditions of a model for balancing feeding rate and predation risk (Milinski & Heller 1978). The decreased attention is assumed to be due to the limited channel capacity of the nervous system in processing sensory information.     

Collision avoidance during group evasive manoeuvres: a comparison of real versus simulated swarms with manipulated vision and surface wave detectors

Coordinated group motion has been studied extensively both in real systems (flocks, swarms and schools) and in simulations (self-propelled particle (SPP) models using attraction and repulsion rules). Rarely are attraction and repulsion rules manipulated, and the resulting emergent behaviours of real and simulation systems are compared. We compare swarms of sensory-deprived whirligig beetles with matching simulation models. Whirligigs live at the water''s surface and coordinate their grouping using their eyes and antennae. We filmed groups of beetles in which antennae or eyes had been unilaterally obstructed and measured individual and group behaviours. We then developed and compared eight SPP simulation models. Eye-less beetles formed larger diameter resting groups than antenna-less or control groups. Antenna-less groups collided more often with each other during evasive group movements than did eye-less or control groups. Simulations of antenna-less individuals produced no difference from a control (or a slight decrease) in group diameter. Simulations of eye-less individuals produced an increase in group diameter. Our study is important in (i) differentiating between group attraction and repulsion rules, (ii) directly comparing emergent properties of real and simulated groups, and (iii) exploring a new sensory modality (surface wave detection) to coordinate group movement.     

Why do males of the dance flyEmpis borealis refuse to mate? The importance of female age and size

Empis borealisfemales form swarms, and males carrying a nuptial gift come to swarms to mate. Males either mated with one of the females (accepted swarms) or left swarms without mating (refused swarms). Males mated with the younger (low wing-wear) and relatively larger females in accepted swarms. They seemed to be able to judge the relative size of the females but to ignore their absolute size. Visiting males stayed shorter in accepted swarms as female size variation increased. This probably reflects their greater ease in choosing a mate among females of relatively different sizes. Females in accepted swarms tended to be larger and to have less worn wings than females in rejected swarms.     

The role of photoreception in the swarming behavior of the copepod Dioithona oculata

The copepod Dioithona oculata forms dense swarms near mangrove prop roots that are centered around shafts of light penetrating the mangrove canopy. Swarms can be created in the laboratory within light shafts created with a fiber optic light pipe. Laboratory observations of swarming behavior were recorded using video cameras, and the swimming behavior of the copepods and density of the swarms were quantified using video‐computer motion and image analysis techniques. Swarm formation results from a combination of phototactic and klino‐kinetic behavior. Dark adapted copepods initially exhibit a photophobic response to a light shaft, but become positively phototactic within 3–5 min after exposure to the light. Copepod aggregation rates under the light fit a saturation model, suggesting that copepods are attracted independently to the swarm marker. Copepods reverse their swimming direction when they encounter light intensity gradients near the edge of a light shaft, which aids in maintaining the swarm. Swarm formation can occur in the laboratory at light intensities as slow as 0.1 μM photons m^‐2 s^‐1, which is similar to light intensities at dawn when they are first observed to form in nature. Swarm formation appears to have an endogenous rhythm, as copepods will not form swarms at night under a light shaft.     

Swarming Behavior, Sexual Dimorphism, and Female Reproductive Status in the Sex Role-Reversed Dance Fly Species Rhamphomyia marginata

Dance flies are predaceous insects which often form male mating swarms. In many species males prior to swarming catch an insect prey, which is presented to the female at mating. In Rhamphomyia marginata, females in contrast to males gather to swarm, while males carrying a prey visit swarms for mating. Here I describe the swarming and courtship behavior in R. marginata and provide data on sexual dimorphism and swarming female reproductive status. Females swarm in small clearings in the forests. There was no specific swarm-maker. The swarming period lasted for 2–3 h and peaked around sunset. Identical swarm sites were used each evening and for several years. The mean number of females in swarms (swarm sites with at least one female) was 9.9 ± 9.1 (range, 1–40; n = 107) in 1993 and 7.1 ± 7.0 (range, 1–35; n = 68) in 1994. No obvious competition between females in swarms was observed. The operational sex ratio in swarms was extremely female biased (all swarms, 0.04). Less than one-third of male visits to swarms resulted in mating and males were found more often in larger swarms. Nuptial prey consisted of male midges. Females seem to mate more than once. Swarming females had undeveloped eggs, whereas mated females in swarms had further developed eggs than unmated females. Amount of sperm in the spermatheca was correlated with egg size. Amount of sperm and egg size did not correlate with wet weight, wing length, or wing load, except for egg size and weight. The wing coloration pattern and shape in R. marginata females are unique among dance flies, being greatly enlarged (1.6 times larger than that of males) and bicolored (gray part, 60% of wing area). When females, instead of males, possess extravagant secondary sexual characters, it is predicted from sexual selection theory that females should compete for males and that males should be selective in their choice of partner. A sex-role reversal will evolve when assess to males limit female reproductive success. The dance fly species R. marginata, like Empis borealis, another dance fly species studied earlier and discussed here, seems to fit these predictions.     

Population structure and swarm formation of the cyclopoid copepod Dioithona oculata near mangrove cays

The cyclopoid copepod Dioithona oculata forms swarms in water>30 on deep among prop roots of red mangroves (Rhizophoramangle) which fringe protected areas of two lagoonal cays, TwinCays, Belize. During 7 of 8 months surveyed by in situ observation,swarms were present but differed in size from small cylindricalswarms (5–10 cm diameter) to bands extending up to 1200m Swarms were never observed at night Swarms formed at dawnwhen light intensities reached an average value of 13.82 (logioquanta cm^–Abstract. s^–1) and dispersed at dusk atsimilar intensities Swarms observed in June formed earlier anddispersed later in the day than swarms observed in January,their swarming behavior followed seasonal changes in light intensityMean dioithonan density in swarms (10 ml^–1) was much higherthan the mean density (0 15 ml^–1) of non-swarming dioithonansaround mangrove prop roots. In open water 3–5 m away fromthe mangroves, mean dioithonan density was 7 9 x 10^–5ml¹ during the day, and 2 68 x10^–3 ml^–1 at nightSwarms were composed predominantly of adults and copepodid stagesIV and V, although younger copepodid stages could be presentNauplii were never present. The ‘average copepodid stage’for all 95 swarms sampled was 5 3, where 6 0 represents a swarmwith only adults In open water 3–5 m away from the mangroves,the youngest copepodids (stage one) dominated the dioithonanpopulation during the day. At night when swarms dispersed toopen waters, average copepodid stage was higher (3 5) comparedwith the day value (1.2) in open waters. Although densitiesin swarms were higher in June than January, average copepodidstage in June was higher (5 6) than that in January (4.9). Ahigher percentage of adults were females during June than January.Therefore higher densities did not result from increases ofsmaller stages in swarms, but perhaps changes in behavior orpopulation structure.