Associative learning in relation to foraging of Neotropical stingless bee Trigona Rebecca Johnson Department of Fish and Wildlife Resources, University of Idaho ABSTRACT Organisms, if capable , will invest in learning when it results in the improvement of overall fitness. L earning ability may allow organisms to outperform competitors or better survive in a variable environment . Order Hymenoptera exhibits a broad spectrum of learning abilities, including landmark navigation, time place learning, and associative learning. These abilities are fairly well studied in North American genera such as Apis , but remain relatively unexplored in Neotropical species such as those in the g enus Trigona , a n essential stingless bee pollinator of rainforest ecosystems. In this study, I test whether Trigona bees in Monteverde, Costa Rica possess the capacity to associatively learn concerning foraging and extraneous markers. I conclude that like other Hymenoptera , the Trigona species studied have the capability to both associate an extraneous landmark with a reward and use this information to maximize foraging strategy. This association minimizes Trigona nt searching for a food source, thus improving colonial resource stores and supporting a larger colon ial population. RESUMEN Los organismos, de ser capaces, van a invertir en el aprendizaje si resulta en una mejorÃa de su Ã©xito reproductivo. La capacidad de aprendizaje pued e permitir a los organismos a superar competidores o a sobrevivir en ambientes variables. El Orden Hymenopthera presenta un amplio espectro de habilidades de aprendizaje, incluyendo navegaciÃ³n por medio de puntos de referencia, aprendizaje tiempo espacio, y aprendizaje por asociaciÃ³n. Estas habilidades son bastante estudiadas en gÃ©neros norteamericanos como Apis, pero permanecen relativamente inexploradas en especies neotropicales como aquellas del gÃ©nero Trigona, una abeja sin aguijÃ³n polinizadora de ecosi stemas del bosque lluvioso. En este estudio, puse a prueba si las abejas Trigona tienen la capacidad de aprendizaje por asociaciÃ³n con lo que respecta a forrajeo y marcas externas en Monteverde, Costa Rica. ConcluÃ que como otras Hymenopteras, la especie e studiada de Trigona tiene la capacidad de crear puntos de referencias externos o por asociaciÃ³n con una recompensa y usar esta informaciÃ³n para maximizar su estrategia de forrajeo. Esta asociaciÃ³n minimiza el tiempo y energÃa invertido por Trigona en busca r fuentes alimenticias, por lo tanto mejorando las reservas de recursos de la colonia y apoyando una poblaciÃ³n mayor de la colonia. INTRODUCTION Learning is the adaptive modification of behavior based on experience (Alcock 2005). The process of learning is a costly investment (Kroodsma & Canady 1985) and therefore animal s will not commit resources to it haphazardly. Only when an animal can account for environment al instability, providing a counter balancing effect to the associated cost, will an organism make an investment in learning (Alcock 2005) ; the benefits must outweigh the cost. Benefits may include maximizing food quality and supply, communication, risk avoidance, and many others. These in better adapt to a variable environment, thus outcompeting competitors and ensuring an ecological niche . Learning itself varies drastically in degrees of complexity and rewards among organisms .
Members w ithin the Order Hymenoptera are able to learn along a broad spectrum of difficulty . Some Hymenoptera exhibit relatively simple learning abilities such as habituation, where the insects l earn to ignore unimportant stimuli (HÃ¶ lldobler & Wilson 1990 ) , while others show much more complex abilities such as t ime place learning , where they learn to associate a reward with a specific time and place (Breed 2002, Murphy & Breed 2008) . Trigona bee s (Apidae: Meliponinae) display time place learning abilities, a skill reserved for those foragers who specialize on food sources renewed ea ch day on a fairly regular schedule, such as nectar supply in flowers ( Murphy & Breed 2008). When regular nectar feeding stations were set up, it was found that Trigona b ees anticipated feeding each day and appeared to previously resource rich flowers when no incentive was being o ffered, showing they learned to associate a location with a reward ( Murphy & Breed 2008) . Somewhat less complex than time place learning is the ability of Hymenoptera to associate an external factor with a given specific result ( Wilson 1971, HÃ¶l ldobler & Wilson 1990 ) . One of the most famous examples is when Pavlov (1927 ) trained his dog to associate a bell with food . Ea ch time he would ring the bell he fed the dog, and eventually, th e dog would salivate when it heard the bell, with or without a reward presented. Again, this learning method is on a spectrum of simple to complex. A simple behavior is the ability of individuals to learn the c olony scent . Scent is learned almost immediately after metamorphos is from the pupa stage , as t his is crucial to not only find and recognize fellow colony members, but also to recognize and defend against intruders (HÃ¶ lldobler & Wilson 1990 ) . On the complex end of the spectrum, H ymenoptera, along with other insects, may also learn to a ssociat e visual cue s with a n expected response ( Weiss & Papaj 2003, L eadbeater & Chittka 2007 ) . I n often occurs when foraging (HÃ¶lldobler & Wilson 1990) . Optimal foraging theory dictates that animals have evolved to select for the highest reward possible with the least cost associated gain the most calories, but spend the least amount of time and energy ( Alcock 2005 ). This allows the animal to most efficiently dedicate resources toward biomass production rather than a waste product of metabolism. Therefore, if an animal can distinguish a pattern that maximize s foraging efficiency, they will commit to the cost of learning this pattern. It has been shown that insects can associate cues , such as flower color or landmarks, with higher caloric rewards (Leadbeater & Chittka 2007). Ants of the genus Cataglyphis (Formicidae) are able to use geographical landmarks to navigate to and from food sites (HÃ¶lldobler & Wilson 1990). Honeybees (Apidae : Apis mellifera ) demonstrate the ability to associate landmarks with caloric rewards if trained (Cheng et al. 1987). Not only do the bees recognize patterns in the landmarks, but also relative sizes of the landmarks. They most often used the landmarks that were closest, then the largest, to the resource. The honeybees would then use these landmarks to calculate their flight vectors back to their colonies. Like honeybees, Trigona bees are eusocial insects. They exhibit age based polyethism : younger bees are assigned jobs within the nest, while older bees carry out the more dangerous foraging duties (Robinson 1992). food, they make their efforts as efficient and rewarding as poss ible. If they are capable, it would be a worthy learning investment to minimize search efforts by associating a visual cue with caloric rewards. I expect that Trigona bees do possess this ability, much like the honeybees and Cataglyphis ants (Cheng et al. 1987, HÃ¶lldobler & Wilson 1990) . Scientists have directly observed the ability of Trigona to associate parameters such as flower color and odor with a reward 1981), but within the genus, species have variable learning and
communicat ion capacities (Biesmeijer & Slaa 2004) and it is not known whether they have the ability to lea rn more complex associations . In this study I investigate whether Trigona bees can associate landmarks with a food source, and then use these landmarks to determine the location of a reward in the absence of the ability to follow a set vector. I expected that after training the bees to associate a given landmark with a sugar incentive, they would follow the landmarks even in the absence of a reward as Murphy and Breed (2008) found in their time place learning experiment. I predicted that Trigona bees do indeed have the ability to associate an extraneous landmark with a reward , as demonstrated by their Formicidae and Apidae relatives ( Cheng et al. 1987, HÃ¶lldobler & Wilson 1990 ) . METHODS Study Sites I worked at three sites in pre montane wet forest (Holdridge 1967) , each with an individual Trigona colony , from April 19 to May 6 , 2011 . Colonies were found in the roots and trunks of mature trees . Two nests were located on the Campbell property and one at the Monteverde Biological Station. On the Campbell property , one nest was located in a regenerating bull pen and the other in the back of the pasture very near the forest edge , while the nest at th e station was located in the garden. Both nests on the farm had large open areas to arrange flower patches, whereas at the station, flower patches were set up on a walkway due to limited area .
Preparation and Observation U sing 60 Impatiens walleriana potted plants of varying colors, I organized two plots (30 plants in each) approximately 5 m away from the nest and 5 m away from each other (Fig. 1) . Though these 60 plants consisted of five different colors, I made the color composition of each patch essentially the same. I chose I. walleriana because , a ccording to Murphy & Breed (2008) , they are effec tive attractors for Trigona bees, and were also relatively inexpensive and easy to maintain. In one patch, I add ed 2 0 microliters of a 50 percent hone y water solution to each flower , as performed by Murphy & Breed (2008) , with another 5 percent per volume of vanilla added . While the honey acted as a reward, the vanilla was an attractant, as suggested by Mui (2005). I used the vanilla with the purpose of attaining a bigger sample size. Around each of these p lants, I placed makeshift yellow landmarks (30 cm tall) (Fig. 1) . I made these landmarks by taping together three painted barbecue skewers , selecting the color yellow because it is attractive to Trigona (Mui 2005). To the second patch of flowers , I add ed 20 microliters of water , again adding 5 percent vanilla per volume to the water . No landmarks were placed in this patch (Fig. 1). I then observe d the site two hour s per day counting how many bees were drinking nectar or hovering over a patch , for four days, except for the first site at which I observed only for three days . I alternated observing each patch (landmarks and no landmarks) every minute . I began taking ob servations when Trigona colonies became active in the mornings ; t his varied among colonie s due to weather and site variation . Each evening after b ees returned to their nests or each morning before they became active for the day, I refill ed the flowers following the same procedure as discussed above. I also sw itched landmark and no landmark flower patch locations each day to prevent the bees from learnin g a set vector for rewar d location . This period served to train the bees to associate the extraneous landmarks with a reward. After this observation period, I had one experimental day. I ref illed the flowers following the same procedure except I transfer red the landmarks to the water patch . I then observe d the bees for only that next morning following the same procedure as discussed above: observe for two hours alternating between patches each minute. Figure 1. Trigona associative learning test site set up. All three sites had essen tially the same set up with 5 m gaps and 60 flowers; however, the B iological S tation site had only 2 m gaps and 30 plants. Vanilla was added to both solutions as an attractant agent for the bees. Solution was added daily, a nd at this t ime, flower patch locations were also switched . On experimental days, the landmarks were transferred to the water solution patch.
After this entire process was completed at the bull pen site, I moved all of my materials and conducted the experiment again in the pasture using the same methods, and then again at the Biological S tation. However, by the time I conducted my experiment at the station, half of the potted I. walleriana no longer had flowers due to temporal and physical stresses as a result of flower maturation and moving plants between sites. Therefore, only thirty pla nts were used at this third site, fifteen in each patch. Due to limited space in the garden, patches and the colony were also only separated by 2 m. RESULTS All together, I made 1680 observations: I collected 1320 data points during the observational period and 660 on the experimental days, ultimately equating 15 days spent observing the bee nests . Figure 2. Mean ( Â± 1 SE ) visitations/minute of Trigona to flower patches a ) with honey and landmarks, and with water and no landmarks during training periods (N = 660 for each patch) and b ) with water and landmarks, and with honey and no landmarks for the experimental day (N = 180 for each patch) in Monteverde , Costa Rica . Differences are significant in both trials. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 Mean Visitations/ Minute 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 Mean Visitations/Minute Landmarks No landmarks a b
The Trigona bees significantly followed the landmarks not only when the landmarks were next to the honey ( One Way ANOVA, F 1, 1319 = 8.17, P = 0.004 ; Fig. 1a ) but also when they were moved to the water patch ( One Way ANOVA, F 1, 359 = 29.99, P < 0.001 ; Fig. 1b ) . The honey nearly 50 percent more often than the patch w ith only water, a mean of 0.30 visitations/ minute (SE = 0.022) at the landmark patch versus a 0.21 mean visitation rate (SE = 0.017) at the patch without landmarks (Fig. 1a). On the experimental day, the mean visitation rate was still higher at the landma rk patch at 0.27 visitations/ minute (SE = 0.035) than the 0.06 visitation rate (SE = 0.018) at the honey patch without landmarks bees were over four times more likely to visit the landmarks patch even without a reward. DISCUSSION Trigona bees can and do learn to associate external factor s with reward s . I established that Trigona prefer sugar filled flowers with landmarks present over water filled flowers without landmarks. Then, e v en though no reward was offered with the landmarks on the experimental day, the bees still followed the landmarks over visiting the flower s offering honey. These findings correlate with the conclusions of Cheng et al. (1987) for their honeybee experiment ( Apis mellifera ) and conclusions for Cataglyphis ants . Trigona bees seem to possess similar landmark learning abilities to other social H ymenoptera. This fairly complex behavior nature ( Wilson 1971 ). If individuals are able to specialize, as they do in a caste system, they can commit more resources to excelling at a given task rather than partitioning resources to attempt all necessary tasks for survival. Within Trigona which exhibits temporal polyethi sm, learning plays a critical role; throughout a lifetime, individuals must learn to fill different roles and all asscociated tasks. As bees age and move into foraging roles within the colony, it is in their best interest to optimize foraging efficienc y, m aking associative learning in relation to landmarks and food sources extremely beneficial . Rel ating these findings back to optimal foraging theory, by discerning an association for where the greatest rewards are found, these foragers decrease the amount of time and energy they invest, increasing their net energy gain. This in turn results in a lar ger resource base for the entire colony, allowing higher reproduction rates and a longer colony lifetime. A higher population would allow a colony to better outcompete or defend against other colonies, and make it more resilient to disturbances. Associati ve learning in foragers could However, finding a bright yellow stick next to a reward does not often occur in natur e, nor do 50 percent nectar rewards and giant patches of flowers directly in front of the nest. Futur e studies may consider gauging Trigona bees markers , rewards, and flower patch locations . We must also consider the repercussions of these findings. If stingless bees are more sensitive to landmark placem ent than we originally thought , this makes them much more vulnerable to habitat disturbances such as deforestation or road and trail development: if they cannot find their previous landmarks, they may have difficulty finding the food. This would be devasta ting to both the environment and human society if Trigona bees were to start disappearing at the same rates as the honeybees in the U.S. dying of Colony Collapse Disorder (CCD) . Bees provide a cosmopolitan ecosystem service; just within the United States , bee
pollination services are estimated at nearly $10 billion per year (Nabhan & Buchmann 1997) . As pollinators, Trigona bees are critical to the rainforest plants, and therefore to all the animals who eat those plants, and so on and so forth. The role of small scale landmarks in the activity of these stingless bees must be understood in greater detail if we are to be able to guess how continuing deforestation and even climate change might affect these pollinators. ACKNOWLEDGMENTS I would like to thank Anjali Kumar for her guidance and expertise throughout this project, along with her patience with my constant supply of sometimes trivial questions. I would also like to thank the Campbell family for gracious ly allowing me to explore their farm, set up two of my work sites, and use their supplies for jobs such as transporting and watering flowers. I also would like to thank Moncho and Gisella FernÃ¡ndez for answering being my available gophers when I need ed supplies they were al ways ready to help in any way possible . A special thanks to Gisella for translating my abstract! LITERATURE CITED A LCOCK , J . 2005. Animal Behavior. Pp. 88, 89, 210. Sinauer Associates, Inc., Sunderland, Massachusetts. B IESMEIJER , J.C. AND E. J. S LAA . 2004. Information flow and organization of stingless bee foraging. Apidologie. 35: 143 157. B REED , M.D., E.M. S TOCKER , L. K. B AUMGARTNER , AND S.A. V ARGAS . 2002. Time place learning and the ecology of recruitment in a stingless bee, Trigona amalthea (Hym enoptera, Apidae). Apidologie. 33: 251 258 C ANE , J. H. AND V. J. T EPEDINO . 2001. Causes and extent of declines among native North American invertebrate pollinators: Detection, evidence, and consequences. Conservation Ecology. 5:1. C HENG , K.,T.S. C OLLETT , A. P ICKHARD , AND R. W EHNER . 1987. The use of visual landmarks by honeybees: Bees weight landmarks according to their distance. Journal of Comparative Physiology A. 161: 469 475. D YER , F. AND J. D ICKINSON . 1996. Sun Compass Learning in Insects: Representat ion in a Simple Mind. Current Directions in Psychological Science. 5: 67 72. H Ã–LLDOBLER AND W ILSON . 1990. The Ants . Pp. 366, 368. The Belknap Press of Harvard University Press, Cambridge, Massachusetts. H OLDRIDGE , L.R. 1967. Lifezone ecology . Tropical Science Center, San Jose, Costa Rica. K ROODSMA , D. E., AND R. A. C ANADY . 1985. Differences in repertoire size, singing behavior, and associated neuroanatomy among Marsh Wren populations have a genetic basis. Auk 102: 439 446. L EADBEATER , E. AN D L. C HITTKA . 2007. Social learning in insects From miniature brains to consensus building. Current Biology. 17: 703 713 M UI , C. Y. What kind of flower attracts stingless bees? An experiment. 2005. Int ernational Field Biology Course, Center fo r Tropical Forest Science. M URPHY , C. M., AND M. D. B REED . 2008. Time place learning in a neotropical stingless bee, Trigona fulviventris GuÃ©rin (Hymenoptera: Apidae). Journal of the Kansas Entomological Society. 81: 73 76. N ABHAN , G.P. AND S.L. B UCHMANN . 2009. Services provided by pollinators. In: Nature's services: societal dependence on natural ecosystems , G. Daily , ed . pp. 37 46. Island Press, Washington, D.C. P AVLOV , I.P. 1927. Conditioned reflexes: An investigation of the physiological activity of the cer ebral cortex. Oxford University Press, London, England. P ESSOTTI I., AND A. M. L Ã‰ Ã‰NÃ‰CHAL . 1981. Aprendizagem em abelhas. I DiscriminaÃ§Ã£o simples em onze espÃ©cies, Acta Amaz. 11: 653 658. As cited in: B IESMEIJER , J.C. AND E. J. S LAA . 2004. Information flow and organization of stingless bee foraging. Apidologie. 35: 143 157. R OBINSON , G.E. 1992. Regulation of division of labor in insect societies. Annual Review of Entomology. 37: 637 665. W EHNER , R., B. M ICHEL , AND P. A NTONSEN . 1996 . Visual n avigation in insects: Coupling of egocentric and geocentric i nformation . The Journal of Ex perimental Biology. 199: 129 140. W EISS , M. R. AND D. R. P APA J . 2003 . Colour learning in two behavioural contexts: how much can a butterfly keep in mind? Animal Behavoiour. 65: 425 434. W ILSON , E. O. 1971. The Insect Societies . Pp. 217. The Belknap Press of Harvard University Press, Cambridge, Massachusetts, and London, England.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00231
El aprendizaje asociativo en relacin con la bsqueda de alimento de la abeja Neotropical sin aguijn, Trigona
Associative learning in relation to foraging of Neotropical stingless bee Trigona
Organisms, if capable, will invest in learning when it results in the improvement of overall fitness. Learning ability may allow organisms to outperform competitors or better survive in a variable environment. Order Hymenoptera exhibits a broad spectrum of learning abilities, including landmark navigation, time-place learning, and associative learning. These abilities are fairly well studied in North American genera such as Apis, but remain relatively unexplored in Neotropical species such as those in the genus Trigona, an essential stingless bee pollinator of rainforest ecosystems. In this study, I test whether Trigona bees in Monteverde, Costa Rica possess the capacity to associatively learn concerning foraging and extraneous markers. I conclude that like other Hymenoptera, the Trigona species studied have the capability to both associate an extraneous landmark with a reward and use this information to maximize foraging strategy. This association minimizes Trigonas time and energy spent searching for a food source, thus improving colonial resource stores and supporting a larger colonial population.
Los organismos, de ser capaces, van a invertir en el aprendizaje si resulta en una mejora de su xito reproductivo. La capacidad de aprendizaje puede permitir a los organismos en superar a los competidores o a sobrevivir en los ambientes variables. El Orden Himenptera presenta un amplio espectro de habilidades de aprendizaje, incluyendo la navegacin por medio de los puntos de referencia, el aprendizaje tiempo-espacio, y el aprendizaje por asociacin. Estas habilidades son bastante estudiadas en los gneros norteamericanos como el Apis, pero permanecen relativamente inexploradas en las especies neotropicales como aquellas del gnero Trigona, una abeja sin aguijn polinizadora de los ecosistemas del bosque lluvioso. En este estudio, puse a prueba si las abejas Trigona tienen la capacidad de aprendizaje por asociacin con lo que respecta al forrajeo y marcas externas en Monteverde, Costa Rica. Conclu que como otras Himenpteras, la especie estudiada de Trigona tiene la capacidad de crear puntos de referencias externos o por asociacin con una recompensa y usar esta informacin para maximizar su estrategia de forrajeo. Esta asociacin minimiza el tiempo y la energa invertido por Trigona en buscar fuentes de alimento, mejorando as las reservas de los recursos de la colonia y apoyando a una mayor poblacin colonial.
Text in English.
Costa Rica--Puntarenas--Monteverde Zone--Monteverde
Monteverde Biological Station (Costa Rica)
Costa Rica--Puntarenas--Zona de Monteverde-Monteverde
Estacin Biolgica de Monteverde (Costa Rica)
Tropical Ecology Spring 2011
Foraging behavior in animals
Ecologa Tropical Primavera 2011
Conducta de alimentacin en animales
Abejas sin aguijn
t Monteverde Institute : Tropical Ecology