Jennifer Fewell, PhD - US grants

A key requirement of division of labor in social insect colonies is the ability to respond flexibly to environmental change. This ability is crucial to survival in variable environments. Recent evidence suggests a significant part of observed behavioral variation in social insect colonies is genetically based. In the research, the role of genetic variation in colony flexibility and task regulation by the honey bee, Apis mellifera will be examined. Dr. Fewell will examine stimulus threshold model which describes how genotypic diversity may function as a mechanism for regulating division of labor in social insects. This model proposes that individual workers have genetically linked thresholds to perform a given task, and that a range of genotypic within a colony provides a range of tendencies to perform that task. From this model, one can predict that the genotypic composition of the population performing a task will change as environmental conditions and, therefore, stimulus environments change. Dr. Fewell will test this prediction by changing colony need for pollen and measuring associated changes in the genetic composition of the pollen and nectar foraging populations. This will be the first experimental test of the role of genotypic variation in social insect foraging regulation. These studies will address the important question of whether current hypotheses on the role of genotype in honey bee foraging apply to naturally mated colonies. They also will involve the application of a new set of molecular genetic techniques, PCR and RAPD analysis, to behavioral questions in social insects.

Division of labor is a fundamental component of complex social systems, in groups as diverse as insects and mammals. In all social groups, division of labor involves specialization by individuals on certain tasks. However, for division of labor to function in a variable environment, it must allow for changes in task performance. In honey bees, Apis mellifera, task specialization depends on: (1) age polyethism, with individuals of different age groups performing different tasks within a colony, and (2) genetically-based individual preference for certain tasks. The long-term objective of this project is to determine the mechanisms for individual and societal task regulation in honey bees. More specifically, it will examine mechanisms for and constraints on individual flexibility in foraging behavior, and determine how individual variation in task performance is integrated into colony-level regulation of foraging. The project will use a variety of behavioral and genetic techniques, including use of instrumental insemination to construct honey bee colonies containing a variety of worker genotypes, and allozyme electrophoresis and DNA analyses (RAPD markers) to identify genetic subfamilies within colonies. Three principle questions will be addressed: (1) What are the relative contributions of individual flexibility and genotypic diversity to colony-level foraging plasticity?, (2) How do genetic and developmental constraints on individual behavioral flexibility constrain foraging behavior at the colony level?, and (3) How well do models of foraging flexibility developed using domesticated honey bee colonies predict the behavior of social groups under conditions of natural selection? This study will provide information crucial to understanding the evolution of division of labor in complex social systems, and mechanisms of behavioral flexibility in social insect colonies.

Genetic and Phenotypic Variation in Foraging Behavior of African and European
Honey Bees

Principle Investigators: Jennifer H. Fewell and Jon F. Harrison, Arizona State University

A fundamental question in integrative biology is how genetic variation relates to variation in an animal's survival and reproductive strategies. One way to answer this is to determine the genetic basis for behavioral differences between animal populations and then assess the ecological effects of those differences. This project examines foraging strategy differences between neotropical African-derived (Apis mellifera scutellata) and European (Apis mellifera ligustica) honey bees. It addresses to the questions of: (1) how genetic variation between African and European workers influences foraging strategy, and (2) how variation in worker foraging strategy affects differences in colony growth rates between the two subspecies.
Population-level research suggests that African honey bees out-compete European bees primarily through faster rates of colony growth and reproduction (swarming). Because honey bees are social insects, these differences in colony strategy are generated by collective differences in the behavior of the colony's individual workers. Prior research by the investigators suggests that a major behavioral difference between African and European workers is in foraging. African workers show a higher tendency than European workers to collect pollen. Because pollen is the primary food source for developing brood, this difference can have profound effects on colony growth. The first objective of this research is to understand the mechanisms producing individual differences in pollen and nectar foraging. The investigators will test the hypothesis that differences in resource choice between African and European foragers are driven by differences in their sensitivity to stimuli affecting pollen collection. They will determine whether co-fostered African workers respond more readily to stimuli known to affect pollen foraging, and examine differences between African and European workers in genomic regions (quantitative trait loci or QTL's) known to be associated with pollen foraging behavior in European bees.
Preliminary data also suggest that African and European workers differ in work effort during foraging, and that this variation may be mediated by differences in metabolic capacity. The second objective is to test the hypothesis that variation in metabolic rate between African and European honey bees is genetically based, and is linked to differences in resource preference and foraging effort. The researchers will compare metabolic rates, foraging load sizes and foraging rates of co-fostered African, European and hybrid workers to determine how these traits are inherited. Then, quantitative trait locus (QTL) mapping will be performed to locate genomic regions influencing variation in flight metabolic rates. These will be compared to known loci affecting body size and preference for pollen versus nectar collection.
The third objective is to determine whether variation in individual foraging behavior affects colony-level growth rates. The investigators will: (a) measure pollen intake rates and colony growth rates in African and European hives, and (b) experimentally manipulate the genotypes of the foragers in a hive (African versus European), to measure the effect of forager genotype on brood production. This research integrates behavior, physiology and genetics to generate a more complete understanding of the proximate mechanisms generating complex behavioral traits such as foraging. It also examines the consequences of individual behavioral differences on the ecological success of these competing subspecies.

0129319
Fewell
This award supports Jennifer Fewell and graduate and undergraduate students from Arizona State University in a collaboration with Juergen Gadau of Institute for Behavioral Physiology at the University of Wuerzburg, Germany. The aim of the international project is to examine the extent and consequences of hybridization between two native seed-harvester ants, Pogonomyrmex rugosus and Pogonomyrmex barbatus. The German and U.S. groups will test the hypothesis that hybridization in that system does not universally produce reduced hybrid fitness, as expected by current models. They will also test the hypothesis that colonies respond to hybridization by preferentially using sperm belonging to different species. They will use morphological, molecular, and behavioral techniques to determine the extent of introgression between species and to measure fitness components of hybridized and parental colonies. In addition, they will perform controlled mating experiments to determine the effect of hybridization on caste determaination. The work plan provides for extensive participation by graduate students in the international travel and research.

Use continuation pages as needed to provide the required information in the format shown below. Start with Principal Investigator. List all other key personnel in alphabetical order, last name first. Name Organization Role on Project Bustoz, Joaquin Arizona State University Project Role: Dr Bustoz will serve as MARC Director. As such, he will represent the MARC program within the university and when necessary outside the university. He will have primary responsibility for recruiting MARC trainees. He will advise MARC trainees majoring in Mathematics, and in related areas such as Statistics. He will be the liaison between MARC and participating mathematically based disciplines such as Physics. As Director of the SUMS Institute he will administer SUMS staff in the

Collaborative Proposals 0446415 and 0446342
Title: Division of labor in communal groups
Jennifer Fewell, Susan Bertram, Penny Kukuk

Project Summary: Social groups from insects to humans divide their labor among individuals so that different group members specialize on different tasks. This division of labor has long been considered one of the key adaptations of sociality. Recent work suggests that its underlying mechanisms involve simple rules of individual behavior, which when combined with interactions among individuals generate complex group level properties (self-organization). Self-organization is thought to occur across all biological levels, but this has proven difficult to test experimentally. Social groups provide an important opportunity to test this assertion, because individuals can be observed and interactions between group members quantified.
This research will examine how local interactions among the members of simple social groups could produce division of labor during early social evolution. The central hypothesis is that division of labor can self-organize if interactions among group members amplify differences in their probability of performing tasks. If so, division of labor could emerge even in the absence of selection, and thus could appear spontaneously at the origins of sociality. The alternate hypothesis will be tested, that division of labor is produced primarily via selection for increased differences in task performance after the formation of social groups.
The research will use two different modeling approaches to explore how interactions among group members could potentially generate division of labor. The first assumes that individuals within any group vary naturally in their tendency to perform tasks, and that amplification of these differences generates division of labor. The second borrows from economics by treating group members as "companies" that compete for tasks; tasks become distributed among group members based not on initial variation, but on differences in individual success at performing them. To test these models and to examine how division of labor changes during early social evolution, the researchers will compare the behavior of groups made up of normally social individuals to those of solitary individuals that are forced to form social groups. Two taxa have been identified in which solitary and social populations can be compared. These are the ground nesting bee genus Lasioglossum, which contains both solitary and communal species, and the seed harvester ant species, Pogonomyrmex californicus, which has two populations of the same species that initiate nests alone versus in communal groups. Both of these taxa can be brought into the laboratory, allowing artificial social groups to be constructed and observed. This provides a unique opportunity to test how social structure and division of labor change during the transition from solitary to social living.

Broader Impact: This work makes an important contribution to our understanding of the mechanisms underlying division of labor, an essential component of social organization within human cultural as well as biological systems. It also addresses the broad question of how properties emerge across levels of biocomplexity. The research actively engages undergraduate and graduate students in the process of science, from data collection to analysis, presentation, and publication. Students will have the opportunity to participate in both the theoretical and experimental components of the research. The PIs have exceptional records of undergraduate mentoring, including directing undergraduate research programs at their respective institutions.

Like the mule, hybrids between species or lineages are often considered an evolutionary dead end. However, recent studies suggest that hybrids can be important in shaping evolution. While most hybrid gene mixtures are unsuccessful, some are beneficial and result in new strategies to cope with changing ecological and social environments. In Pogonomyrmex seed-harvester ants, hybridization has produced a unique social and reproductive structure, in which queens must mate with males of two different genetic lineages to produce both workers and daughter queens. Daughters of the queen and males of the same lineage develop into queens. Hybrid daughters between the queen and males of the alternate lineage become sterile workers. This system of genetic caste determination violates the widely held assumption that whether a female becomes a worker or queen has to be based on differences in nutrition or other environmental factors, and not genetics, because gene combinations producing sterility should be selected against. This system occurs within two closely related harvester ant species, P. rugosus and P. barbatus, leading to the hypothesis that it originally evolved as a result of incomplete hybridization between the two species that lead to the development of two interdependent genetic lineages. However, preliminary evidence suggests that this is not the case and that it may have evolved initially in one species (P. barbatus) then spread to the other. To test these hypotheses, the extent and consequences of hybridization within and between these species of seed-harvester ants will be examined using mitochondrial and nuclear genetic analyses. This work will contribute to our understanding of the role that hybridization plays in evolution, especially the evolution of social systems. This research will also provide opportunities for a graduate student and several undergraduates to learn molecular techniques and analyses with broad applications in evolutionary biology.

Self-organizing animal and human groups have increasingly become the focus of research by scientists interested in social dynamics. While a substantial amount of literature exists on the behavioral interaction patterns found in animal groups, there is not a comparable body of work in the social sciences. From hunter-gatherers to city-dwellers, structured gatherings of humans appear in all cultures. These groups range from married couples and co-workers to large crowds and neighborhoods, with each type having a distinct structure and history. What is not clear, however, is how individuals, each with unique attributes and preferences, contribute to the formation of these groups. Even less is known about how the socio-developmental processes observed in groups modify its constituents. Because of this complexity, this study brings together a multi-disciplinary team that integrates human development, computer simulation, biology, mathematics, and geography to study how fundamental social processes are critical to human development and life-course trajectory. To investigate the fundamental properties of sociality, this study will, for 3 years, observe and catalog how and where 3-5 year old children form groups and dyadic friendships. The study of young children is fortuitous for answering process-driven questions about group formation and group stability for several reasons: (1) This is the first time that many of the children are consistently exposed to a large number of peers -- the sizeable pool of eligibles can provide information about the selection process in the formation and evolution of groups; (2) Given the relative social inexperience of this age group, there should be basic and simple process components common to all social entities (e.g., exchange of communicative signals, role differentiation); and (3) There is long-term societal utility for studying children's abilities to form and maintain relationships with their peers -- this phenomena has been closely associated with academic and social competencies. At the end of three years of data collection, computer models from the behavioral and geo-spatial data will be constructed to inform scientists and policy makers about how these critical social processes emerge and evolve. Although the focus is on young children, the core of this endeavor is an attempt to understand and model the reciprocal evolutionary dynamics basic to understanding all social processes. As such, the research is informed by multiple scientific disciplines ranging from human development, anthropology, and sociology to computer science, physics, biology, and applied mathematics.

Biomimicry is the practice of imitating nature's forms and processes to more effectively and sustainably meet human design challenges. The conference, Social Biomimicry: Insect Societies and Human Design, will explore how social insects such as ants, bees, and termites can inform human design. Over millions of years, the social insects have evolved tightly integrated societies rivaled only by humans in their scope and sophistication. Individual social insect workers are relatively simple in their behavioral capabilities, but through their collective behavior they have solved many practical problems inherent in large social groups organizing and communicating around common goals. Already, computer models based on ant foraging behavior are being used to route traffic in busy communications networks such as the Internet, and architects are mimicking termite mounds to build passively-cooled buildings that save energy. However, the implementation of social-insect-inspired design has been limited, in part, by lack of communication and coordination across disciplinary and professional boundaries. The conference will address this shortcoming by bringing together biologists - including leaders in the field of social insect biology and complexity theory - with designers, engineers, and businesspeople, with the goal of promoting the exchange of concepts, perspectives, and tools. This activity should (1) enrich biology, by revealing new applications of basic research and stimulating new areas of problem-driven research, and (2) advance the field of biomimicry, which is an important engine for innovation and economic growth. The conference also features a public Social Insect Science EXPO! which will connect researchers at Arizona State University with members of the Phoenix community, including underrepresented students from local K-12 schools. The conference is organized and led by a panel of graduate students from ASU's Social Insect Research Group, and encourages participation by undergraduate and graduate students from the U.S. and abroad. The conference provides a unique opportunity in training future scientists with the leadership skills necessary to organize and run a professional meeting and to communicate and apply their work in an interdisciplinary context. Conference results will be disseminated broadly through peer-reviewed and popular articles and Internet resources.

Three major elements, Carbon, Nitrogen, and Phosphorus, determine how organisms grow, and influence how systems, from groups to ecosystems, function. As a result, organisms have developed behavioral and physiological mechanisms to adjust to limits in element availability and grow successfully in the face of nutritional challenges. Like organisms, social groups that collect and store food must also adapt to changes in nutrient availability. This study focuses on how a complex social group, the leafcutter ant colony, uses behavioral strategies to integrate intake of these elements with colony growth. Because social insect colonies must allocate effort to both foraging and brood care, colony growth and success depend on balancing work allocation across activities. This study will employ biochemical and behavioral techniques to determine how relative balances of these three key elements affect colony growth, as well as foraging and brood-care decisions by the individual workers within colonies. It is expected that colonies will respond to limitations by altering growth patterns and shifting efforts between tasks to balance different tasks associated with colony maintenance, growth, and reproduction around nutrient availability. The findings will integrate our understanding of how fundamental biological properties such as growth scale across levels of organization, especially in complex social groups like the social insects. Social insects are paralleled only by humans in their social complexity and level of organization. Thus, they serve as a relevant model to address the question of how resources and social organization are integrated. This research has broader impact in student training and outreach. Multiple undergraduates will be involved in the project through hands-on research and will participate in an associated seminar: the majority of students move on to the Ph.D. or other postgraduate training. The research will be featured in public educational outreach including the Ask-A-Biologist website (askabiologist.asu.edu) and local K-12 demonstrations.

One of the most significant traits of an animal is its body size, which has profound consequences for its structure, function, and ecology. Likewise, when animals form social groups, individual and group properties may change, or scale, in response to changes in group size. In highly social insects such as ants and bees, colony size varies tremendously, from just a few individuals up to thousands or even millions. Moreover, their colonies are so tightly integrated that they function like organisms, and may thus be subject to similar scaling relationships. This project will investigate how the organization and output of work performed by ant colonies scales with colony size. Ant colonies exhibit a division of labor in which different workers specialize on different tasks such as foraging and brood care. By manipulating colony size and measuring task performance, the researchers will test how colony size affects division of labor and other patterns of activity in the seed-harvester ant Pogonomyrmex californicus. Preliminary evidence suggests that as colony size increases, division of labor increases. The project will also analyze the scaling of brood production, the major form of work output in an ant colony and a key parameter in models of social insect life history and evolution. The results will provide insights into the organization, development, and evolution of insect societies and other social groups. Social insects are among the most abundant and ecologically successful animals on earth; they also serve economically important roles (e.g., as pollinators and pests) and are leading models for the study of social behavior and complexity. In addition, the project will provide training opportunities for a graduate student and undergraduate research assistants. The investigators will also engage in public outreach based on their research, including demonstrations to K-12 students and contributions to online educational resources.

Social insects, including ants and termites, as well as many bees and wasps, are among the most diverse and ecologically significant organisms on earth. These species live in complex societies whose governance comes not from a central authority but from the interactions of their individuals with each other and with the environment. From this decentralized system, there emerge consistent patterns of task allocation, division of labor, and resource exploitation. The multiscale study of the emergence of these patterns presents both challenges and opportunities for research and education. The goal of this interdisciplinary research is to develop a general integrated multi-scale dynamical network modeling framework of division of labor in social insects. Rigorous mathematics will be integrated with extensive field and laboratory data to study the complex adaptive system of social insect societies. The research team is exploring: 1. How the underlying topology of the interaction network of a colony evolves and adapts at different scales of the organization; 2. How to characterize the crucial feedback mechanisms linking both structure and dynamics of the division of labor in a dynamical environment and 3. How the decentralized social insect system based on many independent and simple individual interactions leads to highly complex dynamics with great network properties such as scalability, robustness/flexibility and simplicity. The investigators use nonlinear differential equations, spatial stochastic processes and kinetic equations to model at the colony level, the individual level, multilevel and at a spatial and task continuum level, respectively.

This research project lives at the intersection of social science, life sciences and applied mathematics. This modeling framework of social insects provides not only a powerful system for examining how network dynamics contribute to the evolution of complex biological systems but also a great opportunity to explore how behavior evolves within complex systems in general. The methods developed may apply in many domains outside of biology, including network routing, optimization theory and robotics. A template integrating interdisciplinary learning for social sciences students, life sciences students and applied mathematics students will be developed through shared research projects at the undergraduate and/or graduate level.

Social groups are built around cooperation, in which individuals work together to benefit the group. Human success as a social species is built around this mutual benefit, but we know little about its evolution and maintenance. Cooperative groups risk being vulnerable to cheating, if a group member takes advantage of the effort of others to secure a disproportionate amount of resources. In groups of relatives, these effects can be offset by kinship. However, the costs and benefits of cooperation among non-relatives are more difficult to assess and explain. The California harvester ant provides a rare opportunity to examine the costs and benefits of cooperation among non-relatives. In this species, some queens practice primary polygyny, in which unrelated queens cooperatively build a new nest and rear offspring together, forming what is essentially a multi-family colony. The proposed work will investigate what ecological conditions favor the evolution of cooperation in this species, and how these cooperative societies deal with the possibility of cheating by some group members. The proposed research will be combined with undergraduate training, in which students participate in field and laboratory research. This builds on a record of mentoring excellence by the principal investigator and graduate student. The researchers will also work in conjunction with the local parks department to provide educational materials highlighting this unusual ant population. This research is fundamental to our understanding of social evolution, and may provide insight into the issues of cooperation and conflict in human societies.

Prior work has shown that colonies in the polygynous population have lower reproductive output than those in a nearby population without polygyny. This suggests polygyny may be a functional response to local environmental constraints. Polygynous colonies also face issues of cheating, if reproduction is non-equitably distributed among queens. Interpretation of these findings is limited, because the polygynous and single-queen populations investigated so far are in different environments. The researchers will quantify the reproductive output of single- versus multi-queen colonies within a single population, to determine whether variation in reproduction is environmentally driven, or an intrinsic colony attribute. This analysis will be combined with a food supplement experiment, to determine whether reproduction in polygynous colonies is resource limited. Finally, relative queen contributions to reproduction versus worker production within polygynous colonies will be analyzed via microsatellites, to determine whether queens share in reproduction equitably. If not, this suggests that subtle cheating may be taking place within these cooperative associations. Data will be stored locally and on ASU server space for a minimum of 5 years, with the expectation of permanently archiving them. Data will be indexed and made available to an open source data depository after publication.

The pattern of lower energy use per gram with increasing size (known as metabolic scaling) occurs across all animal groups, yet the mechanisms behind it are still not understood. Intriguingly, social insect colonies show the same level of reduced metabolism with increasing colony size. This project will combine experiments and mathematical models to determine the relationships between energy use and the organization of work in ant colonies. Ant colonies serve as an exemplar for this question, because their highly coordinated societies are organized around the work needed to sustain colony growth and maintenance, analogous to metabolic processes in organisms. Understanding the mechanisms underlying metabolic scaling has potential applications in physiology, medicine, and agriculture. The energy-based models and experiments in this project will be relevant well beyond social insect colonies, including informing human population scaling issues. The project will connect this research with a Mathematics and Social Biology Co-mentoring Program, to train research teams of undergraduates together in biology and mathematics. Students will participate in a formal training program on social biological principles and simulation modeling, and then participate as cross-disciplinary teams in collaborative research. Existing connections with research and minority programs at ASU will be used to recruit under-represented students into the program. Teaching modules on group size, social behavior, and metabolism for online K-12 use will also be developed in collaboration with ASU's Ask-a-Biologist outreach program. This project is co-funded by the Animal Behavior program in the Division of Integrative Organismal Systems, the Mathematical Biology program in the Division of Mathematical Sciences, and the BIOMAPS program for proposals at the interface of Biology, Math and the Physical Sciences.

Group size is one of the most fundamental attributes of sociality, with important effects on social organization and fitness. Previous work has found that larger harvester ant colonies (Pogonomyrmex californicus) exhibit allometric changes in task performance and activity patterns, including increased worker specialization. They additionally shift the allocation of effort, possibly away from more expensive tasks. Potentially coupled with this, colonies also show a consistent pattern of hypometric metabolic scaling, which matches that of organismal scaling patterns. Because task performance is directly linked to colony growth and metabolism, these results lead to the hypothesis that larger colonies achieve economies of scale in energy use from scaling changes in the organization of work. To examine how colony size influences work organization, the distribution of workers across tasks, individual task specialization, and worker activity levels will be measured in laboratory colonies of varying size. Colony metabolism, brood production and growth rates will be measured simultaneously to link behavioral organization to colony metabolic consequences. Additionally, colony demographics, including worker size and age distributions, will be assessed for their contributions to variation in metabolism. These parameters will be measured for: (a) colonies that vary in size but not age; (b) colonies changing in size ontogenetically over time; (c) size-manipulated colonies. The empirical research will be combined with simulation, differential equation and optimization models to ask how components of task organization and activity might interact to generate scaling changes in colony metabolism and productivity.

The future of Animal Behavior research depends on the next generation of scientists and leaders. This project is to jump-start a new initiative that advances animal behavior science by supporting early-career investigators and promoting their science. At the core of the initiative is a new series of professional-development workshops for early-career scientists to be held annually as part of the Animal Behavior Society (ABS) meetings. In addition, a new committee of established investigators will promote early-career scientists by nominating workshop participants for awards and organizing symposia that highlight their research at international conferences. The committee will also forge stronger connections between the Animal Behavior Society and other professional societies, such as the Society for Integrative and Comparative Biology, by organizing similar symposia and workshops at their annual meetings.

The project begins with a symposium for 35 early-career scientists in May 2019, followed by the formation of peer-mentoring circles that meet biweekly via conference calls to support these researchers throughout the subsequent year. Professionals who specialize on career development of women and minority scientists will organize this initial symposium, and will simultaneously train the members of the new committee to run similar workshop and peer-mentoring circles at the 2019 and 2020 ABS meetings. The new committee will also propose symposia that highlight the research of workshop participants at the 2020 ABS and 2021 Society for Integrative and Comparative Biology (SICB) conferences.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.