Jeffrey C. Schank - US grants

This research aims to further integrate the study of behavior in both animals and robots. The research investigates how rules of animal behavior can be analyzed, quantified, coded, and instantiated in robots. It focuses on individual and group sensorimotor behavior from a developmental perspective using Norway rats (Rattus norvegicus). Robotic rat pups will be built that instantiate sensorimotor rules discovered by careful observation and computational modeling of pups in isolation and in groups at different stages of development. Specific attention will be paid to how these rules change at different stages of development. For the study of behavior, the aim is to use robotic systems to validate rule-based computational models of behavior and generate testable predictions. For robotics, the aim is to transfer information from organisms that start out life as simple sensorimotor systems, but subsequently develop social behaviors, while developing new sensorimotor, learning, and cognitive capabilities. By studying precisely how these capabilities develop, this research will provide additional insight into the emergence of group behaviors in robotics. In the longer run, this research should facilitate the design of group robotic systems that develop greater and greater sensorimotor, learning, and cognitive capabilities over time.

[unreadable] DESCRIPTION (provided by applicant): There is growing recognition that fundamental issues in basic research in the health sciences must develop increasingly powerful theoretical approaches for both understanding and controlling complex systems at multiple levels of organization. The proposed research addresses this problem by extending, theoretically, methodologically, and conceptually, the use on individual-based modeling to the study of individual, social and group behavior. The proposed research is unique in that one of its central aims is to evolve models (e.g., using simulated annealing or genetic algorithms) from data, resulting in surrogate individuals that behave like the complex organisms studied. The goals of the proposed research will be achieved by using infant Norway rat pups to study (i) group dynamics, (ii) sex preferences, (iii) group and individual behavior on thermogradients, (iv) group and individual behavior on illumination gradients, and (v) the interaction of thermo- and illumination gradients. Finally, these models will be used to develop Monte Carlo statistical approaches for studying the behavioral development of complex adaptive systems. [unreadable] [unreadable]

EMT: Collaborative Research: Primate-inspired Heterogeneous Mobile and Static Sensor Networks

Although previous bio-inspired models have concentrated on invertebrates (such as ants), mammals such as primates with higher cognitive function are valuable for modeling the increasingly complex problems in engineering. Understanding primates? social and communication systems, and applying what is learned from them to engineering domains is likely to inspire solutions to a number of problems.
This research involves studying and modeling modes of group behavior and communication of coppery titi monkeys, rhesus macaques, and other primate models, and applying what the investigators learn to the distributed control of heterogeneous mobile and static sensor networks. The investigators will model the social and communication behavior of these primates, which will provide biological inspiration for solving problems in communication and networking. The phases of this research include: 1) identification, interpretation, and translation of primate behavioral models, 2) assessment of the effectiveness of small and large group formations based on primate grouping models in heterogeneous mobile and static sensor networks, 3) development of bio-inspired message-based communications, and 4) development of bio-inspired behavior-based communications. This research aims to achieve a deeper understanding of effectiveness of bio-inspired communications and networking by studying primates, and to establish interdisciplinary research and education in the fields of biological modeling, sensor networking, and robots control.

DESCRIPTION (provided by applicant): Metrics and Models for Measuring Behavioral Development in Rodents. The proposed research has two basic aims: The first aim is to determine what kinds of behavioral metrics in infant rats are most predictive of adult behaviors. The approach developed here, will have broad application to rodents and should be especially relevant to behavioral phenotyping of rodents during development. The second aim is to develop agent-based models that provide a novel way to analyze behavior in both infant and adult rats, and, which will provide theoretical interpretations that are not possible with traditional statistical techniques. These models will be developed for both infant and adult animals. They will be developed on an open-source platform, which is written in Java and usable on all major operating systems. Users who are not experts in the development of agent- based models will be able to use these models-for appropriate data sets-just as they use popular statistical programs to analyze data. The methodological goals laid out in these aims will be accomplished using vasopressin deficient Brattleboro rats, which are a potentially important model organism for several developmental disorders, including autism spectrum disorders. Since juvenile Brattleboro rats exhibit early locomotor and social differences from wild type rats and adult Brattleboro rats exhibit social and cognitive deficits, this is an ideal model organism to use in developing these methods. It is hypothesized that the measurement and analysis of infant rat locomotion and aggregation will be predictive of adult social and cognitive (e.g., learning and memory) behavior. That is, we will be able to predict from metrics of aggregation and locomotion, differences in performance of wild type and vasopressin deficient rats as adults in several standard behavioral tests of learning, memory, and social behavior. Moreover, the agent-based models developed for this project will greatly enhance traditional statistical analyses, allowing for the novel theoretical interpretation of results. Indeed, it may even be possible one day to test out drugs or other therapeutic treatments on simulated animals before investing the time and resources on actual animal testing. PUBLIC HEALTH RELEVANCE: One aim of the proposed research is to develop metrics of the dynamics of early behavior and connect them to adult behavior in rats and ultimately other rodent animal models. A second aim is to build agent-based models for different behavioral tests, which will help us analyze and explain mechanisms underlying adult behavior, learning, memory and social interactions. Indeed, it may even be possible one day to test out drugs or other therapeutic treatments on simulated animals before investing the time and resources on actual animal testing.