Bulletin of the American Physical Society
APS March Meeting 2014
Volume 59, Number 1
Monday–Friday, March 3–7, 2014; Denver, Colorado
Session W10: Focus Session: Physics of Behavior II |
Hide Abstracts |
Sponsoring Units: DBIO Chair: Joshua Shaevitz, Princeton University Room: 201 |
Thursday, March 6, 2014 2:30PM - 3:06PM |
W10.00001: Wild swarms of midges linger at the edge of an ordering phase transition Invited Speaker: Irene Giardina The most notable hallmark of collective behavior in biological systems is the emergence of order: individuals synchronize their state, giving the stunning impression that the group behaves as one. Birds flocks, fish schools and mammal herds are just a few common examples of polarized animal groups. Mating swarms of mosquitoes and midges, on the other hand, do not display global order and it is therefore unclear whether swarms are a true instance of collective behavior or a mere epiphenomenon of the independent response of each insect to an environmental stimulus. The crucial task for a group, however, is not simply to achieve an ordered state, but to respond collectively to the environmental stimuli. For this to happen, correlation must be large, namely individuals must be able to influence each other's behavioral changes on a group scale. In this work, we experimentally study wild swarms of midges and find that, despite the lack of collective order, swarms display strong correlation, comparable to that found in highly ordered groups of vertebrates. Correlation is orders of magnitude larger in natural swarms than in random systems, indicating the existence of large clusters of insects responding together. We also find that the total amount of correlation, i.e. the susceptibility, increases sharply with the swarm density, a distinctive mark of an incipient ordering phase transition. Swarms, however, live at the near-critical edge of this transition, never plunging in the ordered phase, suggesting that their size and density are tuned to maximize collective response. [Preview Abstract] |
Thursday, March 6, 2014 3:06PM - 3:18PM |
W10.00002: Self Organized Sorting in Swarms Katherine Copenhagen, David Quint, Ajay Gopinathan Swarming behavior extends across multiple length scales in biology ranging from bacteria to whales. Natural swarms are affected by erratic, or dissenting behavior by individuals within the swarm who may display different types of behaviors than the rest of the swarm. This research investigates the introduction of heterogenous behavior amongst individuals within a swarm and their impact on swarm formation and robustness. We model swarms with a finite number of agents utilizing a velocity alignment interaction and a Lennard-Jones potential, which provides both cohesive and repulsive interactions between neighboring agents. Depending on the parameters governing the swarming interactions and the level of heterogeneity in behavior introduced, we found a variety of collective behavior including sharp transitions from swarming to non-swarming regimes and self organized sorting of individuals based on their types of behavior. Our research sheds light on the varied responses of swarms to internal dissent and suggests optimal strategies to tolerate errant individuals. [Preview Abstract] |
Thursday, March 6, 2014 3:18PM - 3:30PM |
W10.00003: Congestion and communication in confined ant traffic Nick Gravish, Gregory Gold, Andrew Zangwill, Michael A.D. Goodisman, Daniel I. Goldman Many social animals move and communicate within confined spaces. In subterranean fire ants {\em Solenopsis invicta}, mobility within crowded nest tunnels is important for resource and information transport. Within confined tunnels, communication and traffic flow are at odds: trafficking ants communicate through tactile interactions while stopped, yet ants that stop to communicate impose physical obstacles on the traffic. We monitor the bi-directional flow of fire ant workers in laboratory tunnels of varied diameter $D$. The persistence time of communicating ant aggregations, $\tau$, increases approximately linearly with the number of participating ants, $n$. The sensitivity of traffic flow increases as $D$ decreases and diverges at a minimum diameter, $D_c$. A cellular automata model incorporating minimal traffic features---excluded volume and communication duration---reproduces features of the experiment. From the model we identify a competition between information transfer and the need to maintain jam-free traffic flow. We show that by balancing information transfer and traffic flow demands, an optimum group strategy exists which maximizes information throughput. [Preview Abstract] |
Thursday, March 6, 2014 3:30PM - 3:42PM |
W10.00004: Extracting distinct behaviors from laboratory insect swarms James Puckett, Nicholas Ouellette Throughout nature, self-organized collective motion in animal groups produces rich and complex behaviors. Many modeling approaches have been proposed from continuum to discrete agent based models which are capable of emulating the behavior observed in flocks and swarms. Most models assume uniformity in the way individuals interact and discard differences between individuals and changes of behavior with time. While in many cases individual differences may average out in large groups of animals, this is not likely the case for small groups. By measuring trajectories and kinematics of individual Chironomids in laboratory mating swarms, we assess the dynamics of individual behavior and discuss the impact of our results on current models. [Preview Abstract] |
Thursday, March 6, 2014 3:42PM - 3:54PM |
W10.00005: Mapping the structure of animal behavior Gordon Berman, Daniel Choi, William Bialek, Joshua Shaevitz Most animals possess the ability to actuate a vast diversity of movements, ostensibly constrained only by morphology and physics. In practice, however, a frequent assumption in behavioral science is that most of an animal's activities can be described in terms of a small set of stereotyped motifs. Here we introduce a method for mapping the behavioral space of organisms, relying only upon the underlying structure of postural movement data to organize and classify behaviors. Applying our method to movies of size closely-related species of freely-behaving fruit flies, we find a wide variety of non-stereotyped and stereo-typed behaviors, spanning a wide range of time scales. We observe subtle behavioral differences between these species, identifying the some of the effects of phylogenic history on behavior. Moreover, we find that the transitions between the observed behaviors display a hierarchical syntax, with similar behaviors likely to transition between each other, but with a long time scale of memory. These results suggest potential mechanisms for the evolution of behavior and for the neural control of movements. [Preview Abstract] |
Thursday, March 6, 2014 3:54PM - 4:06PM |
W10.00006: Predator pursuit strategies: how do falcons and hawks chase prey? Suzanne Amador Kane, Marjon Zamani, Andrew Fulton, Lee Rosenthal This study reports on experiments on falcons, goshawks and red-tailed hawks wearing miniature videocameras mounted on their backs or heads while pursuing flying or ground-based prey. Videos of hunts recorded by the raptors were analyzed to determine apparent prey positions on their visual fields during pursuits. These video data then were interpreted using computer simulations of pursuit steering laws observed in insects and mammals. A comparison of the empirical and modeling data indicates that falcons use cues due to the apparent motion of prey on the falcon's visual field to track and capture flying prey via a form of motion camouflage. The falcons also were found to maintain their prey's image at visual angles consistent with using their shallow fovea. Results for goshawks and red-tailed hawks were analyzed for a comparative study of how pursuits of ground-based prey by accipeters and buteos differ from those used by falcons chasing flying prey. These results should prove relevant for understanding the coevolution of pursuit and evasion, as well as the development of computer models of predation on flocks,and the integration of sensory and locomotion systems in biomimetic robots. [Preview Abstract] |
Thursday, March 6, 2014 4:06PM - 4:18PM |
W10.00007: Physics of \textit{C. elegans }search Maria Panlilio, Frederic Bartumeus, William Ryu Movement is a fundamental feature of life. Organisms must search for prey, avoid predators, or explore new habitats. Using methods from statistical physics we seek to elucidate the behavioral strategies governing \textit{C. elegans} searches and their effects on both ecological and evolutionary timescales. Here we ask: how does the search strategy change under starvation? Animal movement studies are often hindered by difficult observations over large spatiotemporal scales, unknown environmental conditions, and complex behavioral descriptions. \textit{C. elegans} is a powerful model system for overcoming such challenges. Machine vision technologies capture high-resolution images of individuals crawling through a large isotropic environment. Trajectories are reconstructed from the images and behavioral reorientation events are automatically flagged. We find that short-term directional persistence initially increases with time away from food. We also quantify local and global spatial searching scales, which are modulated at least in part by the dynamics of one distinct behavior. Since other reorientation types are known to be suppressed under starvation, we propose that the long-term behavioral strategy acts as a compensatory mechanism to prevent both under and oversampling of the environment. [Preview Abstract] |
Thursday, March 6, 2014 4:18PM - 4:30PM |
W10.00008: The dynamics of the thermal memory of \textit{C.~elegans} William Ryu, Konstantine Palanski, Frederic Bartumeus, Ilya Nemenman \textit{C.~elegans} has the capacity to learn associatively. For example, \textit{C.~elegans} associates temperature with food and performs thermotaxis towards this temperature when placed on a spatial thermal gradient. However, very little is understood how \textit{C.~elegans} acquires this thermal memory. We have developed a novel droplet-based microfluidic assay to measure the dynamics of the thermal memory of \textit{C.~elegans}. Individual animals are placed in an array of microdroplets on a slide, and a linear temperature gradient of 0.5 deg/cm is applied to the array. By measuring the swimming motions of \textit{C.~elegans} in the droplets, we show that they can perform thermotaxis. By calculating an index of this taxis behavior over time, we quantify the worm's thermal memory and measure its dynamics when the animals are exposed to different conditions of feeding and starvation. Over a time scale of hours, we find that the thermal preference of wild-type worms decays and will actually become inverted and that mutations in the insulin signaling pathway perturb the dynamics. This biphasic conditional association can be explained with a reinforcement learning model with independent reinforcement and avoidance pathways with distinct time scales. [Preview Abstract] |
Thursday, March 6, 2014 4:30PM - 4:42PM |
W10.00009: Swimming Eigenworms Frank Van Bussel, Zeina Khan, Mizanur Rahman, Siva Vanapalli, Jerzy Blawzdziewicz The nematode C. Elegans is a much studied organism, with a fully mapped genome, cell structure, and nervous system; however, aspects of its behavior have yet to be elucidated, particularly with respect to motility under various conditions. Recently the ``Eigenworm'' technique has emerged as a promising avenue of exploration: via principle component analysis it has been shown that the state space of a healthy crawling worm is low dimensional, in that its shape can be well described by a linear combination of just four eigenmodes. So far, use of this methodology with swimming worms has been somewhat tentative, though medical research such as drug screening is commonly done with nematodes in fluid environments e.g. well plates. Here we give initial results for healthy worms swimming in liquids of varying viscosity. The main result is that at the low viscosities (M9 buffer solution) the state space is even lower dimensional than that for the crawling worm, with only two significant eigenmodes; and that as viscosity increases so does the number of modes needed for an adequate shape description. As well, the shapes of the eigenmodes undergo significant transitions across the range of viscosities looked at. [Preview Abstract] |
Thursday, March 6, 2014 4:42PM - 4:54PM |
W10.00010: A quantifiably complete repertoire of C. elegans locomotion Andre Brown, Roland Schwarz, Robyn Branicky, William Schafer Visible phenotypes have played a critical role in understanding the molecular basis of behaviour in model organisms. However, most current descriptions of behaviour are based on manually identified events or a limited set of quantitative parameters. Here we report an extension of the concept of behavioural motifs to exhaustively catalogue C. elegans locomotion and derive a repertoire that is quantifiably complete. A repertoire learned for spontaneous behaviour in wild-type worms can be used to fit data from mutants or worms in different environmental conditions and provides a sensitive measure of phenotypic similarity. Repertoire comparison can also be used to assess inter-individual variation and the compositionality of behaviour, that is, the extent to which behavioural adaptation involves the creation of novel repertoire elements or the reuse of existing elements in novel sequences. Repertoire derivation is general, so that given a representation of posture, our approach will apply to other organisms. [Preview Abstract] |
Thursday, March 6, 2014 4:54PM - 5:06PM |
W10.00011: A Quantitative Model of Motility Reveals Low-Dimensional Variation in Exploratory Behavior Across Multiple Nematode Species Stephen Helms, Leon Avery, Greg Stephens, Tom Shimizu Animal behavior emerges from many layers of biological organization---from molecular signaling pathways and neuronal networks to mechanical outputs of muscles. In principle, the large number of interconnected variables at each of these layers could imply dynamics that are complex and hard to control or even tinker with. Yet, for organisms to survive in a competitive, ever-changing environment, behavior must readily adapt. We applied quantitative modeling to identify important aspects of behavior in chromadorean nematodes ranging from the lab strain \textit{C. elegans} N2 to wild strains and distant species. We revealed subtle yet important features such as speed control and heavy-tailed directional changes. We found that the parameters describing this behavioral model varied among individuals and across species in a correlated way that is consistent with a trade-off between exploratory and exploitative behavior. [Preview Abstract] |
Thursday, March 6, 2014 5:06PM - 5:18PM |
W10.00012: Insects traversing grass-like vertical compliant beams Chen Li, Ronald Fearing, Robert Full Small running animals encounter many challenging terrains. These terrains can be filled with 3D, multi-component obstacles. Here, we study cockroaches (\textit{Blaberus discoidalis}) moving through grass-like vertical compliant beams during escape. We created an apparatus to control and vary geometric parameters and mechanical properties of model grass including height, width, thickness, lateral and fore-aft spacings, angle, number of layers, stiffness, and damping. We observed a suite of novel locomotor behaviors not previously described on simpler 2D ground. When model grass height was \textgreater 2 $\times$ body length and lateral spacing was \textless 0.5 $\times$ body width, the animal primarily (probability $P =$ 50{\%}) rolled its body onto its side to rapidly (time $t =$ 2.1 s) maneuver through the gaps between model grass. We developed a simple energy minimization model, and found that body roll reduces the energy barriers that the animal must overcome during traversal. We hypothesized that the animal's ellipsoidal body shape facilitated traversal. To test our hypothesis, we modified body shape by adding either a rectangular or an oval plate onto its dorsal surface, and found that $P$ dropped by an order of magnitude and $t$ more than doubled. Upon removal of either plate, both $P$ and $t$ recovered. Locomotor kinematics and geometry effectively coupled to terrain properties enables negotiation of 3D, multi-component obstacles, and provides inspiration for small robots to navigate such terrain with minimal sensing and control. [Preview Abstract] |
Thursday, March 6, 2014 5:18PM - 5:30PM |
W10.00013: ABSTRACT WITHDRAWN |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700