Session C14: Physics of Social Interactions II

Live

Bird flocks may range in size from only a few birds to hundreds or thousands. Characterizing the ways in which the structure and properties of flocks vary with the size is critical for understanding their dynamics and function. We will describe the structure and dynamics of more than 100 transit flocks of jackdaws, a highly social corvid species, measured in the field. We find that some properties, such as the bird number density, vary with size but saturate quickly. Other properties, such as the typical speeds, appear to be independent of flock size. We interpret our results in the context of the (topological) interaction range in flocks and compare them with similar measurements in insect swarms.   

Live

Two-body fighting behavior occurs throughout the animal kingdom to settle dominance disputes. Qualitatively, fights appear as a sequence of repeated, stereotyped dynamics that lead, ultimately, to a winner. Quantitatively however, little is known about the shorter-time behaviors that compose a fight and the longer-time dynamics of the dominance decision. We study fighting behavior in pairs of male zebrafish imaged at high spatiotemporal resolution in 3D. We leverage advances in convolutional neural networks to track multiple bodypoints of both fish while maintaining organism identity. We use these bodypoints to uncover key features of contest behavior; the fight onset is apparent through static and dynamic displays, followed by escalating aggression, culminating in distinct posture characteristics indicative of submission and dominance. Our approach offers a new system for the quantitative analysis of strongly-coupled social dynamics which can broadly inform theoretical models of mutual assessment.   

Live

Collective dynamics in animal groups has been studied in recent years intensively. Recent works have suggested that such multi-agent systems should operate in a special parameter region, close to critical points. This is relevant because critical systems exhibit unique properties like maximal responsiveness to external stimuli and optimal propagation of information within the group. However, empirical support for critical dynamics in animal groups is still very limited. We study high-density giant schools of fish (sulphur mollies; up to 3000 fish/m2) which in their natural habitat (sulphuric ponds/streams) are mostly confined to the surface. These fish exhibit a collective wave-like diving response to potential threats. We measure and quantify the "resting state" of the collective diving activity driven by environmental fluctuations, strongly resembling noise-driven excitable systems. Our quantitative analysis of experimental data suggests that the system operates close to criticality. By a systematic comparison of the system dynamics with a generic mathematical model, we conclude that this natural system operates indeed in a special parameter region close to a critical point. We explore potential functional benefits with respect to collective predator evasion.   

Live

Collective decision making is a key feature during natural motion of animal groups. A model of collective decision making regarding direction of travel was introduced as an extension of the Ising model where the spin-spin interaction is interpreted as a social force. We study trajectories in the mean field approximation of this model for two and three targets. For two targets there are two types of possible directions of movement: A compromise movement between the two targets when the angle between them is small or a movement towards one of the targets when the angle is large. At a critical angle there is a phase transition that manifests itself in a bifurcation point.
For three targets we find that the decision is broken down to a series of binary decisions and moreover, for a wide class of geometries we find an infinite number of bifurcations. In particular, as a result of the geometry of this series of bifurcations, we find that the mechanism of reaching the central target (out of three) is very different than the one which is used to reach the external ones. In addition, the susceptibility along the trajectories diverges at bifurcation points as expected at phase transitions. We support the mean field results by comparison to simulations.   

Live

Affiliative vocalizations are used by humans and other animals to facilitate social cohesion and help separated individuals reunite. In both the laboratory and field studies, changes in acoustic features of affiliative vocalizations exhibit a consistent relationship with physical distance from conspecifics: longer distances induce louder calls. Thus, vocalizations are likely to be driven by low-dimensional latent states associated with distance. Here, we propose a dynamical model for the use of affiliative vocalizations as a means of active sensing to guide navigation. In this model, vocalization and locomotion are inversely driven by the uncertainty of the locations of communicating conspecifics, and the dynamics of actions are regulated by energy constraints. Using data from captive animals, we show that the context-dependent vocalizations are a natural consequence of such an active sensing system. Optimal tuning for the probability of vocalization exists as measured by the rate of spatial convergence. This study predicts the potential mechanisms for the coordination between vocal production and the navigation system.   

Live

The sensory threshold is one of the most fundamental and well studied computational primitives organisms perform, both as a standalone computation, and as a component of more complex tasks. In social organisms such as bee swarms and ant colonies, which perform computational tasks at the group level, the collective sensory threshold is an emergent property that depends on the responses of individuals in the group and on the interactions between them. Here we study this emergence in the clonal raider ant (Ooceraea biroi), a model system that provides convenient and precise control over the properties of the colony. We show that an ant colony indeed responds collectively to step changes in temperature, and that this response is characterised by a threshold. We further show that this threshold is sensitive to the size of the colony, implying that interactions play an important role. We then replicate the observed size dependency in a mathematical model, and show that it entails a change resisting interaction between the ants. Finally, we discuss how heterogeneity and variability between individuals in the group affect this emergence both in the model and in the experiment.   

Live

Classical models of collective behavior often take a “bird’s-eye perspective,” assuming that individuals have access to social information not directly encoded in their sensory input. Despite the explanatory success of those models, it is now thought that a better understanding needs to incorporate the perception of the individual, i.e., how internal and external information are acquired and processed. In particular, vision has appeared to be a central feature to gather external information and influence the collective organization of the group. Here, we show that a simple vision-based model of collective behavior is sufficient to generate organized collective movements in the absence of any spatial representation. Our work suggests a different approach for the development of purely vision-based autonomous swarm robotic systems and formulates a mathematical framework for exploration of perception-based interactions and how they differ from physical ones.   

Live

How savvy we are in responding to social cues is often determined by prior social experiences. But what kind of experiences shape our social cognition the most? Using a fish species with unique social attributes, we test two competing theories for the development of social cognition (complex vs costly social interactions). Our research enables us to identify the social factors that contribute to the development of adult behavior, cognitive abilities, and the forces that shape the circuitry of the vertebrate brain.   

Live

Understanding social behavior requires tracking and quantifying animals’ movements as they engage in social interactions. Recently, there has been notable progress in developing deep learning algorithms to track multiple animals in social paradigms. Although these methods perform well when the animals are some distance apart or have brief close contact with each other, they exhibit poor tracking accuracy when the animals spend more time with each other, performing behaviors such as huddling, mutual-grooming, and mating. To improve the tracking of animals within social contexts, where the animals are in close proximity for long periods of time, we implemented a pipeline that combines multiple deep-learning-based tracking methods to obtain detailed and high-accuracy postural trajectories of multiple animals. Tested on a data set of prairie voles - a model organism for the study of social interactions - our pipeline robustly maintains animal identities and greatly increases the accuracy of the posture tracking over applying convolutional neural network methods by themselves. With this improved tracking accuracy, we are able to isolate behaviors that are key for understanding the dynamics of social interactions.   

Live

We introduce a mean-field term to an evolutionary spatial game model.
Namely, we consider the game of Nowak and May, based on the Prisoner's
dilemma, and augment the game rules by a mean-field term. This way,
the agent operates based on local information from its neighbors and
non-local information via the mean-field coupling.
We simulate the model and construct the steady-state phase diagram,
which shows significant differences due to the mean-field term.   

Live

The conventional view is that high temperatures cause microorganisms to replicate slowly or die, both autonomously. We show that budding yeasts, despite being single-celled organisms, collectively combat rising temperatures [1]. By cooperating, yeasts help each other and their future generations to replicate and avoid population extinction at high temperatures. As a consequence, even at the same temperature, a yeast population can exponentially grow, never grow or grow after unpredictable durations (hours to days) of stasis, depending on its population density. We measured a phase diagram which summarizes, as a function of both the temperature and the starting population density, when yeasts can replicate and when they cannot. These features arise from yeasts secreting and extracellularly accumulating a heat-damage preventing antioxidant (glutathione). We show that the secretion of glutathione, which eliminates harmful extracellular chemicals, is both necessary and sufficient for yeasts to survive high temperatures. Our study demonstrates how organisms can cooperatively define and extend the boundaries of life-permitting temperatures.

[1] Laman Trip, D.S., Youk, H. Yeasts collectively extend the limits of habitable temperatures by secreting glutathione. Nat Microbiol (2020).