Evidence for a pervasive 'idling-mode' activity template in flying and pedestrian insects

Analyzing Sensor-Based Individual and Population Behavior Patterns via Inverse Reinforcement Learning

Digital markers of behavior can be continuously created, in everyday settings, using time series data collected by ambient sensors. The goal of this work was to perform individual- and population-level behavior analysis from such time series sensor data. In this paper, we introduce a novel algorithm—Resident Relative Entropy-Inverse Reinforcement Learning (RRE-IRL)—to perform an analysis of a single smart home resident or a group of residents, using inverse reinforcement learning. By employing this method, we learnt an individual’s behavioral routine preferences. We then analyzed daily routines for an individual and for eight smart home residents grouped by health diagnoses. We observed that the behavioral routine preferences changed over time. Specifically, the probability that the observed behavior was the same at the beginning of data collection as it was at the end (months later) was lower for residents experiencing cognitive decline than for cognitively healthy residents. When comparing aggregated behavior between groups of residents from the two diagnosis groups, the behavioral difference was even greater. Furthermore, the behavior preferences were used by a random forest classifier to predict a resident’s cognitive health diagnosis, with an accuracy of 0.84.

Quantifying flight aptitude variation in wild Anopheles gambiae in order to identify long-distance migrants

Background

In the West African Sahel, mosquito reproduction is halted during the 5–7 month-long dry season, due to the absence of surface waters required for larval development. However, recent studies have suggested that both Anopheles gambiae sensu stricto (s.s.) and Anopheles arabiensis repopulate this region via migration from distant locations where larval sites are perennial. Anopheles coluzzii engages in more regional migration, presumably within the Sahel, following shifting resources correlating with the ever-changing patterns of Sahelian rainfall. Understanding mosquito migration is key to controlling malaria—a disease that continues to claim more than 400,000 lives annually, especially those of African children. Using tethered flight data of wild mosquitoes, the distribution of flight parameters were evaluated as indicators of long-range migrants versus appetitive flyers, and the species specific seasonal differences and gonotrophic states compared between two flight activity modalities. Morphometrical differences were evaluated in the wings of mosquitoes exhibiting high flight activity (HFA) vs. low flight activity (LFA).

Methods

A novel tethered-flight assay was used to characterize flight in the three primary malaria vectors- An. arabiensis, An. coluzzii and An. gambiae s.s. The flights of tethered wild mosquitoes were audio-recorded from 21:00 h to 05:00 h in the following morning and three flight aptitude indices were examined: total flight duration, longest flight bout, and the number of flight bouts during the assay.

Results

The distributions of all flight indices were strongly skewed to the right, indicating that the population consisted of a majority of low-flight activity (LFA) mosquitoes and a minority of high-flight activity (HFA) mosquitoes. The median total flight was 586 s and the maximum value was 16,110 s (~ 4.5 h). In accordance with recent results, flight aptitude peaked in the wet season, and was higher in gravid females than in non-blood-fed females. Flight aptitude was also found to be higher in An. coluzzii compared to An. arabiensis, with intermediate values in An. gambiae s.s., but displaying no statistical difference. Evaluating differences in wing size and shape between LFA individuals and HFA ones, the wing size of HFA An. coluzzii was larger than that of LFAs during the wet season—its length was wider than predicted by allometry alone, indicating a change in wing shape. No statistically significant differences were found in the wing size/shape of An. gambiae s.s. or An. arabiensis.

Conclusions

The partial agreement between the tethered flight results and recent results based on aerial sampling of these species suggest a degree of discrimination between appetitive flyers and long-distance migrants although identifying HFAs as long-distance migrants is not recommended without further investigation.

Do bumblebees have signatures? Demonstrating the existence of a speed-curvature power law in Bombus terrestris locomotion patterns

We report the discovery that Bombus terrestris audax (Buff-tailed bumblebee) locomotor trajectories adhere to a speed-curvature power law relationship which has previously been found in humans, non-human primates and Drosophila larval trajectories. No previous study has reported such a finding in adult insect locomotion. We used behavioural tracking to study walking Bombus terrestris in an arena under different training environments. Trajectories analysed from this tracking show the speed-curvature power law holds robustly at the population level, displaying an exponent close to two-thirds. This exponent corroborates previous findings in human movement patterns, but differs from the three-quarter exponent reported for Drosophila larval locomotion. There are conflicting hypotheses for the principal origin of these speed-curvature laws, ranging from the role of central planning to kinematic and muscular skeletal constraints. Our findings substantiate the latter idea that dynamic power-law effects are robust, differing only through kinematic constraints due to locomotive method. Our research supports the notion that these laws are present in a greater range of species than previously thought, even in the bumblebee. Such power laws may provide optimal behavioural templates for organisms, delivering a potential analytical tool to study deviations from this template. Our results suggest that curvature and angular speed are constrained geometrically, and independently of the muscles and nerves of the performing body.

Role of the circadian clock in the statistics of locomotor activity in Drosophila

In many animals the circadian rhythm of locomotor activity is controlled by an endogenous circadian clock. Using custom made housing and video tracking software in order to obtain high spatial and temporal resolution, we studied the statistical properties of the locomotor activity of wild type and two clock mutants of Drosophila melanogaster. We show here that the distributions of activity and quiescence bouts for the clock mutants in light-dark conditions (LD) are very different from the distributions obtained when there are no external cues from the environment (DD). In the wild type these distributions are very similar, showing that the clock controls this aspect of behavior in both regimes (LD and DD). Furthermore, the distributions are very similar to those reported for Wistar rats. For the timing of events we also observe important differences, quantified by how the event rate distributions scale for increasing time windows. We find that for the wild type these distributions can be rescaled by the same function in DD as in LD. Interestingly, the same function has been shown to rescale the rate distributions in Wistar rats. On the other hand, for the clock mutants it is not possible to rescale the rate distributions, which might indicate that the extent of circadian control depends on the statistical properties of activity and quiescence.

Distinguishing between apparent and actual randomness: a preliminary examination with Australian ants

Abstract

The correlated random walk paradigm is the dominant conceptual framework for modeling animal movement patterns. Nonetheless, we do not know whether the randomness is apparent or actual. Apparent randomness could result from individuals reacting to environmental cues and their internal states in accordance with some set of behavioral rules. Here, we show how apparent randomness can result from one simple kind of algorithmic response to environmental cues. This results in an exponential step-length distribution in homogeneous environments and in generalized stretched exponential step-length distributions in more complex fractal environments. We find support for these predictions in the movement patterns of the Australian bull ant Myrmecia midas searching on natural surfaces and on artificial uniform and quasi-fractal surfaces. The bull ants spread their search significantly farther on the quasi-fractal surface than on the uniform surface, showing that search characteristics differed as a function of the substrate on which ants are searching. Further tentative support comes from a re-analysis of Australian desert ants Melophorus bagoti moving on smoothed-over sand and on a more strongly textured surface. Our findings call for more experimental studies on different surfaces to test the surprising predicted linkage between fractal dimension and the exponent in the step-length distribution.

Significance statement

Animal search patterns often appear to be irregular and erratic. This behavior is captured by random walk models. Despite their considerable successes, extrapolation and prediction beyond observations remain questionable because the true nature and interpretation of the randomness in these models have until now been elusive. Here, we show how apparent randomness can result from simple algorithmic responses to environmental cues. Distinctive predictions from our theory find support in analyses of the search patterns of two species of Australian ants.

A random acceleration model of individual animal movement allowing for diffusive, superdiffusive and superballistic regimes

Patterns of individual animal movement attracted considerable attention over the last two decades. In particular, question as to whether animal movement is predominantly diffusive or superdiffusive has been a focus of discussion and controversy. We consider this problem using a theory of stochastic motion based on the Langevin equation with non-Wiener stochastic forcing that originates in animal’s response to environmental noise. We show that diffusive and superdiffusive types of motion are inherent parts of the same general movement process that arises as interplay between the force exerted by animals (essentially, by animal’s muscles) and the environmental drag. The movement is superballistic with the mean square displacement growing with time as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\langle {x}^{2}(t)\rangle \sim {t}^{4}$$\end{document}〈x2(t)〉∼t4 at the beginning and eventually slowing down to the diffusive spread \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\langle {x}^{2}(t)\rangle \sim t$$\end{document}〈x2(t)〉∼t. We show that the duration of the superballistic and superdiffusive stages can be long depending on the properties of the environmental noise and the intensity of drag. Our findings demonstrate theoretically how the movement pattern that includes diffusive and superdiffusive/superballistic motion arises naturally as a result of the interplay between the dissipative properties of the environment and the animal’s biological traits such as the body mass, typical movement velocity and the typical duration of uninterrupted movement.

The evolutionary origins of Lévy walk foraging

We study through a reaction-diffusion algorithm the influence of landscape diversity on the efficiency of search dynamics. Remarkably, the identical optimal search strategy arises in a wide variety of environments, provided the target density is sparse and the searcher’s information is restricted to its close vicinity. Our results strongly impact the current debate on the emergentist vs. evolutionary origins of animal foraging. The inherent character of the optimal solution (i.e., independent on the landscape for the broad scenarios assumed here) suggests an interpretation favoring the evolutionary view, as originally implied by the Lévy flight foraging hypothesis. The latter states that, under conditions of scarcity of information and sparse resources, some organisms must have evolved to exploit optimal strategies characterized by heavy-tailed truncated power-law distributions of move lengths. These results strongly suggest that Lévy strategies—and hence the selection pressure for the relevant adaptations—are robust with respect to large changes in habitat. In contrast, the usual emergentist explanation seems not able to explain how very similar Lévy walks can emerge from all the distinct non-Lévy foraging strategies that are needed for the observed large variety of specific environments. We also report that deviations from Lévy can take place in plentiful ecosystems, where locomotion truncation is very frequent due to high encounter rates. So, in this case normal diffusion strategies—performing as effectively as the optimal one—can naturally emerge from Lévy. Our results constitute the strongest theoretical evidence to date supporting the evolutionary origins of experimentally observed Lévy walks.

How organisms improve the search for food, mates, etc., is a key factor to their survival. Mathematically, the best strategy to look for randomly distributed re-visitable resources—under scarce information and sparse conditions—results from Lévy distributions of move lengths (the probability of taking a step ℓ is proportional to 1/ℓ²). Today it is well established that many animal species in different habitats do perform Lévy foraging. This fact has raised a heated debate, viz., the emergent versus evolutionary hypotheses. For the former, a Lévy foraging is an emergent property, a consequence of searcher-environment interactions: certain landscapes induce Lévy patterns, but others not. In this view, the optimal strategy depends on the particular habitat. The evolutionary explanation, in contrast, is that Lévy foraging strategies are adaptations that evolved via natural selection. In this article, through simulations we exhaustively analyze the influence of distinct environments on the foraging efficiency. We find that the optimal procedure is the same in all situations, provided density is low and landscape information is scarce. So, the best search strategy is remarkably independent of details. These results constitute the strongest theoretical evidence to date supporting the evolutionary origins of experimentally observed Lévy walks.

The speed-curvature power law in Drosophila larval locomotion

We report the discovery that the locomotor trajectories of Drosophila larvae follow the power-law relationship between speed and curvature previously found in the movements of human and non-human primates. Using high-resolution behavioural tracking in controlled but naturalistic sensory environments, we tested the law in maggots tracing different trajectory types, from reaching-like movements to scribbles. For most but not all flies, we found that the law holds robustly, with an exponent close to three-quarters rather than to the usual two-thirds found in almost all human situations, suggesting dynamic effects adding on purely kinematic constraints. There are different hypotheses for the origin of the law in primates, one invoking cortical computations, another viscoelastic muscle properties coupled with central pattern generators. Our findings are consistent with the latter view and demonstrate that the law is possible in animals with nervous systems orders of magnitude simpler than in primates. Scaling laws might exist because natural selection favours processes that remain behaviourally efficient across a wide range of neural and body architectures in distantly related species.

Optimal search patterns in honeybee orientation flights are robust against emerging infectious diseases

Lévy flights are scale-free (fractal) search patterns found in a wide range of animals. They can be an advantageous strategy promoting high encounter rates with rare cues that may indicate prey items, mating partners or navigational landmarks. The robustness of this behavioural strategy to ubiquitous threats to animal performance, such as pathogens, remains poorly understood. Using honeybees radar-tracked during their orientation flights in a novel landscape, we assess for the first time how two emerging infectious diseases (Nosema sp. and the Varroa-associated Deformed wing virus (DWV)) affect bees' behavioural performance and search strategy. Nosema infection, unlike DWV, affected the spatial scale of orientation flights, causing significantly shorter and more compact flights. However, in stark contrast to disease-dependent temporal fractals, we find the same prevalence of optimal Lévy flight characteristics (μ ≈ 2) in both healthy and infected bees. We discuss the ecological and evolutionary implications of these surprising insights, arguing that Lévy search patterns are an emergent property of fundamental characteristics of neuronal and sensory components of the decision-making process, making them robust against diverse physiological effects of pathogen infection and possibly other stressors.

Lévy patterns in seabirds are multifaceted describing both spatial and temporal patterning

Background

The flight patterns of albatrosses and shearwaters have become a touchstone for much of Lévy flight research, spawning an extensive field of enquiry. There is now compelling evidence that the flight patterns of these seabirds would have been appreciated by Paul Lévy, the mathematician after whom Lévy flights are named. Here we show that Lévy patterns (here taken to mean spatial or temporal patterns characterized by distributions with power-law tails) are, in fact, multifaceted in shearwaters being evident in both spatial and temporal patterns of activity.

Results

We tested for Lévy patterns in the at-sea behaviours of two species of shearwater breeding in the North Atlantic Ocean (Calonectris borealis) and the Mediterranean sea (C. diomedea) during their incubating and chick-provisioning periods. We found that distributions of flight durations, on/in water durations and inter-dive time-intervals have power-law tails and so bear the hallmarks of Lévy patterns.

Conclusions

The occurrence of these statistical laws is remarkable given that bird behaviours are strongly shaped by an individual’s motivational state and by complex environmental interactions. Our observations could take Lévy patterns as models of animal behaviour to a new level by going beyond the characterisation of spatial movements to characterise how different behaviours are interwoven throughout daily animal life.

Signatures of chaos in animal search patterns

One key objective of the emerging discipline of movement ecology is to link animal movement patterns to underlying biological processes, including those operating at the neurobiological level. Nonetheless, little is known about the physiological basis of animal movement patterns, and the underlying search behaviour. Here we demonstrate the hallmarks of chaotic dynamics in the movement patterns of mud snails (Hydrobia ulvae) moving in controlled experimental conditions, observed in the temporal dynamics of turning behaviour. Chaotic temporal dynamics are known to occur in pacemaker neurons in molluscs, but there have been no studies reporting on whether chaotic properties are manifest in the movement patterns of molluscs. Our results suggest that complex search patterns, like the Lévy walks made by mud snails, can have their mechanistic origins in chaotic neuronal processes. This possibility calls for new research on the coupling between neurobiology and motor properties.