Session P14: Animal Behavior

Live

Plant shoots and roots capitalize on their slender structures to successfully negotiate unstructured environments while employing a combination of two classes of growth-driven movements: tropic responses, growing toward or away from an external stimulus, and inherent nastic movements, such as periodic circumnutations, which promote exploration.
During growth, organs often encounter various mechanical obstacles, such as soil, stones, or other organisms. Forces exerted by the organs on such obstacles, originating from growth processes, subsequently induce mechanical stresses on the organs themselves, yielding a rich and complex dynamics.
In order to emulate these dynamics in a 3D environment, we recently developed a general model for the growth of rod-like organs, adopting the Cosserat rod model. The dynamics are based on the combination of active growth-driven movements and passive elastic interactions with external objects or fields.
This model, together with its robust numerical framework, provides a powerful tool to deepen our understanding of how rod-like sensory-growth systems, such as plants, fungi and neurons, interact with their environments through a combination of both active and passive responses.   

Live

Twining plants do not rely on self-supporting structural integrity to uphold their weight while growing away from gravity and toward a light source. Instead, twining plants rely on external stable structures to hold their weight by twining around them.
An important phase in successfully attaching to such a supporting structure depends on the plant choosing a suitable support.
The plant will move in a periodic circular motion called circumnutation, attempting to collide with potential supports. If a collision occurs, the plant either twines around it or not. Our study suggests that this decision is determined by an integration of the force applied on the plant during contact with the support, a force that is driven by the circumnutational movement.
When the accumulated force does not pass a critical level, the stem will tend to 'slip' off of the support and continue the circumnutation 'search' behavior.
Such behavior acts as a filter for what structures are stable enough to support the plant and is a major evolutionary advantage over similar twining plants with no or less effective filters, resulting in a higher probability of collapse whilst growing on non-stable supports.   

Live

Marine microorganisms control the global biogeochemistry of the oceans through interactions between individual cells and between cells and particles of organic matter. Prominent examples include marine snow formation by phytoplankton or bacterial degradation of marine snow responsible for carbon export from the upper ocean in the biological pump. Here, we first describe the impact of elongation on marine snow formation. We derive the collision kernels between identical and dissimilar rods settling in a quiescent fluid and show that marine snow formation by elongated phytoplakton can proceed efficiently even under quiescent conditions and that the resulting coagulation dynamics can lead to periodic bursts in the concentration of marine snow particles. Later, we describe the encounters between bacteria and sinking particles of organic matter. There, we find that the coupling between cell shape and fluid shear has a considerable effect on the encounter rate and encounter location through the mechanisms of hydrodynamic focusing and screening. Overall, our results demonstrate that elongation and fluid shear must be taken into account to accurately predict encounter rates at the microscale, which govern the large carbon flux in the ocean’s biological pump.   

Live

Behavior exhibits multiple timescales: from fast control by neural activity to slower variation due to neuromodulation or even aging. Can we extract longer-time scale dynamics, driven by such hidden processes, from short-time behavior? We use a recently developed transfer operator approach, akin to a Fokker-Planck description, to perform a top-down coarse-graining of C. elegans posture dynamics: from long timescale ``roaming’' and ``dwelling’' states (minutes), to a ``run’' and ``pirouette’' description (seconds), down to fine scale movements (milliseconds). We use the transfer operator spectrum to illustrate how genetic and environmental perturbations can impair the generation of long-lived behavioral patterns. We study the effects of starvation and of mutation in a neuropeptide receptor gene, and show that although both perturbations lead to the absence of a ``dwelling’' state, there are important fine scale differences that hint at distinct neuromodulatory mechanisms.   

Live

Many elongate terrestrial animals (e.g., snakes, nematodes) generate and propagate waves of body curvature to traverse highly dissipative environments. Particularly, the mm-long nematode worm C. elegans must execute complex behaviors to navigate its natural environment (e.g., rotting fruit). In laboratory studies of locomotion in homogeneous environments, worms use a time-dependent omega-like shape to achieve high in-place rotation. To discover if such “omega turns” are effective in heterogeneous environments, we conducted laboratory experiments in fluid filled PDMS multi-post arrays. Worms effectively performed omega turns in these environments by wrapping their bodies around obstacles. Kinematics and turning performance was comparable to those on the surface of homogeneous agar or buffer, where worms achieved high rotation while minimizing swept area. Preliminary experiments with mechanosensing defective mutants (mec-4) suggest that worms do not need to sense their surroundings to perform omega turns in complex environments. Our results suggest that omega turns are a robust strategy to turn and maneuver in diverse environments.   

Live

Animals traverse large obstacles in complex 3-D terrain by using physical interaction to make diverse locomotor transitions. Understanding how such transitions emerge is challenging, especially at large spatiotemporal scales. Existing terrain testbeds limit observations to small scales (~10 strides or body lengths). Here, we used a novel terrain treadmill to study cockroach traversal of large pillar obstacles over large spatiotemporal scales. Our treadmill consists of a transparent outer sphere and a concentric inner sphere mounted with large pillars and is controlled to keep an untethered animal moving on top. Animals moved on the treadmill freely and continuously, even up to ~500 seconds (250× increase), covering ~50 m (20× increase), while changing direction freely. We reconstructed its 3-D motion to measure the body/antenna contact with obstacles. For sparse pillars, body-pillar interaction was minimal due to wider gaps and antennae use to detect and avoid obstacles. For cluttered pillars, animals interacted and rolled its body into gap and traversed. Behavior such as antenna sweeping, body turning, and pillar climbing were also observed. Our experimental system can measure locomotor behaviors with a high spatial resolution over a long duration and distance.   

Live

Rather than wait passively for signals to be detected by their sensors, animals actively move in order to gather information from their environment. Furthermore, when sensing is performed by means of rhythmic movements, reafferent sensory streams are also rhythmic, which is advantageous for sensory processing. Here, we report on and characterize an hitherto unknown behavioural state of pulse-type weakly electric fish during which electrosensory acquisition becomes rhythmic and is coupled to low-frequency movement. The oscillatory nature of this sensory sampling strategy is in stark contrast to that exhibited during other behaviours.

To characterize the oscillatory dynamics of these rhythmic behavioural states, we identify and extract them from long-term observations of freely-moving fish. By applying t-distributed Stochastic Neighbor Embedding on the wavelet spectra of the inter-pulse interval time series, we achieve a successfully classification of behavioural states. Two dominant frequency bands appear during oscillatory behaviour: around 0.5 and 1 Hz. By analyzing the associated bandpass filtered signals, we find that they exhibit hallmark features of synchronization, including phase slips and limit cycles of the phase dynamics on the torus.   

Live

As a basis for understanding the physics of behaviour in fruitfly larvae, we here develop an effective theory for the animals' motion. We define a set of fields which quantify deformations of the larva's anteroposterior axis, then search the space of possible theories that could govern the fields' low-energy physics. Guided by symmetry and stability requirements, we arrive at a unique free-field theory with few free parameters. Surprisingly, we can explain many features of larval behaviour by applying equilibrium statistical physics to this model. Our theory closely predicts the animals' postural modes (eigenmaggots), as well as distributions and trajectories in the mode space, across several behaviours. Our results show that real animal behaviour can be understood using relatively simple effective physics.   

Live

Many animals use aerial righting to mitigate the risks associated with falling (e.g., predation, starvation and desication), offering inspiration for landing strategies open to biomimetic robots. Spotted lanternflies (Lycorma delicatula) (SLFs) are invasive insect pests that often fall from host plants in response to predators or abiotic factors (e.g., wind). We used high-speed video to study whether immature SLFs (nymphs) land upright more often than expected by chance, and, if so, whether they do so via active or passive mechanisms. SLF nymphs were found to adopt a stereotypic falling posture similar to that used by insects, spiders, geckos, frogs and skydivers. Live nymphs landed upright more often when they released from surfaces than when dropped artificially. The fraction landing upright was significantly higher for live SLFs than for similarly-posed dead specimens, and did not depend significantly either on orientation during release or at first impact. Significantly more SLFs reoriented to upright from other orientations on first impact using a combination of bouncing and adhesion. Videos of nymphs landing on leaves confirmed that these insects use multiple tactics to land upright on host plants.   

Live

How insects navigate complex odor plumes, where the location and timing of odor packets are uncertain, remains unclear. Here, we imaged complex odor plumes simultaneous with freely-walking flies, allowing us to quantify how behavior is shaped by discrete encounters with odor packets. Combining measurements, dynamical models, and statistical inference we found that navigation was stochastic, and did not rely on the continuous modulation of speed or orientation. Instead, flies turned stochastically with stereotyped saccades, whose direction was biased upwind by the timing of prior odor encounters, while the magnitude and rate of saccades remained constant. Further, the timing of encounters was used to modulate the transition rates between walks and stops. Thus, while in more regular environments, flies are known to continuously modulate speed and orientation, our results show that in less predictable environments walking flies instead navigate with random walks biased by the timing of encounters.   

Live

Movies of freely moving animals contain rich information about an animal’s behavior. However, unbiased quantification of the behavior remains challenging. In our work, we investigate two different strategies for behavioral classification used in previous studies: First, we perform wavelet transformations of the distances between body parts, followed by unsupervised clustering of the data in frequency space to identify distinct behaviors. As a second approach, we identify change points in the temporal dynamics, and locally fit the data to autoregressive models, which also can be clustered to assign behaviors. By comparing these two different approaches, we can identify their strengths and limitations and create a more holistic picture of animal behavior. We use our findings to explore the behavioral repertoire of mice and identify differences between the open field behavior of mice with and without a cerebellar perturbation.   

Live

Snakes can traverse complex 3-D terrain by deforming their elongate body. Most previous studies on terrestrial snake locomotion focused on flat surfaces and planar gaits except for sidewinding with small body lifting. How snakes deform their body in 3-D to traverse complex 3-D terrain with large obstacles was less understood. We recently discovered that snakes use vertical undulation to traverse a horizontal ladder (Jurestovsky et al., in prep) and use lateral oscillation and vertical bending to traverse a large step (Gart, Mitchel, Li, 2020, JEB). Here, we hypothesized that vertical body bending is important for traversal of more complex rubble-like terrain and tested it using the generalist corn snake P. guttatus (N = 2 animals, n = 23 trials). The animal moved in a virtual tube, with little slip was observed where the body contacted the terrain despite low surface friction. The animal tended to move through lower valleys and mainly used vertical body bending to push against horizontal ridges. This was only occasionally complemented by lateral pushing against vertical walls. These observations supported our hypothesis. We plan to further study distributed contact forces using a sensorized snake robot as a physical model (see talk by Ramesh et al. in Robophysics sessions).   

Live

Interactions between limbless animals and their natural environments, which are essential for movement, are mediated solely through skin contact. We used atomic force microscopy to investigate how surface textures on shed skins vary across snakes and environments. While most snakes have microscopic spikes oriented from head to tail, a few distantly related snake species have convergently lost these features in favor of a more isotropic structure. We hypothesize that these microstructures affect the frictional interaction with the substrate and we use resistive force theory to model the effects of frictional anisotropy on snake locomotion. For lateral undulation, we predict that an anisotropic frictional interaction in movement along the body is favored over transverse movement, improving performance (measured in distance traveled per cycle), and that larger anisotropies produce larger displacements. In sidewinding locomotion, however, we predict the opposite trend: decreased frictional anisotropy improves performance. These predictions are consistent with our AFM measurements of small-scale features on snake skins and suggest a functional benefit for the convergent loss of structure shared by sidewinding vipers.   

Live

Unconstrained animal behavior is composed of rich dynamics that span multiple timescales. How these long timescale dynamics emerge from collections of short postural movements, however, remains an open question. Using markerless tracking, we measured the postural dynamics of freely behaving rats as they move throughout an enriched arena over 3-hour sessions and train a multi-layer recurrent neural network to predict these dynamics. We use activation values from each layer to create multiple two-dimensional density maps of behavioral states. We find that in these density maps, local maxima correspond to different stereotyped behaviors that the animal performs, including locomotion at different speeds, head movements, grooming, rearing and rest. Density maps from different layers exhibit dynamics at different timescales. Lastly, we extracted the hierarchical structure of rat behavior from these data, creating a generative model that recapitulates the observed movements. The results from this model allow us to make predictions about the internal neural and neuromodulatory states that regulate the rat’s behavior at multiple timescales.   

Live

The roundworm C. elegans learns from its experiences. Worms on a thermal gradient perform
thermotaxis to or away from the conditioned temperature, depending on its thermal history and the
starvation duration during the conditioning phase. To quantify this behavior, we developed a novel assay
that tracks single worms—each experiencing a spatiotemporal thermal gradient with thermal precision
±0.01C—in a small (2.8ul) droplets of buffer, arrayed on hydrophobic-printed microscope slides. CCD
cameras track individual worms simultaneously in many droplets for many hours and at high temporal resolution, each droplet set at 20C midpoint with a thermal gradient of 0.5C/cm. A worm’s thermal preference is
summarized as a thermotaxis index. Initially, worms reared at 25C and 15C exhibit thermophilic or
cryophilic tendencies, respectively, and starvation during conditioning or in the droplet reverses these
tendencies. This reversal indicates multiple dynamic processes for learning that operate on different time
scales. We build a predictive model with multiple time scales and utilize mutants to detangle the various
learning processes. The model predicts the behavior under various training conditions and genetic perturbations.