Session T03: Animal Behavior II

Theory, reimagined: lessons from the low-dimensional dynamics of animal behavior

TBD   

The use of first-axillary steering muscles in Drosophila roll control

Like balancing a meter stick on the tip of your finger, flapping flight is an inherently unstable dynamical system. In order to navigate changing wind conditions and evade determined fly swatters, fruit flies have evolved a stability reflex which can correct for mid-flight perturbations on the timescale of wingbeats (~30 ms). About the fly's most unstable degree of freedom, roll, these corrections begin within 5ms, one of the fastest reflexes in the animal kingdom. Previous investigations have shown that this reflex is well-modeled using a proportional integral controller. We now seek to elucidate the underlying mechanisms of neural control within the wing motor system which make this possible. Using a combination of mid-flight mechanical perturbations and neural manipulation via optogenetics, we investigated the role of the first axillary steering muscles during corrections for perturbations about the fly's body roll axis. We find that both of the first axillary steering muscles drive wing kinematic changes necessary for roll correction, and place this in context of the fly's roll controller.   

Elucidating the role of tergopleural muscles in fruit fly's pitch stability.

Flapping flight is an inherently unstable form of locomotion that requires flying insects to constantly make slight adjustments to their wing motion in order to maintain stability. Previous studies have shown that, in fruit flies, the reflexes used to counteract these instabilities are fast, robust, and well-described by a proportional-integral (PI) controller model. To elucidate the neuromuscular underpinnings of this flight control reflex, we perform behavioral assays in which freely flying flies are simultaneously subjected to external mechanical torques and acute, targeted neuronal manipulations. By analyzing the 3D wing and body kinematics that flies produce in response to these paired perturbations, we assay the role of individual cells in the flight control reflex. Here we apply this methodology to investigate the motor neurons innervating a subset of the wing's control muscles, the tergopleural (tp) muscles. We find that inhibiting the tp muscles leads to an asymmetry in the body pitch response that cannot be captured by a PI model that only utilizes the forward sweep angle. This prompts us to consider an expanded control theory framework that captures the interplay between multiple wing degrees of freedom to explain the asymmetry observed.   

Neuromechanics of Gait Adaptation in the Nematode C. elegans

The mm-scale nematode C. elegans adapts its undulatory gait to changing environmental resistance via a combination of distributed and centralized control mechanisms. We record body kinematics across environments (varying viscosity, and on agar surfaces) using bright field imaging and measure muscle activation patterns using ratiometric calcium imaging to investigate control strategies underlying gait adaptation. Waves of muscle activation display a constant spatial frequency (~0.8 times the body length) across environments while the resulting waves of undulating body curvature shift in response to increasing environmental resistance (from ~0.6 to ~1.5 body lengths). In contrast, undulation frequencies of muscle activation shift in parallel to body wave frequencies as drag increases. This leads to the existence of “neuromechanical phase lags” – phase shifts between muscle activation and body curvature -which depend on the environment. We compare measured phase lags to a resistive force theory model, which allows cross-taxa comparisons to previously studied systems, such as the cm-scale sandfish lizard, which locomotes through granular terrain.   

A minimal, biomimetic, add-on model for C. elegans swimming in a natural environment

The small (1 mm) nematode Caenorhabditis elegans (C. elegans) is widely used as a model organism; in particular its connectome has been completely mapped and its locomotion widely studied (c.f. http://www.wormbook.org/). In order to better understand C. elegans locomotion, we develop a minimal “add-on” model (the opposite of a knockout model) for the C. elegans central pattern generator (CPG) (c.f. Xu et al. 2018, Wen et al. 2012).  This model consists of a small network of simulated FitzHugh-Nagumo neurons, coupled by diffusion, whose topology is based on the Xu et al. connectome (Singhvi et al. 2021).  Here we show that the model admits traveling wave solutions corresponding to forward and backward locomotion, as well as phenomenological omega turns, and that these modes can be largely represented by the first three “eigenworms” of Stephens et al. (2008) and Ahamed et al. (2021). We also show that incorporating a simple Markov model (c.f. Roberts et al. 2016) for translating sensory inputs into dynamics of this simple model generates the observed random searching behavior (Calhoun et al. 2014).  Finally, we compare the locomotion of simulated worms to our (JM and HMH) lab’s experimental results (Magnes et al. 2020). (AS and CEL contributed equally to this work.)   

Spectral mode dynamics of C. elegans

The roundworm C. elegans exhibits diverse behavioral states and motor programs. Recent advancements in imaging platforms have enabled the collection of large amounts of data to record videos of C. elegans motion over long time periods. Constructing models for animal motor programs and connecting these models to behavioral states and neural patterns is a major current area of research. In this work, we present a general method for learning motor programs in a spectral representation. The method is robust to experimental noise and imposes necessary physical constraints and symmetries of the system. The learned models are sparse and physically interpretable. We utilize this learning approach in concert with clustering techniques to identify distinct motor patterns in C. elegans.   

Water surface swimming dynamics in centipedes

Centipedes use body and limb coordination to navigate diverse terrains. In terrestrial environments, the animals use body travelling wave and limb-stepping waves for locomotion. Less is known about how centipedes navigate fluid environments. Here, we challenged a primarily terrestrial centipede, L. foficatus (N=8, L=2.3±0.3 cm, 14 leg pairs), to locomote in water. On the water surface, centipedes locomoted via body waves propagated in the direction of motion. This was surprising since centipedes that use a direct limb-stepping pattern do not use body undulation in terrestrial environments. In contrast, amphibious centipedes swim using body waves that travel opposite to the direction of motion, folding their limbs towards their body (Yasui et al., Sci. Rep., 2019). When L. foficatus used body undulation on the fluid surface, the centipedes achieved speeds of 0.22±0.03 BL/cyc using body waves with maximum amplitude of 3.9±1.5 cm^-1 and 1.3±0.23 waves along their bodies. Without body undulation, the animal’s displacement was negligible. This suggests that surface swimming in this species is facilitated by body waves, not limb flexion. We posit these direct body waves enable the animal to swim by varying the animal’s drag anisotropy (ratio of local perpendicular to parallel forces).   

Walking on a membrane: how to measure ground contact forces for small animals

Measuring spatially distributed forces is crucial to study and understand animal locomotion. For animals like humans, piezo or capacitive pressure mats are readily available, but when it comes to lighter animals that weigh under 100g we lack similar sensors. Optical methods that use photoelastic materials have been developed previously but with limited success. We show a new method to measure spatially distributed in vivo loads that uses the deformation of a thin sheet to back-calculate the applied traction; reminiscent of traction force microscopy but with some notable differences. The substrate is a pre-stressed thin sheet that behaves like a membrane rather than a shell. When the pre-stress dominates the behavior, the deformed curvature is proportional to the normal load according to the Young-Laplace approximation, i.e. -T▽²w=q, with pre-tension T, normal deformation w, and force q to be estimated. In numerical studies, we show the viability of this approach and propose an experimental realization using a non-contact digital image correlation system. Finally, based on analytical models, we develop material selection criteria to satisfy constraints like spatial resolution, minimum load detection thresholds, and animal comfort.   

Simulation of snakes using vertical body bending to traverse terrain with large height variation

Snake moves across various terrains by bending its elongated body. Recent studies discovered that snakes can use vertical bending to traverse terrain of large height variation, such as horizontally oriented cylinders, a wedge (Jurestovsky, Usher, Astley, 2021, J. Exp. Biol.), and uneven terrain (Fu & Li, 2020, Roy. Soc. Open Sci.; Fu, Astley, Li, J. Exp. Biol., in review). Here, to understand how vertical bending generates propulsion, we developed a dynamic simulation of a snake traversing a wedge (height ≈ 0.05 body length, slope = 27°) and a half cylindrical obstacle (height ≈ 0.1 body length). By propagating down the body an internal torque profile with a maximum around the obstacle, the simulated snake moved forward as observed in the animal. Remarkably, even when frictional drag is low (snake-terrain kinetic friction coefficient of 0.20), the body must push against the wedge with a pressure 5 times that from body weight to generate sufficient forward propulsion to move forward. This indicated that snakes are highly capable of bending vertically to push against the environment to generate propulsion. Testing different controllers revealed that contact force feedback further helps generate and maintain propulsion effectively under unknown terrain perturbations.