Skip to main content

Neuromechanical basis of airflow-dependent antennal positioning in hawkmoths

Time: Mon 2020-12-07 10.30

Location: Zoom link for online defence (English)

Subject area: Computer Science

Doctoral student: Dinesh Natesan , Beräkningsvetenskap och beräkningsteknik (CST)

Opponent: Prof. Volker Dürr, Bielefeld University, Bielefeld, Germany

Supervisor: Prof. Örjan Ekeberg, Beräkningsvetenskap och beräkningsteknik (CST); Prof. Sanjay P. Sane, National Centre for Biological Sciences, Bangalore, India

Abstract

Fast behaviors, seen in varied life forms, are often considered to be stereotypic and reflexive and the control neural circuits to be hard-wired. However, many such reflexes have been shown to respond in a context-dependent manner. The work presented in this dissertation focuses on uncovering the principles of one such context-dependent behavior - antennal positioning in insects.

Insect antennae acquire multimodal sensory cues that are required for a wide range of behaviors. These include odor, temperature, humidity, as well as mechanical vibrations from the surroundings. Each modality encodes a different aspect of the environment and is used appropriately to control behavior. Antennal vibrations, for instance, provide feedback relevant for flight stabilization, and is used to modulate wing movements on short, stroke-to-stroke, timescales. Olfactory cues, on the other hand, indicate presence of food and mates, and are used to alter flight trajectories over longer timescales of multiple wing-strokes. Therefore, for a proper behavioral response, the antennae must optimally acquire sensory cues over multiple timescales. Context-dependent modulation of the antennae perhaps enhances their functionality by tuning their dynamic range.

This dissertation focuses on one context, namely airflow, and its effect on antennal positioning. Hawkmoths, and diverse insects, actively position their antennae at the onset of flight by bringing them forward. During flight, they dynamically alter this position based on airflow. Two antennal mechanosensors are involved in this behavior, one being the Böhm’s bristles, which monitors and feeds back the position of the antennae, and the second being the Johnston’s organs, which are stimulated by frontal airflow generated during flight.

The first part of the thesis concerns the control algorithms that underlie the sensory integration of antennal mechanosensory input to produce airflow-dependent antennal positioning. Using the Oleander hawkmoth, Daphnis nerii, as a system of study, the behavior is investigated with a combination of experiments and computational techniques. We find that the dynamics of this behavior can be captured by a tunable feedback loop consisting of two components. One, a negative feedback loop that stably maintains antennae at a preferred position, or set-point, using positional feedback from the Böhm’s bristles. Two, a dynamic set-point that is modulated by airflow (and other context specific cues). Furthermore, a minimalistic model neural circuit based on these components simulate airflow-dependent modulation of antennae. Such circuits could enable moths to maintain stable antennal position on short timescales while retaining context-based flexibility over longer durations.

The latter half of the thesis focuses on each of the individual components. The neural mechanisms underlying modulation of set-point by the Johnston’s organs are investigated using behavioral and electrophysiological experiments. The positional feedback, sensed and encoded by the Böhm’s bristles, is investigated using biomechanical models. These provide an understanding of how airflow-dependent, or more generally, context-dependent antennal positioning arises as a result of these individual components. As a whole, this dissertation provides a conceptual framework that utilizes experimental and computational techniques to formally describe and understand context-dependent behaviors.

urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-285597

Page responsible:biblioteket@kth.se
Belongs to: News & Events
Last changed: Nov 11, 2020