Neural mechanisms of visuomotor transformations in larval zebrafish

The National Eye Institute awarded Duke University $127,718 on December 21, 2025, under the Vision Research program (CFDA 93.867) to support cellular-level dissection of the visuomotor circuit in larval zebrafish, focusing on neural mechanisms that transform continuous visual motion signals into discrete motor commands.

The research investigates how specific motion-selective neurons in the pretectum drive activation of recurrent inhibitory circuits composed of midbrain spinal projection neurons, excitatory hindbrain neurons, and hindbrain inhibitory neurons. The work leverages the optically and genetically accessible larval zebrafish model to characterize the optomotor response—a visually guided behavior in which zebrafish stabilize body position by compensating perceived optic flow through discrete locomotion events. The investigator will implement these circuits into quantitative models and the physics-based neuromechanical simulation SimZFish to clarify functional circuit architecture and mechanisms underlying the transformation from continuous visual input to discrete movement.

Performance of work occurs at Duke University in Durham, North Carolina, with a period of performance from award date through April 30, 2028. The award is a Project Grant, an assistance type within the Vision Research program designed to fund innovative research addressing eye and visual system function, disease prevention and treatment, and vision workforce development.

Abstract

To survive in dynamic environments, animals utilize continuous sensory information to drive precise and coordinated motor behavior. For instance, to catch a flying ball, one must reach with a targeted, discrete arm movement. Therefore, the neural networks processing visual information must transform these continuous motion cues into distinct motor commands to achieve locomotion with appropriate speed, vigor, and duration. So far, there has been no single-cell resolution and mechanistic characterization of these circuits in vertebrates due to the overwhelming size of the mammalian brain. In this proposal, I will leverage the optically and genetically accessible larval zebrafish (danio rerio) for cellular level dissection of the visuomotor circuit underlying a visually guided behavior, the optomotor response (OMR). During the OMR, zebrafish stabilize their body's position by compensating perceived optic flow via discrete locomotion events (i.e. bouts), consisting of undulating tail movements for short periods followed by passive glide phases. Previous behavioral and neural imaging studies have characterized motion-processing neural circuits across the brain, including the retinorecipient pretectum (PT). Recently, we have implemented these circuits into quantitative models and a physics-based neuromechanical simulation, simzfish, which allows for simulated OMR behavior experiments with biologically derived neural circuits. Yet, these virtual whole-brain OMR circuit models lack clarity in the functional circuit architecture and mechanisms that transform continuous visual motion encoded by the PT into discrete movement. Here, I hypothesize that specific motion-selective PT neurons drive the activation of a recurrent inhibitory bout- generating circuit, composed of specific midbrain spinal projection neurons, excitatory hindbrain neurons, and hindbrain inhibitory neurons. To determine the functional neural circuit that transforms visual motion into motor commands, I will use in vivo volumetric two-photon imaging with precise 3D holographic optogenetic photostimulation to map cellular functional connections across brain regions between motion responsive PT neurons and the midbrain cells that initiate bouts (aim 1). To determine how continuous visual motion is converted into discrete locomotor events, I will investigate how the midbrain cells recruit specific excitatory and inhibitory hindbrain neurons using the same all-optical techniques with simultaneous tail tracking (aim 2). Results from both aims will update the simzfish neural circuit and then tested in virtual experiments to validate these findings. Using integrative all-optical, behavioral, and neuromechanical approaches, this proposal aims to provide insights into how single neurons transform sensory information into motor commands and general principles of vertebrate locomotion control. Ultimately, my long-term goal is to comprehensively understand how neurons compute information across brain regions to generate movement for future applications to develop physiologically accurate prosthetics and treat motor neurodegenerative diseases.