Roboticists reverse engineer zebrafish navigation

Using simulations, robots, and live fish, scientists at EPFL and Duke University have replicated the neural circuitry that allows zebrafish to react to visual stimuli and maintain their position in flowing water.

Ecole Polytechnique Fédérale de Lausanne

FacebookXLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

The larval zebrafish robot, Zbot. 2025 BioRob EPFL CC BY SA 4.0

view more 

Credit: 2025 BioRob EPFL CC BY SA 4.0

A paradox of neuroscience is that while brains evolve within specific sensory and physical environments, neural circuits are usually studied in isolation under controlled laboratory conditions. But we can’t fully understand how environmental factors shape brain function without considering the body in which that brain evolved.

The BioRobotics Lab in EPFL’s School of Engineering specializes in developing bioinspired robots to tease apart the brain-body interactions involved in sensorimotor coordination. Now, they have published a study in Science Robotics that provides detailed insight into embodiment, or how the body affects perception, in larval zebrafish.

“Our simulated larval zebrafish provided virtual embodiment, which allowed us to observe its reaction to simulated fluid dynamics and visual scenes. Then, we used a physical robot to observe these interactions in the real world. These connections to the environment can’t be studied with an isolated brain in a lab,” summarizes BioRobotics Lab head Auke Ijspeert.

A fish-eye view

With their translucent bodies, tiny larval zebrafish offer optical access to all their neurons, making them well-studied animal models in biomedical research. For the study, neurobiologist Eva Naumann and her team at Duke University provided a neural network architecture derived from real-time imaging data from the brains of live zebrafish. They also tracked visually driven behavior of the tiny fish, and recorded their reactions when presented with varied visual stimuli that mimicked what they might encounter in flowing water.

The BioRobotics Lab then worked with Naumann and her team to develop a simulation that faithfully reproduced zebrafish visual processing, body mechanics, and neural circuits, from retina to spinal cord. In experiments targeting the optomotor response – the reflexive swimming that helps fish compensate for water currents – the virtual animal closely replicated the behavior of real larval zebrafish.

“It was exciting that we replicated all the different behaviors that Eva and her team observed in the live fish – it suggests we succeeded in reverse engineering the circuitry,” Ijspeert says.

In the process, the team discovered that most neural signals driving zebrafish behavior come from a relatively small part of the retina. Their simulation even predicted two previously unidentified neuron types that explained the live fish’s response to unusual stimuli.

To further validate their work, EPFL postdoctoral researcher Xiangxiao Liu built an 80-cm robotic zebrafish larva equipped with two cameras for eyes, motors to move its tail segments, and the same neural circuits as the simulated zebrafish. In experiments in Lausanne’s Chamberonne River, the robot was able to keep from being swept downstream, even in the disorder of a natural environment.

“The emergence of the optomotor response from our neural circuitry is significant, as some of an animal’s response to any stimulus is random. Despite this randomness, the neural circuitry still converged to reorient the robot and maintain its position,” Liu says.

An open-source platform for studying animal behavior

The BioRobotics Lab is already extending this research to study zebrafish swimming patterns. Their simulation and robot design are also available as open-source tools for researchers to study visuomotor coordination in zebrafish and in other animals. Indeed, Ijspeert emphasizes that the work demonstrates how crucial models and simulations are for understanding which sensorimotor mechanisms are sufficient for certain biological functions.

“In animal experiments, you can only show which sensorimotor mechanisms are necessary to function, but not which are sufficient, but because you can’t remove all mechanisms except one in animals. Here, we have shown that vision alone is in principle sufficient for zebrafish to maintain their position, which is a challenging and non-trivial result.”

The larval zebrafish robot, Zbot. 2025 BioRob EPFL CC BY SA 4.0

Credit

2025 BioRob EPFL CC BY SA 4.0

Journal

Science Robotics

DOI

10.1126/scirobotics.adv4408 

Method of Research

Experimental study

Article Title

Artificial Embodied Circuits Uncover Neural Architectures of Vertebrate Visuomotor Behaviors

Article Publication Date

15-Oct-2025

Eel-Inspired Robots? Study reveals how amphibious animals navigate tough terrain

University of Ottawa

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

“Our study introduces a new model to explain the control of locomotion in elongated amphibious animals”

Emily Standen

— Associate Professor at uOttawa's Faculty of Science

view more 

Credit: University of Ottawa

An international research team has unveiled significant findings regarding the locomotion of elongated amphibious animals. The researchers developed an innovative model that explains how elongated amphibious animals—such as eels—coordinate movement both in water and on land.

This collaborative effort, supported by the Human Frontier Science Program, involved researchers from the BioRob lab at EPFL in Switzerland, the Ishiguro Lab at Tohoku University in Japan, and the Standen Lab at University of Ottawa.

Emily Standen, Associate Professor at uOttawa's Faculty of Science and one of the lead Principal Investigators, led the biological side of the research. “Our study introduces a new model to explain the control of locomotion in elongated amphibious animals,” she says. “We aim to deepen our understanding of the neuromotor control systems used by animals that can adapt their movements between aquatic and terrestrial environments.”

The research, which has spanned multiple years, involved a comprehensive approach combining simulation modeling at Tohoku University, robotics testing at EPFL, and animal observation at the University of Ottawa. “In my lab, we observed eels to better understand their motor control systems and observe how brain signals, local spinal pattern generators and sensory feedback systems influence undulatory locomotion,” Professor Standen explains. “By using eels as a living model, we were able to guide the simulation and robotics models with biological data.”

The models in this study show that basic components of the motor system, like coordination in the nervous system, as well as pressure feedback and stretch feedback, allow for redundant coordination during swimming. This redundancy and the capacity of stretch feedback to allow the exploitation of heterogeneity in the environment to help move forward, may explain why elongated fish like the eel and lamprey can move in terrestrial environments. “These animals are remarkably resilient,” she notes. “Our models point to sensory feedback as the key to allowing them to maintain their locomotor performance.”

Bio-Inspired Robotics

Beyond animal biology, the findings could help engineers design flexible robots for challenging environments. “This research provides new ways of understanding neuromotor control in animals, which can have far-reaching implications for both scientific research and technological advancements,” says Professor Standen. Imagine robots that crawl, slither, or swim through tight spaces, using nature’s engineering to stay flexible and strong.

The study, “Multisensory feedback makes swimming circuits robust against spinal transection and enables terrestrial crawling in elongate fish,” is now published in the Proceedings of the National Academy of Sciences (PNAS). It’s a leap forward in understanding movement and could inspire innovative robot designs in the future.

---

Journal

Proceedings of the National Academy of Sciences

DOI

10.1073/pnas.2422248122 

Method of Research

Experimental study

Subject of Research

Not applicable

Article Title

Multisensory feedback makes swimming circuits robust against spinal transection and enables terrestrial crawling in elongate fish

Robots offer clues to the impressive robustness of eel locomotion

A neural circuit model tested in amphibious robots developed at EPFL shows how multisensory feedback enables eels to swim after a spinal cord injury, while also providing new insights into the evolutionary transition of vertebrates from water to land.

Ecole Polytechnique Fédérale de Lausanne

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

 

The EPFL BioRob Lab's amphibious eel-like robot. 2025 BIOROB EPFL CC BY SA

view more 

Credit: 2025 BIOROB EPFL CC BY SA

Elongated fish like eels and lampreys are remarkable movers. Eels demonstrate exceptional locomotor ability, not only when swimming but also crawling on uneven ground. They can even swim after the part of the spinal cord responsible for controlling movement is damaged, which would lead to paralysis in most vertebrates. But the neural mechanism behind these incredible abilities has long been a mystery.

Previous studies have suggested that sensations of skin pressure and muscle stretching modulate the activity of neural networks called central pattern generators (CPG), which are distributed along the spinal cord and are believed to control the adaptive movement of these undulating species. But these theories have not been fully explored due to the technical difficulty of studying multiple sensory feedback in living animals.

Now, a team including researchers from EPFL’s School of Engineering, Tohoku University (Japan), and the University of Ottawa (Canada) have published a mathematical model of a neural circuit in the Proceedings of the National Academy of Sciences that integrates both stretch and pressure sensation for motion control in eels and their relatives. Specifically, the researchers assumed for their model that each body segment has a CPG-like neural circuit that generates rhythmic movement patterns that are regulated autonomously by stretch and pressure sensory feedback.

Then, the researchers used their model to run computer simulations and experiments using amphibious eel-like robots developed in EPFL’s BioRobotics Laboratory. In aquatic experiments, the scientists confirmed that their model quickly produced stable swimming patterns, and that the stretch feedback especially was central to this rapid stabilization. Remarkably, the same neural circuit involved in swimming also enabled the robot to crawl on land and navigate around obstacles, with the stretch feedback again being vital for pushing against obstacles to generate forward thrust.

“The finding that a neural circuit for swimming can also enable terrestrial movement suggests that the evolutionary transition of vertebrates from water to land may not have required the development of entirely new neural circuits,” says BioRobotics Lab head Auke Ijspeert. “Instead, existing aquatic circuits could have been repurposed – a principle that contributes to our understanding of the evolutionary origins of motor control.”

Building more resilient robots

The team also used their robots and simulations to investigate possible mechanisms that enable real eels to swim even after the spinal cord is severed – an injury that would leave most vertebrates paralyzed. Their findings suggest that if the neural circuits distributed throughout the body have a certain degree of spontaneous rhythm generation ability, then this could combine with stretch and pressure feedback to produce coordinated swimming movements in the simulated and robotic fish, before and after the spinal cord severance site.

In addition to providing new information on the evolution and mechanics of animal motor control, the researchers say their findings could be used to develop robots that are resilient to physical damage, and that can move robustly in unpredictable environments like disaster sites. In particular, control methods that integrate multisensory feedback could help researchers develop robots that can move just as easily underwater as on uneven terrain.

Ijspeert adds that the insights into motor function following spinal cord injury are not only biologically intriguing but may also offer principles for designing decentralized motor control systems that do not rely on brain-based control: “If we can understand how biology controls complex movement using senses in the body (without a brain), we may be able to use this information to better control autonomous machines.”

---

Journal

Proceedings of the National Academy of Sciences

DOI

10.1073/pnas.2422248122 

Article Title

Multisensory feedback makes swimming circuits robust against spinal transection and enables terrestrial crawling in elongate fish

The BioRob Lab's amphibious eel-like robot. 2025 BIOROB EPFL CC BY SA

Credit

2025 BIOROB EPFL CC BY SA

Eel-astic robots? Stretch and pressure are the keys to eels' remarkable locomotive abilities

Tohoku University

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Overview of this study. (A) Schematic of the body and neural circuit model of elongated fish such as eels. (B) CAD image of the developed eel-like robot. (C) Our interdisciplinary approach combining animal, simulation, and robot experiments. (D) Tested various neural circuits. 

view more 

Credit: Kotaro Yasui et al.

A spinal cord injury in most vertebrates likely inhibits locomotion and induces paralysis. Not eels. They not only possess the ability to move through water and surprisingly across land when intact, but can also continue to swim even if their spinal cord is severed.

The neural mechanisms behind these incredible abilities have long remained a mystery. Revealing more about the control mechanisms behind eel-like locomotion could radically improve the development of robots to better enable them to navigate diverse and challenging environments.

An international group of researchers has done exactly this by integrating two types of sensory feedback into a neural circuitry model for eel-like elongated fish and testing it with computer simulation and experiments with a real robot. Their findings revealed eels rely on signals from their bodies - like the feeling of stretch and pressure on the skin - to adjust to different environments. These signals, together with the nervous system's built-in rhythm, may be enough to keep movement coordinated even after a serious spinal cord injury.

Details of the findings were published in the Proceedings of the National Academy of Sciences of the United States of America (PNAS) on August 18, 2025.

"Our findings will help design highly adaptive robots capable of navigating complex and unpredictable environments," explained Kotaro Yasui, an assistant professor at Tohoku University's Frontier Research Institute for Interdisciplinary Science (FRIS), and lead author of the study.

Yasui and colleagues first set out to find the control principle behind eels' movement. They first developed a mathematical model of a neural circuit that integrates two sensory feedbacks: stretch and pressure. The model assumed that each body segment has its own neural circuit, like a Central Pattern Generator, which produces movement rhythms regulated by these sensory signals.

They tested their model with computer simulations and robotic experiments, where the model quickly produced stable swimming thanks to sensory feedback. This same neural circuit also enabled the robot to crawl on land and navigate around obstacles, with the stretch feedback being vital for pushing against obstacles to generate forward thrust.

To explore how eels maintain movement after spinal cord injury, the group conducted spinal cord transection experiments with real eels and corresponding simulation and robot experiments using the neural circuitry model with the stretch and pressure feedback. Simulation and robot experiments revealed that the combination of multisensory feedback and the circuits' own intrinsic rhythm-generating ability allows the body to synchronize its movements across the injury site, even without input from the brain.

The study had the added benefit of furthering our evolutionary understanding of locomotion. "The discovery that a swimming neural circuit also supports movement on land suggests that vertebrates may not have needed an entirely new neural circuit when they transitioned to land; rather, flexible swimming circuits were repurposed, reducing the need for complex top-down control while enabling effective movement across different environments," explained Akio Ishiguro, a professor at Tohoku University's Research Institute of Electrical Communication (RIEC) and co-author of the paper.

The team from Tohoku University also included researchers from Future University Hakodate, the École Polytechnique Fédérale de Lausanne, and the University of Ottawa.

---

Journal

Proceedings of the National Academy of Sciences

DOI

10.1073/pnas.2422248122 

Article Title

Multisensory feedback makes swimming circuits robust against spinal transection and enables terrestrial crawling in elongate fish