Fred Delcomyn - Research Overview

Since I am retired, I am no longer accepting students or postdocs. This section describes the research I was engaged in when I was active.

My research focuses on the way in which the nervous system of an animal is able to generate, coordinate, and control the individual movements that together make up the animal's behavior. In particular, I study the way in which insects walk because walking is a relatively simple behavior that is common to most animals yet at the same time is not too complicated to understand. I use physiological techniques, computer simulation, and robotic engineering to study the insect locomotor system.

This overview of my research is aimed at a general audience. For more detailed and technical information, click on the links in the research menu or in the text below.

What is the neural basis of locomotion? Put in more common terms, how does an insect determine which leg to move next as it walks? Research has shown that two mechanisms are at work. First, small groups of nerve cells, called neural networks, work to produce the fundamental alternating back-and-forth movement of each leg. These networks interact with one another to coordinate the movement of one leg with that of the others. Second, and usually just as important, information about the actual performance of each leg is provided by sense organs in the legs and sent back to the nervous system. There it helps to influence the actual movement that is to be carried out for each particular step.

In order fully to understand how walking is generated and coordinated, it is necessary to understand the role of sense organs that contribute to it. Much of the focus of my work in recent years has been to investigate the role of sensory feedback in walking and to study sense organs that are thought to be important in coordination.

Computers can make a significant contribution to our understanding of biological processes quite apart from their obvious use in data acquisition and analysis. One way they can do this is through the use of computer simulation. By formalizing hypotheses about functional relationships between components of a system that is being studied, a researcher can test hypotheses in a systematic way. Hence, a researcher who has an idea about how a neural network may control leg movement can formalize this idea in terms of equations that describe the output of the network as a function of input from elsewhere in the nervous system and from sense organs in the legs. The equations can then be used as the basis of a computer simulation of the network, and the simulation tested to see how well it produces an output similar to that generated in a live animal. Such testing can lead to predictions about what physiological connections there may be between different neural centers and then to a better understanding of how the system works.

In my own research, I am developing a simulation of an entire walking insect, including not just the elements of the central nervous system that control walking, but the physical body of the insect and its muscles and sense organs. This simulation will be used to test ideas about how coordinated walking over rough or irregular terrain can occur.

What does robotics have to do with the walking of insects? Actually, quite a bit. Roboticists have a strong interest in building robots that can walk, because robots with legs can go where those with wheels cannot, such as into structurally damaged buildings for search and rescue missions, or on rugged terrain or off planet for exploration. For an engineer designing the control system for a walking robot, one of the biggest challenges is to devise a way for the legs to move properly while at the same time making sure that their movements are flexible enough to adapt to the specific surface on which the robot is walking.

In principle, this is the same problem faced by an insect or other walking animal. How can the legs be moved properly to achieve a smoothly coordinated gait, while at the same time allowing the flexibility of adjusting movements and foot placement as circumstances warrant? This is the important question.

From this point of view, roboticists are interested in animal, especially insect, walking because it may provide information that will help them to develop better controllers for their machines. Neurobiologists, on the other hand, may find that trying to design a flexible controller for a walking robot will help them think about what the nervous system must have to do in order to achieve the same results.

This link will give you information about personnel, students and others who have worked with me.