The language that all humans speak

Prof. Tamar Flesh develops mathematical models to formulate the legality of everyday human movements and the way the brain puts together complicated movements

The ability to move the limbs is a basic tool in our daily functioning. Without noticing, we perform thousands of movements every day, with each movement characterized by varying intensity, speed, and acceleration. Every movement involves and requires coordination between a large number of muscles and joints, and the central nervous system is the one that is responsible for the precise management of the complicated campaign.

Professor Tamar Flesh, from the Department of Applied Mathematics at the Weizmann Institute, tries to formulate in mathematical language the principles of brain action involved in learning and performing movements. The research includes working with subjects and searching for the legality that dictates their movements, while at the same time an attempt is made to locate and record the electrical activity that occurs in the brain when performing the movements. The goal is to develop mathematical models to formulate the legality of everyday human movements.

One way to develop such models is to look for recurring patterns of movement. "You can look at the execution of complex movements in two principle ways," says Flesh. "There are researchers who believe that the brain learns the complex movements in their entirety, stores them in memory, and pulls out the required movement when needed. But the direction of our research is different. We assume that there is a basic pool of simple movements from which the brain assembles the more complicated movements. Writing the letter a For example, it is a complex action that requires the hand to move the pencil alternately in a straight line and in a rounded line, according to the assumption, the brain is dealing with The challenge is that he builds the complicated movement from a number of basic movements that are performed in sequence."

The basic movements are called "elements of movement", and Flesch describes them as letters of the body's movement language. As the language in our mouth is made up of words and grammar rules, so the movements we perform are based on a limited number of simple movement components, and the language of body movement also has its own rules. The brain contains an "action plan" for the use of the movement elements, and throughout life it refines the abilities to build actions, some of which are very complicated such as dancing, eating and writing.

Flesh and her research students aim to isolate and describe "elements of movement" by tracking how a single movement is performed by a large group of subjects. The laboratory equipment makes it possible to monitor parameters that change over time, such as the speed of movement of the subject, the acceleration of the movement and its intensity. In a study dealing with movement learning, conducted in collaboration with Dr. Avi Karni and research students Ronan Sosnik and Maria Korman, some of the subjects performed the movements while they were strapped into an f-MRI brain imaging device. With this method, "it is possible to locate areas of the brain that participate in learning the element of movement and carrying it out Flesh said.

Assuming that a pool of basic movement components is indeed available to the brain, how does a person choose the composition of the specific movement he performs out of a huge number of possible combinations? Flesh proposed a solution to this question already in the early 80s when she wrote her doctoral thesis at the Massachusetts Institute of Technology (MIT).
In a study conducted in collaboration with Prof. Neville Hogan, she developed a mathematical formula describing the optimization of movements. "The optimization model states that the brain's choice allows a person to make the smoothest movement in space," says Flesh. "In a smooth movement, there is a minimal change in acceleration over time, and the relative speed of the movement is maintained throughout the movement. When we asked subjects in the laboratory to repeat the same movement thousands of times, their movement gradually became smoother as the experiment progressed. They experienced a learning process, and eventually They also reached an optimization of the movement that matches the model we developed."

Using the model, it is possible to predict with great precision the components of the movements that the brain chooses to perform. So, for example, when subjects write a curve in an optimal way, their hand slows down as it approaches the peak of the curvature, and beyond the peak it again increases speed.

The optimization model was first published in 1985 in the Journal of Neuroscience and has since been extensively developed by Flesh and many other researchers. It seems that a diverse formulation of mathematical models helps, among other things, to characterize the difference between a movement performed by a healthy person and the same movement performed by people suffering from motor activity disorders, such as those with Parkinson's disease. "The field of robotics is also influenced by the study of human movement systems," says Flesh, "because the design of the robot's movement systems can be based on the same 'optimal' laws that dictate the working mechanisms of the human brain."