Home

Scientific Fields of Interest

• • Computational Neuroscience

  • Neural Control of Movement

  • Statistical Machine Learning

  • Differential Geometry

  • Financial modeling

"How do the Motions of the Body follow from the Will, and whence is the Instinct in Animals?"

(Isaac Newton, "Opticks")

What are the brain mechanisms that underlie movement generation? In particular, how do these mechanisms give rise to the remarkable flexibility apparent in human motion? How is it that we are able to learn to generate stereotypical movements, follow them over periods of time, adjust and correct them as needed when circumstances dictate? These general questions motivate my research.

My research interests include gaining a better insight into neural coding and motor control. In particular, I am interested in the neural representations of movements, motor learning and optimality principles in motor control. Of special interest is the question of the metrics in which movements are represented and encoded. Another goal of the study is to find out what are basic primitives from which the movements are constructed and their compositionality rules. I believe that the studies of movement primitives should be implemented in the framework of decision-making regarding the choice and tuning of the movement primitive to be composed with the ongoing movement chunk.

I am also interested in applications of differential geometry, information theory and statistical machine learning techniques to the analysis of movements generation, their compositionality rules and neural representations.

The results of my earlier research have been published in peer reviewed journals. The empirical part of the research was based on the analysis of free planar scribbling movements in monkeys and simultaneous extracellular recordings from motor cortical areas. Experimental recordings were performed in the laboratory of Professor Moshe Abeles. Particular features of these data are that the monkey was not presented with any template to follow or a moving dot to track and that the monkey underwent extensive practice during which its free drawings became stereotypical. Neural representation of such movements may differ from the ones proposed in earlier works (e.g. tuning relationships), in which a target or a stimulus were presented to the monkey.

"If we turn to the human activity - conscious, but not following the rules of formal logic, i.e. intuitive or semi-intuitive activity, for example to motor reactions, we will find out that high perfection and sharpness of the mechanism of continuous motion is based on the movements of the continuous-geometric type ... One can consider, however, that this is not a radical objection against discrete mechanisms. Most likely the intuition of continuous curves in the brain is realized based on the discrete mechanism".

(Andrey Kolmogorov "Mathematics - science and profession", 1988)

I have derived a differential equation that describes a movement path compatible with the minimum-jerk criterion and constant rate of accumulating a geometric measurement s along the path (Polyakov et al. Biological Cybernetics (2009) 100(2)):