Towards Cognitive Implants: a Case Example on Learned Sensorimotor Transformations

Tarih: 16.04.2019
Yer: Boğaziçi Üniversitesi Kandilli Kampüsü BME Binası AZ-19

Towards Cognitive Implants:

a Case Example on Learned Sensorimotor Transformations

Prof. Tansu ÇELİKEL

Donders Institute for Brain, Cognition and Behavior, Radboud University, Netherlands

About The Seminar:

Sensorimotor transformations are the first complex computations that animals learn. As baby brains perform thousands of trial and error to generate stimulus (and eventually context) specific actions, distributed networks in the brain implement a set of transfer functions to excel in sensorimotor transformations. Although these computations are at the basis of all goal-directed behavior throughout life, the unifying computational principles are yet to be discovered. In this seminar, I will give an update on our progress in addressing this question. The talk is divided into four parts: First we will introduce a sensorimotor task which rodent brains solve in <90 ms (by treating the moving body parts as coupled manifolds). Performing quantitative analysis and mechanical modeling of the sensory input to the brain and the motor output from it, we will determine that sensorimotor transformations are learned after the maturation of intracortical projections in the somatosensory cortex. In the second part, we will introduce a novel method for neural network tracing; using this approach we will map the brain-wide sensorimotor circuits as a weighted directed graph. By recording and controlling the neural activity in key structures (nodes) along this circuit, we will identify how information is processed, transferred and recovered for sensorimotor transformations. Next, we will deduce the neural transfer functions for translating sensory inputs to motor output, and test their performance for adaptive motor control using a robotic navigator.  Porting some of the circuit and computational insights from these studies on a CMOS chip, we will open the discussion with a road-map for cognitive implants.

About the Speaker:

Tansu Celikel received his PhD in Systems Neuroscience from SISSA (Italy) in 2001.  After postdoctoral research at the University of California, San Diego and the Max-Planck Institute for Medical Research, he started his first laboratory at the University of Southern California in the Departments of Neurobiology and Biomedical Engineering.  Since 2012, Dr. Celikel has been with the Donders Institute for Brain, Cognition and Behavior at the Radboud University in the Netherlands, where he is a university professor, the Founding Chair of the Department of Neurophysiology, and the Founding Director of the binational (Dutch/German) Graduate School of Bionics. Dr. Celikel’s research is focused on reverse engineering the brain circuits and neural computations that are responsible for sensorimotor control, and forward engineering brain inspired in silico circuits that perform active sensing and mapless navigation.  He is a Sloan Fellow (USA), Alexander von Humboldt Fellow (Germany), and a Whitehall Investigator (USA) and a recipient of the young investigator award from the Italian Ministry of University and Scientific Research and Technology.