Closed-loop brain-behavior system Computational Neuroscience

Description

We live in the world that is continuously changing and often challenging. We utilize flexible behaviors to survive despite the challenges we confront. Our brain gives rise to these flexible behaviors via its functionally distinct regions that process various types of sensory and motor information. A great deal of influential research has studied the brain and behavior, however, the focus has been only on one of them overlooking the importance of the other. One reason for this is the current technical difficulties in empirical studies. For instance, finding the neural mechanism that underlie a particular type of behavior is difficult because the recorded neural activity is very local. Since the brain regions are tightly interconnected and is only partly dissociable during behavior, this neural activity might stem from integration of inputs from other regions rather than driving the observed behavior. Computational modeling is a tool that can be helpful overcoming empirical difficulties by providing biologically plausible predictions given that the model is to a needed extent biologically constrained. We are developing a closed-loop system associating neural activity to behavior and vice versa. The neural network in our system is biologically realistic, therefore, its predictions potentially improve our understanding of the particular brain regions we include in our system. We use spiking neurons as building blocks of the neural network and for simulating this neural network we use NEST, which is a widely used simulator in computational neuroscience community worldwide (www.nest-simulator.org). We offer projects at bachelor and master level to further develop this system incrementally by adding neural networks that mimic the activity of particular brain regions that are involved in particular functions. These functions include (but not limited to) reward learning (reinforcement learning in biology), goal-directed spatial navigation and decision making.

Required skill:

Very good programming skill is required (preferably in Python).

Literature:

Jordan, J., Weidel, P., & Morrison, A. (2019). A closed-loop toolchain for neural network simulations of learning autonomous agents. Frontiers in computational neuroscience13, 46.

Brzosko, Z., Zannone, S., Schultz, W., Clopath, C., & Paulsen, O. (2017). Sequential neuromodulation of Hebbian plasticity offers mechanism for effective reward-based navigation. Elife6, e27756.

Supervisors:

Prof. Dr. Sen Cheng and Dr. Mohammad M. Nejad

The Institut für Neuroinformatik (INI) is a central research unit of the Ruhr-Universität Bochum. We aim to understand the fundamental principles through which organisms generate behavior and cognition while linked to their environments through sensory systems and while acting in those environments through effector systems. Inspired by our insights into such natural cognitive systems, we seek new solutions to problems of information processing in artificial cognitive systems. We draw from a variety of disciplines that include experimental approaches from psychology and neurophysiology as well as theoretical approaches from physics, mathematics, electrical engineering and applied computer science, in particular machine learning, artificial intelligence, and computer vision.

Universitätsstr. 150, Building NB, Room 3/32
D-44801 Bochum, Germany

Tel: (+49) 234 32-28967
Fax: (+49) 234 32-14210