What is a neural scheme

Theoretical and experimental analysis of neural circuits

Research Report 2013 - Max Planck Institute for Brain Research

Tchumatchenko, Tatiana; Letzkus, Johannes
Theory of neural network dynamics (Tatjana Tchumatchenko); Neocortical Circuits (Johannes Letzkus)
The brain is by far the most complex system we know. At the MPI for Brain Research, two new research groups are using complementary approaches to better understand the function of neural circuits: Tatjana Tchumatchenko’s group uses theoretical approaches to research neural information coding. Johannes Letzkus ‘team uses experimental approaches such as 2-photon microscopy and optogenetics to elucidate which activity patterns occur in neocortical circuits in different behavioral patterns and how this activity influences the behavior of the animal.


Research into the brain has made great strides in a number of areas over the past few decades: We currently have a good understanding of the regions in which certain functions are located. At the same time, the physiology and biochemistry of individual nerve cells was characterized in great detail. In contrast, we still know very little about how neural circuits - groups of neurons of different types - work and process information. The analysis of these processes in the mesoscopic range - a size range from one nanometer to one micrometer - has only been made possible in recent years by technical advances and is now a central area of ​​neuroscience. The groups of Tatjana Tchumatchenko and Johannes Letzkus at the Max Planck Institute for Brain Research use complementary approaches to understand the fundamental functions of neural circuits.

Tatjana Tchumatchenko's group focuses on theoretical modeling and mathematical analysis of individual neurons and the dynamic properties of neural networks. The group uses analytical methods and computer simulations to investigate how neurons represent the incoming stimuli, how they interact and how dynamic circuits arise from them. On the method side, her focus is on linear and non-linear differential equations, multi-dimensional stochastic integrals and information theory.

An important property of neural activity is shown by the observation that the activity of several neurons is subject to a fine temporal and spatial coordination. How exactly the coordination comes about and which parameters influence it is the subject of research. In a series of current works, Tatjana Tchumatchenko asked herself how well the coordination of digital spike patterns of pairs of neurons (see Fig. 1) - d. H. the sequence of their action potentials - can be represented by the threshold crossings of a Gaussian random process [1; 2; 3]. The work has shown that this model not only describes neural coordination well, but can also be fully described using compact mathematical expressions. The aim of the group is to verify or falsify the theoretical results obtained with the help of experimental data. The group is therefore in direct exchange with the experimental colleagues at the MPI for Brain Research who work with intracellular single or multi-cell recordings.

Experimental analysis of neocortical circuits

A fundamentally important property of neural circuits is that they are composed of many different types of neurons, which presumably also fulfill very different functions in information processing. However, for technical reasons, these specific functions are still poorly understood. Johannes Letzkus' team is investigating how activity in different types of neurons contributes to brain functions such as learning and attention. To do this, the scientists use a recently developed combination of transgenic mouse lines and viral vectors [5; 6] to identify different types of neurons (Fig. 2). In combination with 2-photon microscopy [7], this approach makes it possible to record the activity patterns of the different neuron types in different behavioral examples (Fig. 3). This provides information on whether the type of neuron examined could play a role in the behavior of the animal.

For example, researchers recently found that certain inhibitory interneurons in the auditory cortex are inhibited when the animal is learning [8]. To get from this correlative analysis to causal relationships, they use optogenetics [9], a technique that enables the activity of defined neuron types to be remotely controlled by means of light (Fig. 4