How single neurons compute information?
NeuronsThe neuron is the basic building block of the brain. Unlike the chip of the computer which is a simple computing element, the neuron is a complex computing machine. It is complex both anatomically and functionally. We seek to understand how different neurons in different brain regions such as sensory, motor and olfactory integrate their incoming inputs.

For this purpose, we study dendrites, which are the main structures of the neurons that are the main recipients of synaptic information. Dendrites can be regarded as the computation powerhouse of the cortex and thus are crucial for understanding the input-output function, and ultimately encoding and decoding of cortical information. One of the main issues my lab is concentrating on is to determine the role of dendrites in cortical processing both in-vitro and in in-vivo. This approach is unique in that it connects theoretical concepts with novel experimental techniques to make the direct link from neuronal coding at the dendritic level to network level and hopefully also to animal behavior.

We use advanced microscopy, electrophysiology, light activated manipulations and computer modeling to get insight into these complex, intricate and marvelous brain structures called dendrites.

We believe that understanding how dendrites work is essential for our mechanistic understanding of how the brain works.

Plasticity mechanisms in principle cortical neurons
apical dendrite patchOne of the most basic characteristics of the brain is its ability to ever change throughout life from birth to death. In the lab, we are interested to decipher the cellular and molecular mechanisms responsible for long term plasticity changes in cortical neurons, changes that underlie the capacity of the brain to change and adapt to new experiences and challenges.

We study how synapses of different cell types and cortical regions change in response to previous experiences. We study a diverse brain regions and cell types as our main hypothesis is that different brain regions solve different learning and memory problems thus mechanisms may differ markedly between brain regions and cell types.

We use brain slices to record from soma and dendrites combined with advanced light activated and imaging methods along with computer simulations.

Mechanisms underlying sensory-motor learning in the brain

A major effort in the lab is to decipher how information is represented in the brain and what are the changes that occur in the brain that are responsible for learning and memory of new experiences. To get insight into these questions, we record brain activity of behaving mice using high resolution imaging with two photon fluorescence microscopy in-vivo. Combined with fluorescently tagged protein sensors we are able to monitor that activity across different scales, molecular, cellular and network.

Presently, we are concentrating on the motor system since it is an extremely important and conserved system throughout evolution of mammals.

Understanding the mechanisms underlying cortical motor control may assist in understanding the mechanisms of various brain diseases such as Parkinson’s and in turn may assist in development of novel treatment strategies.

In a recent study using advanced electrophysiological and imaging methods, the group succeeded in recording in unprecedented resolution from tuft dendrites of cortical principal pyramidal neurons. Tuft dendrites are the main recipients of feedback and neuromodulatory information of the brain. Despite their importance due to their small size they were never been studied directly. dendrite patchIn this study, we and our collaborators (Larkum’s group from Bern) present a generalized new conceptual framework of how information is integrated in the principle neocortical pyramidal neurons. This framework unifies the properties of all fine dendrites in the tree. The thin basal dendrites receiving bottom-up information and the distal tuft dendrites receiving top-down information serve as parallel information integrators where information is locally amplified. This pre-integrated information is passed on to the two main integration zones, the apical Ca2+ initiation zone which serves as the site of integration for feedback connections via NMDA spikes and the axo-somatic integration zone which serve as the main decision zone of the neuron.
Synaptic Integration in Tuft Dendrites, Larkum et al. (2009) Science