Lab of Cerebral Cortex Research, Institute of Experimental Medicine


on Control

  • Why does the same network produces different periodic activity patterns, i.e. sharp-wave ripples (SWRs) and gamma oscillation?
  • How do different types of neurons (principal cells and inhibitory neurons) contribute to the dynamics of hippocampal network?
  • How do interneurons integrate convergent synaptic inputs and generate action potentials during different network activity patterns?
  • What is the difference between physiological transient recurrent synchronous states (SWRs) and pathological, highly synchronous states (epileptic events)?
  • How can inhibition exert effective control in one state and fail in the other?

on Coding

  • How do principal neurons represent and procecss information?
  • What are the rules of interactions of neuronal activity patterns?


Associate Professor

Laboratory of Cerebral Cortex Research, Department of Cellular and Network Neurobiology, Institute of Experimental Medicine.

P.O.B. 67




tel: 36-1-tel: 36-1-2109413

fax: 36-1-2109412

E-mail: gulyas[at]

laboratory web page


We use an improved in vitro slice preparation that upon pharmacological manipulations can switch among different activity patterns similar to physiological and pathological in vivo patterns. This arrangement allows selective and quick manipulation of the network's behavior and parallel multichannel recording of network activity (field potentials and multiunit) as well as recording of cellular output (loose-patch recording of action ptentials) and input (voltage and current clamp) from identified neurons or pairs of neurons. We started recording the activity of large population of CA3 pyramidal neurons in the CA3 area using 2P-imaging of calium signals with the intention to understand how are principal neurons representing and processing information. In the near future the in vitro experiments will be complemented with in vivo head restrained imaging of population activity.


I am a "descendant" of the school of Janos Szentagothay. He imaged the cortex as a well designed machines with all the nuts and bolts in the proper place. His inspiring aquarels show that the cortex is built from two types of neurons:

- the majority of the neurons (~80%) are excitatory principal cells with similar features. They form the information processing framework of the cortex. They establish massive long-range projections among cortical areas and the periphery. Their firing pattern codes information and their excitatory synapses implement Hebbian learning to store information in the form of long-term modifiable synaptic weights.

- the rather diverse minority (~20%) are formed by inhibitory neurons. They control the activity of principal cell populations and are important in the implementation if switching among different behavior associated activity patterns.


Since I joined the Laboratory of cerebral Cortex in 1991 we studied the role of inhibitory neurons in the control of the activity of the cortex. We work in the hippocampus, since it is a simplified version of the neocortex with similar building elements and connectivity rules.

I was working for nearly 15 years as an anatomist trying to map the building elements of neuronal networks and their connectivity.

  • - we mapped the properties of different types of inhibitory neurons
  • - studied their connectivity with different cell types
  • - described quantitatively their input and output properties

- afferents from the median raphe and the medial septum are controlling specific subsets of inhibitory neurons, that selectively transmit and amplify the effect of these centers (1,2,3,5,6)

- we described and studied several subsets of hippocampal inhibitory neuron types (4,7,10,13,15,21) often with correlated anatomical and physiological methods (17)

- most notably we described that Calretinin containing interneurons form a new class of interneurtons: they selectively innarvate othe interneurons (11,12,19,22)

- we described that a specialized hippocampal inhibitory cell population, the hippocampo-septal cells (HS cells) project outside the hippocampus, while locally innervate other inhibitory neurons

- we proved that in the hippocampus principal cells innervate inhibitory neurons through a single synapse (16,18)

- we proved that while perisomatic inhibition is responsible for the control of the output of principal cells, denritic inhibition modulates input integration and might influences synaptic plasticity (20)

- in a series of quantitative studies we mapped the number and ratio of synaptic inputs on different domains of identified cell types. Large differences were uncovered in the rules of synaptic convergence onto principal cells and different types of inhibitory neurons (27,28,29,33,38,42)

- described the distribution of endocannabinoid-hydrolyzing enzymes and KCl cotransporter distribution on hippocampal neurons (30,35)

Recently I switched to the field of electrophysiology with the intention to understand how these building elements are integrating synaptic input and how do they control the behavior of the network?

- demonstrated the combinatorial explosion of inhibitory neuronal types (24), when anatomical, physiological and pharmacological properties are all taken into account

- demonstarated that parvalbumin-containing basket cells are responsible for the generation of gamma oscillations and that their transmission is modulated by mu-opiate receptors and cholinergic activation (39,40)

- we examined how different neuron types fire during sharp-wave ripples [SWRs](43), we showed that SWRs are initiated by stochastic pyramidal cell activity and that ripple oscilaltion is generated in the reciprocally connected inhibitory network of PV-containing basket cells (46)

- we demonstrated that SWRs and interictal like (epileptiform events) are different types of highly synchronous events (THAEs), pathologic HFOs observed during epileptiform events form when PV-basket cells get into depolarization block and pyramidal cells start to fire in pseudo synchrony (44). The differences were umamrized in a review (47).