Jerzy Konorski poster




Copyright ©2013 Designed and cretaed by SOJUZ Studio

Symposium 2013

On 19th September 2013, in the Nencki Institute we organized symposium entitled “1st Nencki Symposium on: Jerzy Konorski’s Contribution to Modern Neuroscience”. This event commemorated the outstanding achievements of Jerzy Konorski and marked 40th anniversary of his death. The conference was organized by the Department of Neurophysiology, Nencki Institute of Experimental Biology and sponsored by BIO-IMAGing in research INnovation and Education Project, supported by the EU FP7 Capacities Programme (Research potential).

The scientific achievements of Jerzy Konorski had worldwide impact on the concepts of fundamental brain function. His career started before World War II when he discovered, jointly with Stefan Miller, instrumental (Type II) conditioned reflexes and carried out intensive research on this topic. Indeed, the distinction between the classical (Pavlovian or Type I condition reflexes) and instrumental (Type II) learning appears to be of fundamental importance since huge body of data has indicated that classical conditioning and instrumental learning involve quite different neural networks. In 1948, that is, before the ‘Iron Curtain’ substantially isolated intellectual exchange between the East and West, Cambridge University Press published Konorski’s 1st seminal book entitled: “Conditioned reflexes and neuron organization” in which he attempted to integrate classical Pavlovian theory with Sheringtonian neurophysiology. In this book Konorski established the concept of the „plasticity” of neuronal networks. Almost 20 years later, in 1967, University Chicago Press published Konorski’s 2nd book entitled: “Integrative activity of the brain” in which he formulated many of the ideas which underlie heuristically fruitful programs in the 21st century neuroscience.

The symposium consisted of seven lectures in which a number of important concepts formulated by Jerzy Konorski were discussed. The Program and abstracts of these lectures are listed below.

Andrzej Wróbel
Chair of the Organizing Committee


to commemorate this outstanding neuroscientist on the 40th anniversary of his
death and 95 anniversary of establishment of the Nencki Institute

19 September, 2013 at the Nencki Institute of Experimental Biology

Bolek Srebro, University of Bergen, Bergen
Reflections on Jerzy Konorski „Integrative Activty of the Brain” – nearly half a century after

Rodrigo Quian Quiroga, University of Leicester, Leicerster
Gnostic cells in the XXI century

Guy Orban, University of Parma, Parma
Gradient selective neurons: From single cells to human fMRI

Yadin Dudai, Weizmann Institute of Science, Rehovot
Konorski’s heritage: How Reactivity and Plasticity permit memory to anticipate the future

Wulf Eckhard Haubensak, The Research Institute of Molecular Pathology, Vienna
Emotional mechanics in limbic circuits

Ewelina Knapska, Nencki Institute of Experimental Biology, Warsaw
Social modulation of learning

Stephen Maren, Texas A&M University, Texas
Brain circuits for contextual control of fear

Opening of the symposium and delivery of the Nencki Award to Bogdan Dreher
see video

Round Table Memoirs by Konorski’s friends and pupils
see video


Bolek Srebro
Reflections on Jerzy Konorski „Integrative Activty of the Brain” – nearly half a century after

The Department of Biomedicine, Neuroscience Research Group, University of Bergen, Bergen, Norway
Konorski’s monograph “Integrative Activity of the Brain” was published by The University of Chicago Press in 1967. Its Polish translation was published by PWN two years later. The book was a bold effort to summarize and conceptualize in neuroscientific terms then current state of knowledge on motivational and cognitive processes in the mammalian brain. It reviewed a large body of new experimental work on classical (Pavlovian) and instrumental (Konorski and Miller) conditioning, presented new results on transient (dynamic) memory and, finally, it proposed a grand model of integrative function of the brain. In my presentation I will review the background of the book and Konorski’s motivation for writing it. After a short summary of the most important ideas presented in the book and I will try to relate and confront some of these ideas with the recent developments in brain sciences.

To some disappointment of the author, the book did not receive a worldwide attention as it probably deserved at the time of publication. One significant factor that might have contributed to this was Konorski’s death a few years after publication of the book and hence his inability to provide an updating commentary. The other inevitable factor was an explosive development of brain sciences in the next decades which brought a wealth of new findings. Another possible factor is a general dislike of and empirical skepticism to any grand models. Looking back after almost half a century at the importance of Konorski’s “Integrative Activity of the Brain” one can safely state that it was almost a “Summa Neurobiologica” of its time. Today this book belongs to a collection of documents on development of important concepts in modern neuroscience. Furthermore, some of the ideas presented in the book “seeped” into the subconsciousness of modern neuroscientists’ and are still relevant today as this symposium will attempt to show.


Rodrigo Quian Quiroga
Gnostic cells in the XXI century

In this short review*, I revise the notion of gnostic cells posited by Konorski, together with similar arguments by James, Lettvin and Barlow – namely, the idea of pontifical, grandmother and cardinal cells, respectively. I then discuss how the characteristics of the recently discovered concept cells, i.e. neurons in the human medial temporal lobe with a very high degree of specificity and invariance, fits the conjecture of gnostic or grandmother cells and then discuss the key role of concept cells in memory formation.

*The review is available on the site of Acta Neurobiologiae Experimentalis vol. 73, No. 3 (http://


Guy A. Orban
Gradient-selective neurons: From single cells to human fMRI

Gradient-selective neurons

Gradient-selective neurons were discovered in the nineties at the end of the Insight projects, supported by FET. Xiao and others (1997) showed that MT/V5 neurons were selective for linear speed gradients representing planes tilted in depth and relied upon their antagonistic surround for this selectivity, thus providing a mechanistic explanation. Subsequently, similar neurons were found in MSTd (Sugihara et al. 2002), and invariance for mean speed was demonstrated. The most complete description of speed-gradient selective neurons was provided by Mysore and colleagues (2010), who compared first and second-order speed-gradient selective neurons in MT/V5 and FST. Neurons were more selective and this selectivity was more invariant in FST compared to MT/V5, in agreement with the anatomical link between these two areas. Many of the second-order neurons in FST were selective for saddle-shaped surfaces, an intriguing finding. Given that many joints are in fact saddle-shaped, at least from certain view points, this suggests a manner in which shape signals can be injected into the motion stream to extract action-related signals, which combine motion and shape.

Similar gradient-selective neurons were also described for texture and disparity gradients. For texture only first order gradient selective neurons have been described in CIP (Taira et al. 2000) and TEs in infero-temporal cortex (Liu et al 2004). First-and second-order disparity gradient-selective neurons have also been documented in the ventral (TEs, Janssen et al. 2000) and dorsal visual pathways (CIP, Taira et al. 2000; and AIP, Srivastava et al. 2009), and even in ventral premotor cortex (Theys et al. 2012). The ventral neurons play a role in discrimination (Verhoef et al. 2010), while those in the dorsal pathway likely analyze the 3D shape to control prehension (Sakata et al. 1995).

Using parallel functional imaging to extend knowledge to the human brain

In order to show that similar neurons exist in human cortical areas, we proceeded in two steps: demonstrating that in the model system, here the macaque, the presence of gradient-selective neurons can be captured by fMRI responses in a given paradigm, and then applying this paradigm in human subjects to identify the corresponding areas. As paradigm for investigating 3D structure from motion we introduced the comparison between viewing randomly-connected lines rotating in depth and those same lines translating in the fronto-parallel plane (Orban et al. 1999). It turns out that in the monkey this contrast activates only a few higher-order visual areas: notably MT/V5 and FST (Vanduffel et al. 2002). In fact, gradient-selective neurons in FST respond differentially to these two stimuli (Mysore et al. 2010) thus validating the paradigm. Recently, the human homologues of MT/V5 and some of its satellites have been mapped retinotopically (Kolster et al. 2010). Human MT/V5 and to some degree putative FST (pFST) are also activated by the comparison of rotating vs. translating random lines, suggesting that these areas house speed-gradient selective neurons, as do their monkey counterparts.

In the case of disparity, the fMRI data are even more impressive. Disparity gradient-selective neurons have been reported in TEs, AIP and ventral premotor cortex and all three areas show interactions in the fMRI (Joly et al. 2009) between the factors shape of the surface (flat vs. curved) and stereopsis (present /absent). The same interaction in humans (Georgieva et al 2009) activates phAIP and DIPSA, which together represent the homologue of monkey AIP (its anterior motor and posterior visual side, respectively, Durand et al 2009), as well as ventral premotor cortex. The absence of a temporal activation in that study is probably due to susceptibility artifacts that prevented recording fMRI responses in the anterior part of the temporal lobe. Thus there is reasonable evidence that disparity gradient-selective neurons occur in phAIP and ventral premotor cortex of humans.


Gradient selective neurons are a family of neurons which represent 3D surfaces as populations (for review see Orban 2011). They fall short of being gnostic units as Konorski (1967) defined them. However they go a long way in showing that the visual system of primates provides more powerful and complex analyses of the retinal image than generally assumed (for another example see Sary et al. 1993). Even if several of these neurons need to act together for uniquely representing a 3D surface, they indicate that visual neuronal populations can encode complex visual features and not just low-level features such as orientation, as is often assumed. It may well be that this is the rule in non-human primates, while the increase in size of the brain in humans may have allowed the development of grandmother neurons or gnostic units, at least for some very familiar visual objects. In this respect it is possible that the human hippocampus and medial temporal lobe (MTL) is very different from that of animals and certainly of rodents (Orban, in press). The expansion of the volume of this region as well as the change in the nature of afferent signals, may explain why the first recordings of single neurons in human MTL (Quiroga et al. 2005) were met by most neuroscientists with skepticism, because they only were familiar with animal results.


Durand JB, Peeters R, Norman JF, Todd JT, Orban GA (2009) Parietal regions processing visual 3D shape extracted from disparity, Neuroimage 46: 1114–1126,

Georgieva S, Peeters R, Kolster H, Todd JT, Orban GA (2009) The processing of three-dimensional shape from disparity in the human brain, J Neurosci 29: 727–740.

Janssen P, Vogels R, Orban GA (2000) The three-dimensional shape coding in inferior temporal cortex, Neuron 27: 385–397.

Joly O, Vanduffel W, Orban GA (2009) The monkey ventral premotor cortex processes 3D shape from disparity, Neuroimage 47: 262–272.

Kolster H, Peeters R, Orban GA (2010) The retinotopic organization of the human middle temporal area MT/V5 and its cortical neigbors, J Neurosci 30: 9801–9820.

Konorski J (1967) Integrative Activity of the Brain: An Interdisciplinary Approach, University Chicago Press, Chicago, IL.

Liu Y, Vogels R, Orban GA (2004) Convergence of depth from texture and depth from disparity in macaque infero-temporal cortex, J Neurosci 24: 3795–3800.

Mysore SG, Vogels R, Raiguel SE, Todd JT, Orban GA (2010) The selectivity of neurons in the macaque fundus of the superior temporal area for three-dimensional structure from motion, J Neurosci 30: 15491–508.

Orban GA (2011) The extraction of 3D shape in the visual system of human and non-human primates, Ann Rev Neurosci 34: 361–388.

Orban GA (in press) Which animal model for understanding human navigation in a 3D three-dimensional world?, Behav Brain Sci.

Orban GA, Sunaert S, Todd JT, Van Hecke P, Marchal G (1999) Human cortical regions involved in extracting depth from motion, Neuron 24: 929–940.

Quiroga RQ, Reddy L, Kreiman G, Koch C, Fried I (2005) Invariant visual representation by single neurons in the human brain, Nature 435: 1102–1107.

Sakata H, Taira M, MurataA, Mine S (1995) Neural mechanisms of visual guidance of hand action in the parietal cortex of the monkey, Cereb Cortex 5: 429–438.

Sáry G, Vogels R, Orban GA (1993) Cue-invariant shape selectivity of macaque inferior temporal neurons, Science 260: 995–997.

Srivastava S, Orban GA, DeMaziere PA, Janssen P (2009) A distinct representation of three-dimensional shape in macaque anterior intraparietal area: fast, metric and coarse, J Neurosci 29: 10613–10623.

Sugihara H, Murakami I, Shenoy KV, Andersen RA, Komatsu H. 2002 Response of MSTd neurons to simulated 3D orientation of rotating planes, J Neurophysiol 87: 273–285.

Taira M, Tsutsui KI, Jiang M Sakata H (2000) Parietal neurons represent surface orientation form the gradient of binocular disparity, J Neurophysiol 83: 3140–3146.

Theys T, Pani P, van Loon J, Goffin J, Janssen P (2012) Selectivity for three-dimensional shape and grasping-related activity in the macaque ventral premotor cortex, J Neurosci 32: 12038–12050.

Vanduffel W, Fize D, Peuskens H, Denys K, Sunaert S, Orban GA (2002) Extracting 3D from motion: differences in human and monkey intraparietal cortx, Science 298: 413–415.

Verhoef BE, Vogels R, Janssen P (2010) Contribution of inferior temporal and posterior parietal activity to three-dimesnional shape perception, Curr Biol 20: 909–913.

Xiao DK, Marcar VL, Raiguel SE, Orban GA. 1997 Selectivity of macaque MT/V5 neurons for surface orientation in depth specified by motion, Eur J Neurosci 9: 956–964.


Yadin Dudai
Konorski’s heritage: How Reactivity and Plasticity permit memory to anticipate the future

Powerful ideas often metamorphose and acquire new meaning as science advances. Jerzy Konorski (1948) recognized two metaprinciples that underlie the operation of the nervous system: Excitability (later renamed Reactivity, 1967), which is the capacity to be activated by stimulation of receptive organs, and Plasticity, which is the capacity to change reactive properties as a result of successive activations. Over half a century later, we can now attempt to evaluate, extend and translate these metaprinciples into distinct molecular, cellular and systems processes and mechanisms. Furthermore, new findings provide an opportunity to integrate these metaprinciples to propose how the brain, while reacting to the world, also anticipates it to render plasticity more parsimonious and reaction more efficient. This emerging dynamic and multilevel model of the brain impacts our understanding of the stability of memory, its veracity, and, furthermore, the biological bases and phylogenetic advantage of mental time travel to the future, i.e. imagination.


Wulf Eckhard Haubensak
Emotional mechanics in limbic circuits

Emotions motivate behaviors necessary for survival, are central to our mental self and linked to a variety of psychiatric conditions. Not surprisingly, there has been a longstanding interest in deconstructing how emotions are wired in the brain. However, due to the brains’ internal complexity, this turned out to be a formidable challenge: while the limbic system components have been known as major emotion centers for quite some time, it proved much more difficult to resolve the mechanics of information processing at the level of individual neural circuits within these structures. In a foray into this problem, we use circuit genetic technologies to dissect circuits in the extended amygdala, a key forebrain emotion relay. Combining viral tracing, pharmaco-, optogenetics and electrophysiological recordings we are characterizing a local inhibitory network of two antagonistic neuronal populations in the lateral central amygdala (CEl). Results from combined pharmacogenetics and in vivo electrophysiological recordings suggest this network alternates between two states: in the absence of a conditioned stimulus (CS), so called CEl-off neurons, identified by the expression of PKCδ are active, inhibiting their counterpart CEl-on neurons and amygdala output; in the presence of the CS, CEl-on neurons are active, inhibiting CEl-off neurons, which disinhibits amygdala output and fear signals to the brain stem. Tracing experiments suggest that this network integrates bottom-up and top-down signals from subcortical and cortical inputs, respectively. Optogenetic manipulation of the two major CEl cell types, or their subcortical and cortical afferents, inversely controls avoidance and approach behaviors. Collectively, these data suggests that CE operates like a bi-stable switch which encodes fear vs. reward in opposing circuit states. This work sheds light on the circuit organization of emotions and behavior and the functional design of inhibitory networks.


Fig. 1. Proposed model of inhibitory gating in CE. CEl integrates cortical and subcortical inputs and encodes fear vs. reward in opposing circuit states. Solid lines, direct connection. Dashed lines, direct or indirect connection.


Ewelina Knapska
Social modulation of learning

Neuronal plasticity, a notion put forward by Jerzy Konorski, is very important for understanding changes in animal behavior, from the simplest responses to complicated interactions in social groups. It is well known that social behaviors are regulated, among other factors, by emotions and that the emotion displayed by a conspecific influences the behavior of other animals. However, the mechanisms by which the animals that are not directly endangered share emotions remain largely unknown. We developed a model of between-subject transfer of fear (Fig. 1), in which a naive rat interacts with its cage-mate, which was subjected to fear conditioning. In this models, the rats were kept in pairs and one animal („demonstrator”) was treated to specific behavioral training of either foot-shock-reinforced context conditioning or just exposure to a novel context. The influence of the demonstrators on the brain activation and learning of their cagemates (called „observers”) were examined. Within this model of socially transferred fear, it was shown that several brain structures of the observers mirror the activity of the demonstrators’ brains (as revealed by c-Fos mapping), and that the observers have most of their amygdala and prefrontal cortex activated to the same level as the demonstrators and, in the case of the central amygdala, to an even higher level. In further development of the model, it was also demonstrated that a brief social interaction with a cage mate that has undergone an aversive learning experience promotes aversive learning in an otherwise naive animal. The reasoning that led to this conclusion is based on the following observations. First, during social interaction with a recently fear-conditioned partner, observers and demonstrators exhibit social exploratory behaviors rather than aggressive behaviors. Second, learning and memory in a shock-motivated shuttle avoidance task are facilitated in rats that underwent a social interaction with a partner that had been fear conditioned. Finally, a brief social interaction with a recently fear-conditioned partner immediately before fear conditioning increases conditioned freezing measured on the next day. In subsequent set of experiments, differences in both activation patterns within the limbic system and modulation of learning depending on the sex and estrus cycle phase were observed. In yet another paradigm, the IntelliCage automatic behavioral assessment system was used to study social interactions in more natural settings. The IntelliCage allows studying behavior of mice living in social groups over long periods of time, thus reducing social isolation and experimenter’s exposure stress. In two different experimental settings we showed that the mice are capable of emotional information transfer. It was shown that the presence of a fearful cage mate results in a robust renewal of avoidance responses that were previously successfully extinguished. Moreover, it was observed that mice with spatial memory deficit were capable to learn normally when co-housed with the well-performing littermates, the result suggesting socially transferred information. The reported results are deeply rooted in the scientific heritage of Jerzy Konorski. His views about neuronal plasticity and instrumental reflexes still provide a useful guidance in planning, executing and interpreting the experiments. Of course, as knowledge is accumulated, the detailed research problems change. The reported results constitute an example how classic scientific concepts (Konorski’s neuronal plasticity) conquer new research territories (social interactions) offering a logical framework for the incoming data (social modulation of neuronal plasticity).

Figure 2

Fig. 2. Model of socially transferred fear. Demonstrators were subjected to fear conditioning (A). Subsequently, emotionally aroused demonstrators interacted with their cagemates (observers), whose brain activation and learning were examined (drawings by Tomasz Werka).


Stephen Maren
Brain circuits for contextual control of fear

While it is generally adaptive to rapidly learn about threats in the environment, this form of learning can lead to psychopathology including post-traumatic stress disorder (PTSD). In the clinic, exposure therapy is an effective method for suppressing pathological fear, but relief can be transient and prone to relapse. Recent work from my laboratory has explored the neural mechanisms underlying fear relapse after extinction, a form of learning that models exposure therapy in humans. Interestingly, extinction memories are labile and fear relapses upon the passage of time and changes in context. The return of fear after extinction is consistent with Konorski’s proposal that extinction results in a new inhibitory memory that is formed alongside the excitatory fear memory. We have now identified a network of brain structures in the rat including the amygdala, hippocampus, and prefrontal cortex that contribute to regulation of fear responses after extinction. In particular, we show using electrophysiological and cellular imaging approaches that reciprocal hippocampal-prefrontal circuits control fear output by regulating amygdala neurons involved in fear expression.


Opening of the symposium and delivery of the Nencki Award to Bogdan Dreher



Round Table Memoirs by Konorski’s friends and pupils