You are on page 1of 6

This article appeared in a journal published by Elsevier.

The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
Author's personal copy

Neurocomputing 71 (2008) 3367–3371

Contents lists available at ScienceDirect

journal homepage:

Network properties of a model for conscious and unconscious

mental processes$
Roseli S. Wedemann a,, Luı́s Alfredo Vidal de Carvalho b, Raul Donangelo c
Instituto de Matemática e Estatı́stica, Universidade do Estado do Rio de Janeiro, R. São Francisco Xavier, 524, 20550-013 Rio de Janeiro, Brazil
Programa de Eng. Sistemas e Computac- ão - COPPE, Universidade Federal do Rio de Janeiro, Caixa Postal 68511, 21945-970 Rio de Janeiro, RJ, Brazil
Instituto de Fı́sica, Universidade Federal do Rio de Janeiro, Caixa Postal 68528, 21941-972 Rio de Janeiro, RJ, Brazil

a r t i c l e in fo abstract

Available online 15 July 2008 We have previously described the mental pathology known as neurosis, in terms of its relation to
Keywords: memory function. We proposed neural network mechanisms, whereby neurotic behavior is described as
Consciousness and unconsciousness a brain associative memory process, and the symbolic associativity involved in psychoanalytic working-
Neuroses through can be mapped onto a corresponding network reconfiguration process. Microscopic
Hierarchical memory mechanisms that control synaptic properties self-organize the memory networks to a hierarchical,
Self-organized learning clustered structure. Modules corresponding to sensorial and symbolic memories interact, representing
Complex neural network properties unconscious and conscious mental processes. Here, we review these concepts and illustrate, with
simulations, some of these complex networks’ behaviors and properties.
& 2008 Elsevier B.V. All rights reserved.

1. Introduction able to obtain relief and cure of painful symptoms through a

mechanism called working-through. This procedure aims at
Although inconclusive, psychodynamic theories [7,9,10,17] developing knowledge regarding the causes of symptoms by
seem to suggest correlations between creativity, associativity, accessing unconscious memories, and understanding and chan-
psychopathology and the unconscious. We explored these com- ging the analysand’s compulsion to repeat [10]. The technique
monalities and proposed, in a previous paper [22], a schematic involves mainly free associative talking during analytic sessions
model for some concepts related with neurotic mental processes, and interpretation of dreams, among other processes.
as described by Freud [7,9,10]. Our description is based on the In [22], memory was modeled by a Boltzmann machine (BM)
current view that the brain is a cognitive system composed of represented by a complete graph. It is known, however, that brain
neurons, interconnected by a network of synapses, that cooperate neuronal topology is selectively structured. Neurons interact
among themselves to process information in a distributed fashion. mainly with spatially close neighbors, having fewer long-range
Mental states are thus the result of the global cooperation of the synaptic connections to more distant neighbors [5,12,14]. We
brain’s distributed neural activity [5,13,14]. The emergence of a further developed the memory model by including some known
global state of the brain’s neural network generates a bodily microscopic mechanisms that control synaptic properties, so that
response, which we call an act. the network self-organizes to a hierarchical, clustered structure.
Psychoanalytic research regarding the transference neuroses has We propose an organization, where two hierarchically structured
found that traumatic and repressed memories are knowledge modules corresponding to sensorial and symbolic memories
which is present in the subject, but which is symbolically interact, producing sensorial and symbolic activity, representing
inaccessible to him. It is therefore considered unconscious knowl- unconscious and conscious mental processes. In proposing this
edge [7,9]. Freud observed that neurotic patients systematically organization, we followed Freud’s idea that unconscious mem-
repeated symptoms in the form of ideas and impulses, and called ories are those which we cannot access symbolically, i.e. cannot
this tendency a compulsion to repeat [10]. He related the talk about [7–10]. In this paper, we represent brain mechanisms
compulsion to repeat to repressed or traumatic memory traces, involved in neurosis as a complex system, and analyze them
caused by a mental conflict [9]. Neurotic analysands have been according to recent methods developed for the study of complex
A review of recent developments in the scientific under-
This research was developed with grants from the Brazilian National Research standing of consciousness, as well as a model for attention as a
Council (CNPq), the Rio de Janeiro State Research Foundation (FAPERJ) and the basic function related to conscious activity may be found in [18].
Brazilian agency which funds graduate studies (CAPES).
 Corresponding author. Tel.: +55 2122382180; fax: +55 2125877212. Kinsbourne [15] discusses how Freud’s attempt at proposing a
E-mail addresses: (R.S. Wedemann), neural substrate for mental processes [8] can be viewed in light of
(L.A.V. de Carvalho), (R. Donangelo). modern developments in neuroscience, such as the understanding

0925-2312/$ - see front matter & 2008 Elsevier B.V. All rights reserved.
Author's personal copy

3368 R.S. Wedemann et al. / Neurocomputing 71 (2008) 3367–3371

of forebrain functioning. Aleksander and Morton review their 3. Self-organizing hierarchical memory structures
Axiomatic Consciousness Theory (ACT) in [1], which addresses
phenomenology in neuroscience (relating symbolic representa- As a first approximation [22], we modeled memory by a BM
tion to subjective experience), and apply it to modeling of visual represented by a complete graph. In a further refinement, we now
phenomenology. The Global Workspace Theory [2] is based on the consider that neurons belong to two different subsets, corre-
computational science concept of global workspace, where sponding to sensorial and symbolic memories. Memory traces
material needed to be worked on by a number of processors is stored in sensorial memory represent mental images of stimuli
held, and gives an useful view on certain aspects of consciousness. received by sensory receptors (located in eyes, ears, skin and other
Our main contribution with respect to current work regarding parts of the body) from the environment and the body itself. This
machine models of consciousness is to propose a neuronal includes information regarding affects and emotion. Sensorial
mechanism that describes conscious and unconscious memory memory represents brain structures such as the amygdala,
activity involved in neurosis. The unconscious compulsion to cerebellum, reflex pathways, hippocampus, and prefrontal, limbic
repeat is explained as a neural network associative memory and parieto-occipital-temporal cortices, which synthesize visual,
mechanism, where an input stimulus of any kind associates to a auditory and somatic information. Symbolic memory stores
pattern in sensorial memory, which cannot activate symbolic representations of traces in sensorial memory, i.e. symbols, and
brain processing areas. Neurotic (unconscious) acts are isolated refers to another level of representation. It represents brain
from symbolic representation and association (similar to reflexes). structures such as areas of the medial temporal lobe, the
We illustrate by our network model, how Freud’s ideas regarding hippocampus, Wernicke’s and Broca’s areas and other areas of
the unconscious show that symbolic processing, language and the frontal cortex. These latter areas are associated with language
meaning are essential for consciousness. and, because of them, we can associate a word such as ‘‘red’’ to the
We first review our model for neurosis [22] and the algorithm sensation of seeing a red object. Attentional mechanisms [18],
for generating hierarchically clustered memory modules [20]. We which we did not model, select stimuli to sensorial memory and
then present new simulation results and model analysis, with allow them to become conscious, if associated to a symbolic
some important properties of these complex networks. Finally, we memory trace. Sensorial stimuli which cannot associate to
draw conclusions and perspectives for future work. symbolic memory remains unconscious [7,9,14]. We refer to the
subset of memory traces in sensorial memory which can associate
to a symbol as the ‘‘conscious’’ subset. Freud called this the system
2. Functional and computational model for the neuroses conscious and defines it as the set of psychical information which
is capable of becoming conscious (in some of his works he refers
We review the model described in previous work [22], where we to this set as the preconscious) [7,9].
proposed that the neuroses manifest themselves as an associative The neurotic compulsion to repeat [9,10] is explained here as a
memory process, a mechanism where the network returns a stored bodily response (an act) to an access to sensorial memory, which
pattern, when it is shown another input pattern similar to the stored does not activate symbolic memory, as in a reflex. This accounts
one [13]. We modeled the compulsion to repeat neurotic symptoms, for the fact that neurotics say they cannot explain their neurotic
by supposing that such a symptom is acted, when the subject is acts, such as in ‘‘I don’t know why I was so upset and overreacted
presented with a stimulus, which resembles a repressed or with Mary’’. When a stimulus S 0 which causes the retrieval of a
traumatic (unconscious) memory trace. The stimulus causes a sensorial memory trace can activate also a pattern in symbolic
stabilization of the neural net onto a minimal energy state, memory, it can become conscious, the output is not as in reflexive
corresponding to the memory trace that synthesizes the original behavior and there is another level of processing. This mechanism
repressed experience, which in turn generates a neurotic response is similar to ideas proposed by Edelman [5] and strongly reflects
(an act). The neurotic act is not a result of the stimulus as a new Freud’s concepts of conscious and unconscious mental processes
situation but a response to the repressed memory. and the role of language in psychoanalysis [9,10].
We mapped the linguistic, symbolic associative process In order to model the organization of the topology of each
involved in psychoanalytic working-through into a corresponding memory, we consider the following microscopic biological
process of reinforcing synapses among memory traces in the mechanisms. Brain cells in many animals have a structure called
brain. These connections should involve symbolic memory, on-center/off-surround, in which a neuron is in cooperation,
leading to at least partial transformation of repressed memory through excitatory synapses, with other neurons in its immediate
to consciousness. This has a relation to the importance of neighborhood, whereas it is in competition with neurons that lay
language in psychoanalysis and the idea that unconscious outside these surroundings. These mechanisms are found stati-
memories are those that cannot be expressed symbolically. We cally hardwired, and also as part of dynamical processes, where
propose that as the analysand symbolically elaborates manifesta- neurons compete for certain chemicals [12,14]. In synaptogenesis,
tions of unconscious material through transference in psycho- for example, substances generically called neural growth factors
analytic sessions, he is reconfiguring the topology of his neural are released by stimulated neurons and, spreading through
net, by creating new connections and reinforcing or inhibiting diffusion, reach neighboring cells, promoting synaptic growth.
older ones. The network topology which results from this Cells that receive neural growth factors make synapses and live,
reconfiguration will stabilize onto new energy minima, associated while cells that have no contact with these substances die [14].
with new acts. The process of slowly and repeatedly reconfiguring A neuron releasing neural growth factor guides the synaptic
synaptic connections to elaborate knowledge accounts for the formation in its tri-dimensional neighborhood, becoming a center
long durations of psychoanalytic processes, where repetition is of synaptic convergence. When neighboring neurons release
specially important. different neural growth factors in different amounts, many
Memory functioning is modeled by a BM with N neurons, synaptic convergence centers are generated and a competition is
where node states take binary values. Pattern retrieval on the net established between them, through the surrounding synapses. A
is achieved by a standard simulated annealing process, in which signaling network triggered by environmental stimulation is
the network temperature T is gradually lowered by a factor a. We established, which controls development and plasticity of neuro-
refer the reader to Ref. [13] for a description of the BM algorithm nal circuits, and this suggests an idea of the way environment
and its properties. represents itself in the brain. We thus developed the following
Author's personal copy

R.S. Wedemann et al. / Neurocomputing 71 (2008) 3367–3371 3369

clustering algorithm, to model the self-organizing process which allows stabilizing onto the many local minima of the network
results in a structured topology of each memory. energy function. These initially stored patterns, associated as they
Step 1: Neurons are uniformly distributed in a square bi- are to two weakly linked subnetworks, represent neurotic
dimensional sheet. memory states.
Step 2: To avoid the unnecessary and time-consuming To simulate working-through, we stimulate the net by means
numerical solution of the diffusion equation of the neural growth of a state change in a randomly chosen ni belonging to the
factors, we assume a Gaussian solution. Therefore, a synapse is ‘‘unconscious’’ (sensorial) set of a neurotic pattern. This stimulus
allocated to connect a neuron ni to a neuron nj according to a is then presented to the network and, if the BM retrieves a pattern
Gaussian probability, given by P ij / expððr~j  r~i Þ2 =2s2 Þ, where r~j with symbolic configuration different from that of the neurotic
and r~i are the positions of nj and ni , respectively, in the pattern, we interpret this as a new conscious association, and
bi-dimensional sheet and s is the standard deviation of the enhance all weights from ni to the changed nodes in the symbolic
distribution, which is considered here a model parameter. If a set. Following Hebbian learning, increment values are propor-
synapse is allocated to connect ni and nj , its strength wij is tional to the products of the states of the connected neurons and
proportional to Pij . Synaptic weights are symmetrical, wij ¼ wji . the learning parameter b. New knowledge is learned only when
Step 3: We verified in previous work [4,21] that cortical maps the stimulus from the analyst is not similar to the neurotic
representing different stimuli are formed, such that each stimulus memory trace. This procedure must be repeated for various
activates a group of neurons spatially close to each other, and that reinforcement iterations in an adaptive learning process, and also
these groups are uniformly distributed along the memory sheet. each set of reinforcement iterations must be repeated for various
We thus randomly choose m neurons which will each be a center initial temperature values.
of the representation of a stimulus. The value of m is chosen
considering the storage capacity of the BM [13].
Step 4: For each of the m centers chosen in Step 3, reinforce 4. Simulation illustrations and network behavior
adjacent synapses according to the following criteria. If ni is a
center, define sumni ¼ j jwij j, where wij is the weight of the We illustrate the model with a simulation for a network with
synapse connecting nj to ni . For each nj adjacent to ni , increase N ¼ 50 neurons, such that Nsens ¼ Nsymb ¼ 25 belong to each
jwij j by Dwij, with probability Probnj ¼ jwij j=sumni , where Dwij ¼ memory. In the brain, memory modules are not symmetric and
Z Probnj and Z 2 R is a model parameter chosen in ½0; 1. After neurons and synapses function differently in different brain
incrementing jwij j, decrement Dwij from the weights of all the regions. We concentrate on the interactions between and within
other neighbors of ni , according to: 8kaj; jwik j ¼ jwik j  Dwik , the sensorial and symbolic modules to describe how the
where Dwik ¼ ð1  jwik j= kaj jwik jÞDwij . possibility of memory traces to become conscious is sensitive to
Step 5: Repeat Step 4 until a clustering criterion is met. the connection intensities, and how they can be reconfigured in a
In the above clustering algorithm, Steps 2 and 3 are justified in learning process to enhance or inhibit associativity. Although N is
the algorithm’s description. Step 4 regulates synaptic intensities extremely small, initially, it seems to suffice to illustrate the basic
by strengthening synapses within a cluster and reducing synaptic concepts and mechanisms at a semantic level. The short-range
strength between clusters (disconnects clusters). Neurons that microscopic mechanisms are scalable and, since our algorithms
have received stronger sensorial stimulation, and are therefore are based on microscopic biological mechanisms, we expect that
more strongly connected, will stimulate their neighborhoods and this level should be amenable to mapping to a biological
promote still stronger connections. substratum. Some simulations of system configurations have
The growth of long-range synapses is energetically more costly taken a few days on a single processor, even for these small
than short-range synaptic growth, and therefore the former are systems. We plan to parallelize these algorithms, in order to
less frequent in the brain than the latter. For allocating long-range simulate significantly larger systems.
synapses which connect clusters, we should consider the basic In our simulations, s ¼ 0:58, b ¼ 0:3, Z ¼ 0:1 and the other
learning mechanism proposed by Hebb [5,13,14] based on the fact parameters for annealing in the BM are attributed the same values
that synaptic growth among two neurons is promoted by we have used in [22]. To configure neurotic traces having weak
simultaneous stimulation of the pair. Neuronal groups whose connections to symbolic processing brain areas, synapses con-
states store certain memory traces, which receive simultaneous necting different memory modules are multiplied by k ¼ 0:5,
stimuli, should enhance synaptic growth among the neuronal defining the initially stored patterns as neurotic. For smaller
groups representing these traces, allowing association among values of k the network has learning difficulties, suggesting a
traces. Since memory traces represent both sensorial information difficulty in associating unconscious traces among themselves and
and concepts (symbolic memory), we are also representing with symbolic memory.
association of ideas or symbols by long-range synapses. In Fig. 1(a), we show one topology generated after executing
This suggests a mechanism through which the external world, only the clustering algorithm and, in Fig. 1(b), the corresponding
culture and basic language structure may be mapped onto brain topology after long-range synaptic generation. The figures
topology. Since we are still not aware of the synaptic distri- illustrate the clustered, hierarchical self-organization of the
butions that result in such topologies, as a first approximation, we network. For this initial topology, 42% of the patterns stored
allocate long-range synapses randomly among clusters. Within initially were still stored after working-through, which shows that
a cluster C, ni is chosen for the connection with probability the network adapts with the reconfiguration, freeing itself from
P i ¼ j jwij j= nj 2C k jwjk j. If the synapse connects clusters in some of the ‘‘neurotic’’ states. Figs. 2(a) and (b) give a more
different memories (sensorial and symbolic), its randomly chosen quantitative view. We generated 10 000 different neurotic topol-
weight is multiplied by a factor k 2 R in ½0; 1, reflecting the fact ogies, for different values of N and spatial dimensions, from the
that some sensorial information is weakly accessible to con- same initial system parameter values specified above, and
sciousness, i.e. repressed. measured the average node degree (k) distributions. In Fig. 2(a)
Mechanisms of memory storage and retrieval by the BM (in logarithmic scale) the discrete symbols represent the values
function as mentioned in Section 2. Once the network is found in our simulations. For larger values of N, Fig. 2(a) shows
initialized, we find the stored patterns by presenting many a power law distribution for large k, with exponent g  4.
random patterns to the BM, with an annealing schedule a that This power law behavior is characteristic of many biological
Author's personal copy

3370 R.S. Wedemann et al. / Neurocomputing 71 (2008) 3367–3371

7 positions sensorial memory 7 positions sensorial memory

positions declarative memory positions declarative memory
6 clustered synapses in memory 6 synapses connecting memories

5 5

4 4

3 3

2 2

1 1

0 0

-0.5 0 0.5 1 1.5 2 2.5 3 -0.5 0 0.5 1 1.5 2 2.5 3

x x

Fig. 1. (a) On the left, network topology after clustering for N ¼ 50. (b) On the right, network topology with long-range synapses.

10000 0.5
10**7/x**4.1 Simulation
N = 32 1/x**0.23
Average number of nodes with k links

N = 50
N = 100
Average clustering coefficient

N = 282 0.4
10 N = 1000
N = 2000
N = 4000



1e-05 0.1
1 10 100 0 1000 2000 3000 4000 5000
Number of links (k) Total number of neurons (N)

Fig. 2. (a) On the left, node degree distributions in logarithmic scale, for various values of N. (b) On the right, average clustering coefficient for different N values.

systems and indicates scale independence. It is known that through the preferential attachment criterium, for which C / 1=N
random graphs follow the Poisson distribution of node degrees [11]. To illustrate this point, we show as a full line in the graph the
lk expðlÞ=k! [16]. The deviations from Poisson in Fig. 2(a) for function 1=N 0:23 . We notice that, even if it follows the main trend
higher k values may be attributed to the competitive biological of C obtained in our simulations, this function overpredicts the
mechanisms, and confirm that the resulting networks have a drop for large N. The clustering algorithm presented in this work
structure which is not random and may be scale free [16,19]. thus appears to lead to networks, for which the biological
The average clustering coefficient C [16,19] for 10 000 networks mechanisms establish some different topological properties than
of 50 neurons is 0.38. We compare C generated by our clustering the traditional algorithms, studied in complex network theory.
algorithm to that of a real network, where we have considered the
neural network of the well-studied worm C. elegans, which is
composed of 282 neurons. This worm has measured C ¼ 0:28 [16]. 5. Conclusions
We obtained an average C ¼ 0:24 within our model over 10 000
networks, also of 282 neurons. These two values are surprisingly We have further developed a previous memory model [22] to
similar, considering that our algorithm is based on very general include microscopic biological neuronal mechanisms, and verified
microscopic biological mechanisms. This worm’s network would that the memory complex network self-organizes into a hier-
present C ¼ 0:049, if it were a random graph [16], and the higher archical clustered structure. Two hierarchically structured mod-
measured value indicates the existence of network structure. ules corresponding to sensorial and symbolic memories interact,
In Fig. 2(b) crosses depict the calculated average C values, for producing sensorial and symbolic activity, representing uncon-
the initial network configurations, as a function of N. As it is well scious and conscious mental processes. This memory structure
known, C is negligible in the case of an Erdös–Rényi random and functioning, with an adaptive learning process, is used to
network. We also notice that the dependence for larger N is much explain a possible mechanism for neurotic behavior and psycho-
weaker than that for a network in which linkage is performed analytical working-through.
Author's personal copy

R.S. Wedemann et al. / Neurocomputing 71 (2008) 3367–3371 3371

The model is in agreement with psychoanalytic experience [13] J.A. Hertz, A. Krogh, R.G. Palmer (Eds.), Introduction to the Theory of Neural
that working-through is a slow process, where the individual Computation, Lecture Notes, vol. I, Santa Fe Institute, Studies in the Science of
Complexity, Perseus Books, Cambridge, MA, 1991.
slowly elaborates knowledge by re-associating weakly connected [14] E.R. Kandel, J.H. Schwartz, T.M. Jessel (Eds.), Principles of Neural Science,
memory traces and new experiences. It thus illustrates a McGraw-Hill, USA, 2000.
phenomenological view of mind, with intentionality [3]. Language [15] M. Kinsbourne, Taking the project seriously: the unconscious in neuroscience
perspective, Ann. NY Acad. Sci. 843 (1) (1998) 111–115.
is an important process, and is represented by the functioning of [16] M.E.J. Newman, Random graphs as models of networks, in: S. Bornholdt, et al.
symbolic memory. Repetition is also important, and is illustrated (Eds.), Handbook of Graphs and Networks, from the Genome to the Internet,
in the model by the repetitive adaptive reinforcement learning, in Wiley-VCH, USA, 2003, pp. 35–68.
[17] M. Spitzer, A cognitive neuroscience view of schizophrenic thought disorder,
the simulation of working-through. From the simulations, we saw Schizophr. Bull. 23 (1) (1997) 29–50.
that the working-through when understood as a network [18] J.G. Taylor, Mind and consciousness: Towards a final answer?, Phys. Life Rev. 2
reconfiguration process, partially frees the analysand from the (2005) 1–45.
[19] D.J. Watts, S.H. Strogatz, Collective dynamics of ‘small world’ networks,
original neurotic states and respective acts. The new network
Nature 393 (1998) 440–442.
topology which results from this reconfiguration process will [20] R.S. Wedemann, L.A.V. Carvalho, R. Donangelo, A complex neural network
stabilize onto new energy minima, associated with new conscious model for memory functioning in psychopathology, in: Proceedings of ICANN
or unconscious acts. It is the purpose of the analysand and the 2006, Lecture Notes in Computer Science, vol. 4131, Springer, Berlin, 2006, pp.
analyst that these new minima are reached and generate acts [21] R.S. Wedemann, L.A.V. Carvalho, R. Donangelo, Complex Networks in
which are more pleasant and comfortable to the analysand. The Psychological Models, Progress of Theoretical Physics Supplement, vol. 162,
analysand is rewriting his life history, i.e. his past, through a Yukawa Institute for Theoretical Physics, The Physical Society of Japan, Kyoto,
2006, pp. 121–130.
process of new significations, and his present and future by [22] R.S. Wedemann, R. Donangelo, L.A.V. Carvalho, I.H. Martins, Memory
creating new possibilities [6]. This new story is now embedded, functioning in psychopathology, in: Proceedings of ICCS 2002, Lecture Notes
i.e. written, in his biological neural network. in Computer Science, vol. 2329, Springer, Berlin, 2002, pp. 236–245.
We are proceeding in further model refinement and analysis. It is
Roseli S. Wedemann received her undergraduate
necessary to verify more deeply the dependence of model behavior degree in Physics in 1985, her Master of Science degree
on parameters such as T, s and k. These parameters represent, at in Physics in 1988, and her Ph.D. in Systems Engineer-
least partially, the effects of neurosubstances such as neural growth ing and Computer Science in 1995, at the Federal
University of Rio de Janeiro (UFRJ). She has been
factors and neuromodulators. Although we do not yet have
visiting professor at the Institute of Physics at UFRJ in
experimental indications of their values, tuning their relative values 1996 and is currently professor at the Computer
is fundamental for model stability and functioning. This can give Science Department of the Institute of Mathematics
of the State University of Rio de Janeiro (UERJ). She has
some insight to basic mechanisms in real neural networks and the
been active in the research group in Computational
emergence of behavior. We are also interested in trying to map Neuroscience with Prof. Luı́s Alfredo Vidal de Carvalho
language structure and processing into network topology and since 2000. She is referee of the Brazilian National
dynamics, although we are not yet sure whether this is possible. Research Council since 2001 and of the Rio de Janeiro
State Research Foundation since 2002. Her main research interests include
Although biologically plausible, in accordance with many aspects Computational Neuroscience, Neural Networks, Artificial Intelligence, Distributed
described by psychoanalytic clinical experience, and experimentally Algorithms, Computational Science, Physics of Complex Systems and High
based on simulations, the model is very schematic. We do not Performance Computing.

sustain or prove that this is the actual mechanism that occurs in the
Luı́s Alfredo Vidal de Carvalho received his under-
human brain. It nevertheless seems to be a good metaphorical view graduate degree in Mechanical Engineering in 1982, his
of facets of mental phenomena, for which we seek a neuronal Master of Science degree in Optimization in 1984, and
substratum, and suggests directions of search. Our investigations his Ph.D. in Systems Engineering and Computer
Science in 1989, at the Federal University of Rio de
strongly indicate the importance of the connection of symbolic Janeiro (UFRJ). He has acted as Chairman of the
processing, meaning and language for consciousness. Graduate Program of Systems Engineering and Com-
puter Science at UFRJ in 1995–1996 and 2000–2001,
where he is full professor. He is currently Chairman of
References the Interdisciplinary Graduate Program of History of
Science and Epistemology at UFRJ, since 2000. He has
[1] I. Aleksander, H. Morton, Phenomenology and digital neural architectures, received the Latin America Biological Psychiatry Award
Neural Networks 20 (2007) 932–937. in 1998, awarded by The World Biological Psychiatry
[2] B. Baars, In the theatre of consciousness: global workspace theory, a rigorous Society, The Latin America Biological Psychiatry Society and Lilly - Neuroscience
scientific theory of consciousness, J. Consciousness Stud. 4 (4) (1997) 292–309. Pharmaceutical Laboratory, and also the IEEE - New Educational Technologies
[3] F. Brentano, Psychology from an Empirical Standpoint, Routledge, London, Award in 2000. He has lead the main research group in Computational
1995 (German ed. in 1874). Neuroscience in UFRJ since 1993. He is member of the New York Academy of
[4] L.A.V. Carvalho, D.Q. Mendes, R.S. Wedemann, Creativity and delusions: the Sciences since 1995, and referee of the Brazilian National Research Council since
dopaminergic modulation of cortical maps, in: Proceedings of ICCS 2003, Lecture 1992. His main research interests include Computational Neuroscience, Neural
Notes in Computer Science, vol. 2657, Springer, Berlin, 2003, pp. 511–520. Networks, Artificial Intelligence, Data Mining and History of Neuroscience and
[5] G.M. Edelman, Wider than the Sky, a Revolutionary View of Consciousness, Epistemology.
Penguin Books, London, 2005.
[6] J. Forrester, The Seductions of Psychoanalysis, Freud, Lacan and Derrida, Raul J. Donangelo is born in Montevideo, Uruguay in
Cambridge University Press, UK, 1990. 1947. He graduated as an industrial engineer at the
[7] S. Freud, The Unconscious, Standard Edition, vol. XIV, The Hogarth Press, Universidad de La República, Montevideo, Uruguay, in
London, 1957 (First German ed. in 1915). 1971. He received a Ph.D. in Physics from the University
[8] S. Freud, Project for a Scientific Psychology, Standard Edition, vol. I, The of California at Berkeley in 1977. He has worked at the
Hogarth Press, London, 1966 (First German ed. 1950). Instituto de Fı́sica da Universidade Federal do Rio de
[9] S. Freud, Introductory Lectures on Psycho-Analysis, Standard Edition, W. W. Janeiro since 1978, and is presently working as a full
Norton & Company, New York, London, 1966 (First German ed. 1917). professor. His main areas of research are theoretical
[10] S. Freud, Beyond the Pleasure Principle, Standard Edition, vol. XI, The Hogarth nuclear physics and physics of complex systems.
Press, London, 1974 (First German ed. 1920).
[11] A. Fronczak, J.A. Holyst, M. Jedynak, J. Sienkiewicz, Higher order clustering
coefficients in Barabási-Albert networks, Physica A 316 (1–4) (2002) 688–694.
[12] W.F. Ganong, Review of Medical Physiology, McGraw-Hill, USA, 2003.