Gli atti del congresso saranno pubblicati da Kluwer / Proceedings of the Conference will be published by Kluwer.



Fabio Badalamenti, Giovanni D’Anna, Carlo Pipitone,

Marine Biology Lab., IRMA-CNR, via G. da Verrazzano, 17, 91014 - Castellammare del Golfo, Italy - {fbadala, danna, pipitone};

Salvatore Di Gregorio, Giuseppe A. Trunfio

Department of Mathematics, University of Calabria, Arcavacata, 87036 Rende, Italy -


A quite general Cellular Automaton is in the course of development to model and simulate the behavior and interaction of several fish species in the area of the Gulf of Castellammare in NW Sicily. The model, based on the definition of local interactions that generate global behavior’ is devoted to account for relevant changes in the species distribution after a long period without trawling and the effects for a re-introduction of fishing with bottom trawl nets. The model relies on a vast amount of survey data available over the sea species living in the Gulf.

This paper describes the part of the model concerning the red mullet behavior. First results show that this model part captures some important mechanisms by the definition of few local rules, so that large-scale phenomena, as red mullet concentration and moving, emerge.

KEYWORDS: Cellular Automata, trawling environmental impact, red mullet, fish behavior, Gulf of Castellammare



Erminia Bendicenti, Salvatore Di Gregorio, Angela Iezzi

Dept. of Mathematics, Univ. of Calabria, Arcavacata, 87036 Rende, Italy (

Francesco M. Falbo,

Comando Provinciale Vigili del Fuoco CS, viale Repubblica 56, 87100 Cosenza, Italy


Forest fires represent a serious environmental problem, whose negative impact is becoming day by day more worrisome.

Forest fires are very complex phenomena; that need an interdisciplinary approach. The adopted method to modeling involves the definition of local rules, from which the global behavior of the system can emerge. The paradigm of Cellular Automata was applied and the model ABBAMPAU was projected to simulate the evolution of forest fires.

First simulations account for the main characteristics of the phenomenon and agree with the observations. The results show that the model could be applied for the forest fire preventions, the productions of risk scenarios and the evaluation of the forest fire environmental impact.

KEYWORDS: Forest Fire, Cellular Automata, Complex Systems.



Gianfranco Minati,

Associazione Italiana per le Ricerca sui Sistemi (AIRS), 42 Viale P. Rossi, 20161 Milano, Italy

Sabre Brahms, M.A.; 368 Museum Drive, Los Angeles, CA, 90065, USA; Consulting Faculty, Phillips Graduate Institute, Los Angeles, CA, USA; Doctoral Student, Saybrook Graduate School, San Francisco, CA, USA


The concept of DYnamic uSAge of Models (DYSAM) is introduced.  It applies to modeling systems having dynamic behavior like having different systems emerging from the behavior of the same agents: they cause the arise of multiple systems (Collective Beings). Such a kind of emergence takes places when agents provided with Cognitive Systems are also provided with the same Cognitive Model. The topic is really in the Systems Science and Cognitive Science dominion. The DYSAM approach is useful to model systems changing not only in time as usual for dynamic systems, but in structure.

Because the observer may also be interested in different aspects of the same phenomena, of the same (non-multiple) system, with no reference to agents having or not Cognitive System, it is important the availability of different kinds of models: continuously changing systems, continuously changing observers and continuously changing interests of the same observer are conceptually the same problem.

Distinctions must be made between (1) continuously changing observers (or users) and (2) observers (users) asking for heuristic-generating models converging to the more effective one.

A general schema of DYSAM is introduced.

The case of application of this approach to handwritten digits automatic recognition is discussed.

Cases of application of the DYSAM approach to social systems are also discussed. The systems to be modeled in such a cases are the users.

In many applications having communication and interaction problems, the approach to deal with different kinds of unexpected users is only based on standardizing and on inducing standardization. The case of simultaneous, systemic use of different models to model the receiver in communication processes is discussed with reference to modeling in education. Applications to marriage and family therapy, educational activities,  E-books and  Virtual business (supermarket, banking, telecommunication) technologies are discussed.

KEYWORDS: agents, collective beings, dynamic modeling, emergence, multiple systems



Gianfranco Minati

Associazione Italiana per le Ricerca sui Sistemi (AIRS), 42 Viale P. Rossi, 20161 Milano, Italy


In the framework of modeling Collective Behaviors a new conceptual framework is proposed in this working paper based on the concept of Ergodicity.

A discussion about the applicability of concepts related to set theory to systems is introduced in order to explore the possibility to apply the concept of ergodicity, originally introduced in statistical mechanics with reference to sets, to systems.

After introducing the concept of Ergodicity as presented in different disciplines, in Physics where the concept has been invented, in Theoretical Geomorphology and in Economics, the concept is considered as basis to model collective behaviors emerging from agents provided with cognitive systems, for instance, flocks, swarms, fish schools.

A distinction between emergence processes established by agents interacting in (1) organized, structured  way and at (2) different dynamic levels of ergodicity is discussed proposing as a key point the levels of ergodicity and their dynamics.

The purpose of this contribution focuses on the possibility to identify an indicator of ergodicity able to detect the level of ergodicity of a process or set in the considered time-space, making the observer able

·         to make correspondent sets, systems or processes of very different kind having the same ergodicity or having similar ergodic evolution in time; and

·         to detect the dynamics of different dynamic levels of ergodicity assumed by interacting agents establishing emergent systems.

The hypothesis introduced is that the continuous changing of values of the indexes of ergodicity are, in case of emergence processes, in a “confidence interval”. The values of the indexes of ergodicity assume constant values when the system emerges from a structure, an organization. The confidence interval and the dynamics of changing of the assumed values may possibly correspond to the emergence of some different dynamic systems from collective behaviors, such as flocks, swarms, herds.

The differences between these two cases must be taken in count when considering the opportunity to manage, keep, start, modulate, control and replicate processes of emergence.

In the last part some possible lines of research on emergence, Collective Behaviors and ergodicity are introduced. Special reference is to the possibility to detect the level of ergodicity assumed by the process or set in time and to make correspondent sets, systems or processes of very different kind having similar ergodicity: that’s make simulation important for any kind of system.

Different possible lines of research and related possible modeling and simulation projects are mentioned.

Keywords: collective behavior, emergence, ergodic, indicator of ergodicity, model.



Mario R. Abram


CESI S.p.A - Centro Elettrotecnico Sperimentale Italiano,

Via Rubattino, 54 I-20134 Milano, Italy. URL: E-mail: Tel. +39 02 2125.5737 Fax. +39 02 2125.5813


The decomposition of a system into subsystems is not unique and is related to the goals that one chooses to reach.

The development of a methodology to evidence the relations between the subsystems can be useful to investigate the various available decompositions of a system in order to identify the best one before starting the detailed analysis involved in a design process.

Once a decomposition is fixed, each relation between  the subsystems can be described and formalized using the various analytic methods available from the different fields of science and technology.

This decomposition methodology is qualitative and it aims to give the designer the opportunity to keep in mind a global point of view of the system while the development of a project asks for an increasing detailed analysis of each subsystem.

The methodology is applied to the preliminary analysis involving the control of an industrial process and some peculiarities and general considerations are evidenced.



Roberto Serra

Centro Ricerche Ambientali Montecatini, Ravenna



The use of random networks to model the web of interactions in complex systems dates back to several years ago. A particularly brilliant example is provided by random boolean networks, which were introduced by Stuart Kauffman to model the generic features of interacting genes.

The basic assumption underlying most models of this kind is that the connections are entirely random: the network is composed of N nodes, and there is a fixed probability p that any two nodes be connected by an edge.

This gives rise to a peaked Poisson probability distribution, p(k), for the number k of connections per node. The random graph model predicts

- that the average path length between two nodes is small (it scales with logN)

- that the clustering coefficient (which measures the fraction of nodes,

which are neighbours of the same node, which are also connected among themselves) is small.

In recent years it has become apparent that many real networks have the property that the average path length is short, but the clustering coefficient is high. Examples of these ‘small world’- networks include social networks, Internet, the World Wide Web, electric distribution networks, cellular metabolic networks, etc.

Two theoretical models (each one with several variants) have been proposed to account for these properties:

- the Watts-Strogatz model, where the presence of short paths is mainly due to the introduction of shortcuts between distant nodes

- the scale-free model, where the connectivity follows a power law distribution and there are a few highly interconnected nodes (‘hubs’) which lead to short average path length.

The major distinguishing feature among the two models is the distribution p(k) which allows one to find out, for every real world network, which model (or combination of models) provides a better approximation.



Dr . Eng . Umberto Di Caprio

Stability Analysis, Via A. Doria, 48 /A - 20124 Milano, Italy


An outline is given of the main development of stability theory over the last 50 years . Reference is done to the basic work by M. A. Lyapunov. The following topics are discussed: stability and non-linearity, stability and energy, non symmetric and multivariable systems. Stability theory represents an effective and satisfactory contribution to the union of the great theories of the twentieth century: Special Relativity, General Relativity and Quantum Mechanics. As is well known Einstein acknowledged the inability of General Relativity to unify gravitation and electromagnetism and he denied theoretical value to Quantum theory. The discovery of sub-nuclear particles (quarks) put an end to an already critical situation. As a matter of fact Quantum Chromodynamics hasn’t stood up the latest experimental results. We need a new theory, which is given by stability theory.



Paolo Camiz

Università di Roma "La Sapienza" - Department of Physics

INFN Sezione di Roma – ECONA


The behavior of a perturbative neural network (PNN) describing synaptic development in perceptive systems is critically analyzed, and the equivalent variational problems are discussed. Possible strategies for its implementation in a real living system are analyzed and some more realistic alternative version are proposed as well. Analytic solutions of updating equations are given for some simple cases.



Renza Cambini,

Organizzazione per la Preparazione Professionale degli Insegnanti (OPPI)

Associazione Italiana per la Ricerca sui Sistemi (AIRS)

Adalberto Codetta Raiteri

Organizzazione per la Preparazione Professionale degli Insegnanti (OPPI) Via Orseolo, 4  20100-Milano.


Educational systems are receiving special attention by governments and social designers in all the so called advanced countries. Education is more and more emerging as a strategic resource in the international economic scenario.

The  need to professionally train the young generations in order to make them able to compete in the global market asks for having goal oriented, comparable and measurable educational activities.

English speaking countries particularly have long experiences in evaluating by using procedures based on tests. They also experienced procedures to evacuate single schools and educational systems. Numerical rankings produced by such educational testing procedures (also called “test machineries”) have great influence in decision making activities because of their role of objective synthesis.  This aspect reinforce the importance of such an approach.

The international usage of educational testing procedures may have a role in making generally accepted, if not standardized, educational systems approaches similarly to what is already established in the international business community (quality, industrial standards, information to the user, trade procedures, etc.). Quality certification in education has difficulties to take off: it is intended not as an opportunity, but as an unavoidable procedure, for instance applying for founding (quality certification is one of the items setting the global evaluation when submitting requests to the European Social Found). Besides the need to make students able to change course of studies and kind of school is a strong stimulus to validate different educational systems thanks to equivalence among different educational qualifications.

The adoption of evaluation systems from other cultural contexts without any adapting processing has the risk to weak or destroy cultural specific properties able to make different communities able to compete and cooperate in the global market. Therefore the attention to cultural identities is matter of effectiveness and quality for the evaluator.

The selection of questions and topics, the way of wording, unavoidably related to cultural, racial, religious, sexual issues privilege particular aspects of some of the many ways in which the human intelligence is expressed.

This consideration introduces the awareness that any kind of evaluation, even trough testing procedures, does not assume the characteristic of objective measurement. Moreover even natural sciences (physics first of all) are not based anymore on the objectivity principle only. In the case of education any evaluation has meaning in a particular context, locally defined by relational processes established by the process of evaluating.

To deal with the just mentioned problem OPPI set up a well-tested and many times readjusted course, also focused on the usage of information systems solutions.  This course has the purpose to make available supports to design, to give multiple answer tests, able to manage interactions generated by the evaluation process itself.

In the course held at the end of August 2001 an external evaluator has been required to attend, evaluating the consistency of purposes and activities. It has also been asked for collecting the evaluations expressed by the attendants. 

The external evaluator decided to prepare tools and supports to carry out the constructivist approach. The goal was to introduce new epistemological issues.

The idea is that each approach changes nature and borders if considered from the perspective of another one. The different approaches must be considered not alternative, but as a system of solutions to be used depending on contents, objectives, timing, contexts, in which the evaluation process takes place



C. Palma

Istituto Tecnico Industrile Statale “CARTESIO” , Elettronica e Telecomunicazioni, Informatica-Abacus, Liceo Scientifico Tecnologico - 120, via C. Lombroso, Roma, Italy

Cattedra di Intelligenza Artificiale. I° Facoltà di Psicologia Università degli Studi di Roma  “La Sapienza”, Via dei Marsi  78, 00185  Rome, Italy. e-mail:

Associazione Italiana per la Ricerca sui Sistemi (AIRS) Italian division of the International Society for the Systems Sciences (ISSS), USA – 42, via Pellegrino Rossi – 20161 Milano -  Italy

P. Ribelli, G. Solari, A. Bistarelli

Istituto Tecnico Industrile Statale “CARTESIO” , Elettronica e Telecomunicazioni, Informatica-Abacus, Liceo Scientifico Tecnologico - 120, via C. Lombroso, Roma, Italy


This paper contains a discussion about an experience to realise a spontaneous form of  systemic paradigms and organisation induced in a group of students of the Italian secondary school. The pupils have had to realise a neural net work with a back propagation learning and then make a hardware perceptron. We have used the two paradigms of Systems Science, the paradigm of input/output system and the paradigm of self-organising system, to produce in a spontaneous way two different thinking organisations. The first one come from a hierarchic mind, the second one come from an autonomous mind (auto-organisation).



Adalberto Codetta Raiteri

Organizzazione per la Preparazione Professionale degli Insegnanti (OPPI) Via Orseolo, 4  20100-Milano.

Renza Cambini,

Organizzazione per la Preparazione Professionale degli Insegnanti (OPPI)

Associazione Italiana per la Ricerca sui Sistemi (AIRS)


The authors will refer in the first part to the research conducted for five years on the understanding of the meanings of number zero in conflict with those of common sense. In the second part , starting from the experience conducted, they will expound certain reflections on the methodology of educational research which can be incisive only if it adopts methods of system approach.

The research began in 1995 based on a theme proposed in CIEAM 47 (Mathematics and Common Sense), studying the learning of the meanings of the number zero. The same method of enquiry (centered on questions such as: What does zero mean for you? If zero didn’t exist, what difference would it make?) was used for every pupil from age 6-18 and also for some adults. The 2463 answers collected were so surprising that a long period of work, classification and interpretation was necessary involving the use of skills outside the mathematical field, in order to produce a report for imminent publication. Useful indications have emerged for reflection on the curriculum and classroom practice centered on mathematical literacy for all that can perhaps also be extended to the learning of other mathematical concepts

Thirty five teachers from infant to secondary school took part in this research. One of the aims of the research was to actively involve all the researchers, setting up forms of collaborative work which allowed a widespread repercussion of the results of the research. For this reason the research was not governed by the criteria of efficient statistic  based on the representative ness of a significant sample. The instrument of the research (a questionnaire) was planned in a such a way as to guarantee answers which were not prescribed and not pre-classifiable so as to bring out the beliefs of the interviewees  and not the answers expected by the interviewers. The procedures for the research were chosen in a such way as to ensure  an effective involvement of the teacher -researcher, in this way we did not adopt the classical methods of research because we considered the interaction between the observer  ad the observed as a resource and not as a disturbance.

Thank to this procedure, the questionnaire was examined by a variety of working methods  which came out after the first examination and which made it possible to interpret the results from a point of view which made it possible to interpret the results from a point of view which had not been envisaged at the beginning  of the research. More generally, this research has suggested formulating a model of educational research which of necessity involves a numbers of actors (universities, teacher's training centers, schools) each with its own aims (research, training, teaching-learning) and with its own instruments (theories and models, professional instruments, subjects).

KEYWORDS: Educational research, teaching learning processes, cognitive map, number zero



Maria Pietronilla Penna,

Dipartimento di Filosofia, Sezione di Psicologia, Università di Pavia

Vera Stara

Facoltà di Scienze della Formazione, Università di Cagliari

Natale Bonfiglio

Dipartimento di Filosofia, Sezione di Psicologia, Università di Pavia


The use of computer as a new technological tool is becoming more and more  frequent and important in Italian schools, in order to facilitate the learning. However, the integration of PC should be further supported by a positive perception of it from users. The question of how to describe and study the perception of tool by his/her users is, however, a typically systemic one, as it involves the study of a very complex human-machine system.

In this work, we analyze the perception that 306 children and 26 teachers of elementary school have about the PC, using a questionnaire specially designed. We assumed that learning process was deeply influenced by user’s perception, and the obtained results confirmed our hypothesis. Besides, we claim that a good PC perception is not enough to give rise to learning benefits without a good training of the users that will utilize the computer for learning purposes.  

Therefore, we argued that the “agreeable acceptation” attitude by teachers and students about the new technology – the PC in this case – must be related to a good training of teachers that will use the tool within their teaching activity.



Alberto Trotta

Istituto Tecnico Commerciale Statale “E.Loi”

Via E. Loi

00048 Nettuno (Roma), Italy


This essay deals with the teaching methodology concerning the students of secondary school and of the first two university years. It regards the role of deduction, induction, analogy and of the new technologies in the cognitive process, especially within the logic of scientific discovery.

To this purpose the function of a systemic planning is analysed.

It underlines the fact that the objective to which teaching contents must aim is the acquisition of a method for making science rather than for learning science.



Paolo Bouquet

University of Trento - Dept. of Computer Science

Via Inama, 5 - I-38100 Trento  


 Knowledge Management (KM) aims at developing methods and technologies for fostering organizational learning. Recent theories from the area of social learning (see for example, Wenger's work on the communities of practice) show that learning and knowing within an organization is a distributed process, which happens in communities and even at individual level.

However, most technological architectures of KM system tend to ignore this fact, and propose centralized approaches, in which the multiplicity of perspectives from which a piece of information can be looked at is completely neglected. This result in products that are perceived either as oppressive, or irrelevant by their users. In the talk I will present a general architecture in which the distributed component of KM is taken into account. I'll start from the well-known work by Boland and Tenkasi on "Perspective making and perspective taking", and show that KM system must be built on two main principles: AUTONOMY (knowing is a local, contextual process, and as such be dealt with), and COORDINATION (knowledge sharing is indeed a process of "negotiating meaning" across autonomous, heterogeneous perspectives).

  I'll briefly illustrate the practical application of these ideas in a demonstrator that we developed in a joint project between the University of Trento, the Istituto per la Ricerca Scientifica e Tecnologica (ITC-IRST) of Trento, and Arthur Andersen.



Sesini Enrico


La pervasità delle tecnologie web sulla organizzazione aziendale stà profondamente modificando il modo di essere dell'azienda e risulta funzionale all'attuazione di modelli di organizzazione evoluti che senza il supporto tecnolgico mostravano molte difficoltà ad attecchire. Si assiste anche ad un effetto di trascinamento della tecnologia, di deriva quasi incontrollata delle sue potenzialità che porta a subire l'innovazione in unalogica di emulazione dell'altro (del competitors) piuttosto che in una logica di ragionata innovazione.

La teoria dei sistemi può fornire uno scenario per provare a progettare (e non a subire) l'evoluzione organizzativa aziendale.



M.P. Penna,

Università degli studi di Pavia

Università degli studi di Cagliari

Libera Università Maria SS. Assunta

E. Celani,

Libera Università Maria SS. Assunta

N.S. Bonfiglio*

Università degli studi di Pavia


In this paper we first investigate the dynamic changes in the business environment using a systemic approach; our aim is to point out the need for a new perspective when it comes to analyze the landscape faced by different firms and individuals. Going through the biological metaphor, we’ll show the key roles played by the self-organization and learning processes in the effort towards evolution; we focus on the value of communication and knowledge in order to suggest a way for business organizations to adapt to the environmental complexity. We support our thesis with a multiple-choice test, administered to 40 employees in a large firm. 



M.P. Penna,

Università di Cagliari, Facoltà di Scienze della Formazione. e-mail:

Università di Pavia, Facoltà di Lettere e Filosofia

Lumsa, Corso di Laurea in Scienze della Comunicazione

D. Zandona’,

Lumsa, Corso di Laurea in Scienze della Comunicazione


Dipartimento di Visione Artificiale, Facoltà di Informatica Università di Ancona. e-mail:

N.S. Bonfiglio

Università di Pavia, Facoltà di Lettere e Filosofia.

Cattedra di Intelligenza Artificiale. I° Facoltà di Psicologia Università degli studi di Roma

<<La Sapienza>>, via dei Marsi 78,00185 Rome, Italy.


A new systemic approach to the human action recognition is presented. The work is motivated by the observation that the mechanistic approach of most Computer Vision systems is unable to reproduce the human skill in extracting a semantics from the observed movements. We claim that movement must be studied within a systemic approach taking into consideration both the movement image and the cognitive system of the human observer. Therefore we analyzed how the interaction between the two systems changes when a feature of the image-system is manipulated. The image-system features studied are its dynamics, its shape and the spatial relationships occurring between its components.

In order to investigate the effects of these features on human subjects we showed to 200 students movies and slides depicting 12 different actions, quantified through four levels using a mosaic technique that masks shape and spatial relationships.

The goals were to know if humans are able to recognize actions only getting information from dynamics, and at what shape definition level an human observer is able to distinguish two movements that show the same dynamics but are associated to different actions. An hypothesis asserting the existence, within subjects, of movement schemes operating in the recognition process is suggested.


Multidisciplinary Techniques for Classifying Styles in Descriptive Texts: a Preliminary Study

M. Nardon, E. Not, F. Pianesi, M. Zancanaro

IRST, Istituto per la Ricerca Scientifica e Tecnologica,

I-38050 Povo TN, ITALY.

{nardon, not, pianesi, zancana}


It is generally noticed that, even at a subconscious level, writers tend to adopt their writing style to appeal to the specific audience for which they are writing. The style of a text emerges as a combination of various elements that range from lexical choices to discourse organization. In this respect, we consider writing style a systemic variable.

In this study we try to identify classes of writing styles for a more objective classification of text, without reducing this stylistic classification to a mono-disciplinary analysis. We choose a multi-disciplinary approach in which different techniques cooperate. To this purpose we have isolated some stylistic indicators which characterize the style of a text, according to their relative frequencies in the corpus. Moreover we used the linguistic text analysis proposed by van Dijk to identify, for texts on a certain subject, whether there are different distributions of topics for texts belonging to different writing styles.



Daniele Gugelmo Dias

Lecture to the Universidade Federal do Rio Grande do Norte (UFRN) - Brasil

Research in the Núcleo Interdisciplinar de Comunicação Sonora (NICS) – Unicamp - Brasil
student in Ph D at Universidade Estadual de Campinas (Unicamp) - Brasil


This work projects in perspective a discussion of the compositional process created in the XX century using the General System Theory (GST). From the perspective of the research presented here, the break of tonal process in the beginning of the century, as referential framework, the dodecafonism, the integral serialism, the indeterminism and the minimalism between others, are examples of “Compositional Systems”. The aim of this work describes the compositional methods themselves, but it shows a referential theory that is hand to analyze and synthesize the complexity of them. The GST can be useful to classify and understand this complex universe. The major subject of this paper is focusing instrumental music; discussing the Compositional System in the XX Century after of the dissolution of the tonality besides the systemic fragmentation of the XX century music.



Graziano Terenzi

Dipartimento di Filosofia – Sezione Psicologia

Università degli Studi di Pavia

Piazza Botta, 6 (Palazzo S. Felice)

27100 Pavia, Italy



The problem of emergence of the symbolic function is dealt with in the context of a neural solution to the symbol grounding problem. We argue that emergence can be better understood as the transition, taking place within logically open systems from unordered collective behaviors to ordered collective behaviors, i.e. phase transitions. By appealing to the latter analysis, we illustrate a neural network architecture based on two interconnected modules: one deputed to symbol categorization, and another deputed to associate thematic roles to sentence constituents. The operation of the model is based on the idea according to which the emergence of entities endowed with semantic content (i.e. symbols) depends both on the formation of a suitable mapping from  linguistic stimuli to perceptive meanings (and the other way down), and on the further semantic processing to which such entities undergo within a speaker’s cognitive system. Clearly, the centrality of the notion of emergence suggests the relevance of a systemic approach to natural language.



G. Tascini, A. Montesanto, P.Puliti, N.Rabascini

Institute of Computer Science

University of Ancona



The sistemics and the reactive robotics point of views underline the meaning of emergent behaviour trought the interaction between sensorial systems and the action in the environmental context. The global behavior is the sum of a lot of micro-behaviors.

This work addresses the problem of developing an autonomous system to navigate using only visual information. A modular navigation for a visually guided mobile robot is proposed: a control system for reactive navigation, obstacle avoidance and target achieve. Brooks' approach focalize the concept of knowledge without representation, so we try to study the self-organization process of the information using the clustering capacity of Self-Organizing Maps to make a gross self-location and to reactive navigate. Location of object position is achieved using the depth perception resulting by the differences in the image of a trinocular visual sensor. A system for goal recognition is used too. All those behaviors let emerge an autonomous navigation behavior without collisions.



M. C. Tofoni,

Corso Interfacoltà di Scienze della Comunicazione, Università degli studi di Macerata,

A. Montesanto

Istituto di informatica, Università degli studi di Ancona,



The aim of this paper is to provide methodological indications in projecting and implementing on-line hypertexts containing clear and easy retrievable information and also able to foster the idea of co-operative writing.

From a systemic point of view, a web-site can be considered as a global system made up of sub-systems, i.e. all the pages in the site. Coherence to the context emerges from the interaction of the single sub-systems. Although remaining globally coherent, the latter have also their own independence and internal coherence. That is to say, once opened, the single page becomes, in turn, a global system containing different sub-systems (i.e. the links).

The web-site, as a complex system, owns features as abundance and heterogeneity of information, interconnection of information, variety of experiences and professionals working on its organization and/or facing its reading/navigation. In addiction to this, it provides different levels of reading of the information.

We will hereby point out our attention on two levels: a) textual information in the pages; b) information given by the files and directories organization structure. Both of them need settled methodological principles so that the hypertext is highly readable. This work illustrates two different approaches to the structuring of the hypertextual information. The two combined, taking into account what said above, have the purpose to foster the realization of hypertexts with high degrees of usability.



Giancarlo Mauri

Department of Information Sciences of the University of Milano


Computing with membranes is a recent branch of Molecular Computing, proposed at the end of 1998 by Gh. Paun, and inspired from the structure and the functioning of the alive cells. In short, in the compartments defined by a membrane structure one has multi-sets of symbol-objects which evolve according to given rules also associated with each compartment; the objects can pass through membrane under precise conditions, the membranes can be dissolved or divided, their thickness/permeability can be modified. The rules are applied in a non-deterministic, maximally parallel manner: in each time unit, all objects which can evolve should do it. In this way, transitions from a configuration of the system to another configuration are defined; a sequence of transitions defines a computation. A result is associated with a halting computation in the form of the vector of natural numbers describing the multiplicity of certain objects in a certain membrane. The objects can also be described by strings and then the evolution rules are string processing rules and the result of a computation is a language.

Several variants were introduced, having in mind the three (contradicting) goals: to have models as realistic as possible, but as powerful as possible, and as efficient as possible.

Many of these variants are computationally complete, they can generate all recursively enumerable sets of vectors of natural numbers (all recursively enumerable languages, in the case of string-objects).

Several of the P systems (with enhanced parallelism, for instance, by membrane division) are able to solve NP-complete problems in polynomial---often linear---time (of course, by trading space for time and making use of an exponential space).

The lecture will have three parts:

Part I - BASICS:

     Elements of formal language theory used in membrane computing (Chomsky      hierarchy, matrix grammars, normal forms), the structure and the      behavior of bio-membranes, the basic idea of a P system, definitions,      examples.

Part II - POWER:

     Variants of P systems with symbol-objects and with string-objects,      characterizations of recursively enumerable sets of vectors of natural      numbers and of recursively enumerable languages. Hierarchies on the      number of membranes.


     Solving NP-problems in polynomial/linear time by P systems with      membrane division and by P systems with duplication of string objects.



Gianfranco Minati

Associazione Italiana per le Ricerca sui Sistemi (AIRS), 42 Viale Pellegrino Rossi, 20161 Milano, Italy,

Alberto Ricciuti

Attivecomeprima – Onlus (Breast cancer Association) – Milano -

Associazione Italiana per la Ricerca sui Sistemi (AIRS) – Milano -


Some basic distinctions between the processes of balancing and compensating are introduced.

The two concepts are elaborated with reference to subjects provided with cognitive models, that’s able to make internal information processing as assumed by Cognitive Science. Typical area of interest for this issue are people having some disabilities.

The topic of  considering balancing and compensating processes in systems provided with Cognitive Systems bring necessarily the attention to the classical discussion between empiricist and rationalist philosophers.

The dichotomy of empiricism versus rationalism influenced theories of the mind in the form of 'nature/nurture debate'.  In literature a new point of view has been introduced focusing on information processing. Strategies of using resources available in the context are developed by agents, distinguishing between reorganizing and restructuring.

Reference is made to the DYnamic uSAge of Models (DYSAM) approach, that is simultaneous usage of the available ones in order to decide, to take actions the more suitable as possible to the context. For instance people usually walk in the context: interest of  this contribution is for people using their context to walk, inventing ways of using it.

On the basis of what introduced, the project to add processing abilities, ability to have memory and to be inquired, to prostheses and tools used by handicapped people to walk, are proposed. It has the purpose to allow physicians to check the daily behavior.

Physical Rehabilitation must also be completed with Cognitive Rehabilitation inducing in the disables new appropriate models to be checked by psychological analysis trough questioners.

KEYWORDS: balance, context, compensation, cognitive model, disability, dynamic usage of models, equilibrium, intelligent interactive tool, multiple sclerosis, use




Medico chirurgo

Membro A.I.R.S.

Via Quarto dei Mille, 10

25128 Brescia (Italia)

Tel. 030/3384621 – Fax. 030/3391721



If emergence is one of the most peculiar feature we find in the complex systems, individuality follows as natural consequence especially in human beings.

The goal of a correct therapy should be meeting these fundamental characteristics and use them in order to bring the system back to its own equilibrium. By this way we can successfully try to avoid the increase of entropy taking place when we work on it.

The previous step, in order to prescribe the correct remedy to the dis-functioning whole, is to know how it works in standard circumstances and in far-from-equilibrium states. So we are in need to "push" the system and excite emergence.

Strange, but true, we own a method to reach this goal perfectly suited to human beings and using it we can draw exhaustive pictures of functioning-dis-functioning wholes.

By a second step, prescribing the pattern of the dis-equilibrium to the system we can reach the correction automatically in accordance to the first Ashby' s principle.

 This systemic therapy gets to the roots and to the effectiveness of any self-regulating device in the most harmless conduct.



Alberto Ricciuti

Attivecomeprima – Onlus (Breast cancer Association) – Milano -

Associazione Italiana per la Ricerca sui Sistemi (AIRS) – Milano -


Chemotherapy is burdened with heavy side effects that often cause both rejection by the patient and reduction, by the doctors, of the therapeutic dosage that is considered to be optimal, or even interruption of the therapy. The illness (“fatigue”) experienced by the patients, the causes of which are still unclear and which is often underestimated even by the oncologists, induces them to look for alternative support therapies without communicating it to their doctor in charge, often with unsatisfying outcomes. Moreover, such therapies are usually opposed to by oncologists because they often use languages and methods that are not shared by the scientific community. On the other hand though, the fact that chemotherapy causes toxic effects at a level which is deeper and more general than that usually considered, is overlooked. In fact, there are reasons to believe that at least most of its effects on the organism are due to systemic toxic actions developing at cell and tissue levels, which produce a heavy disturbance of self-organization, self-repair, and self-defense abilities that define the entire organism as autopoietic system. A systemically oriented  therapy is therefore proposed, that, referring to concepts and instruments shared by the scientific community and introducing a new knowledge model centered on the person and not on the disease, yields a reduction of the physical pain experienced by patients affected with cancer during and after chemotherapy, an enrichment with new communicative and operational contents of the doctor-patient relationship, an effective and concrete answer to the need to be listened to and to the need of hope the cancer patient looks for, the appearance of a clinical anthropological look in the medical praxis.

The anti-tumoral chemotherapy, that is the use of medicines aimed at the destruction of neoplastic cells, is regarded as a cornerstone in tumor therapy. This therapy is nevertheless burdened with such frequent and intense side effects, that serious pain is induced in the patients and that doctors are often forced to space further out in time the treatments, to reduce the dosage, or even to interrupt the therapy. Besides, this is lived by the patients as a reduction of their life expectancy.



P. L.  Bandinelli,

Dipartimento di Salute Mentale ASL Roma “E” c/o Servizio Psichiatrico di Diagnosi e Cura, Azienda Complesso      Ospedaliero S. Filippo Neri. Via G. Martinotti 20, 00135 Roma, Italy

Cattedra di Intelligenza Artificiale. I° Facoltà di Psicologia Università degli Studi di Roma  “La Sapienza”, Via dei Marsi  78, 00185  Rome, Italy. e-mail:

Associazione Italiana per la Ricerca sui Sistemi (AIRS) Italian division of the International Society for the Systems Sciences (ISSS), USA – 42, via Pellegrino Rossi – 20161 Milano -  Italy

C. Palma

Cattedra di Intelligenza Artificiale. I° Facoltà di Psicologia Università degli Studi di Roma  “La Sapienza”, Via dei Marsi  78, 00185  Rome, Italy. e-mail:

Associazione Italiana per la Ricerca sui Sistemi (AIRS) Italian division of the International Society for the Systems Sciences (ISSS), USA – 42, via Pellegrino Rossi – 20161 Milano - Italy

M.P. Penna

Cattedra di Intelligenza Artificiale. I° Facoltà di Psicologia Università degli Studi di Roma  “La Sapienza”, Via dei Marsi  78, 00185  Rome, Italy. e-mail:

Dipartimento di Filosofia – Sezione Psicologia. Università degli Studi di Pavia. Piazza Botta 6 (Palazzo S. Felice) 27100 Pavia, Italy

Associazione Italiana per la Ricerca sui Sistemi (AIRS) Italian division of the International Society for the Systems Sciences (ISSS), USA – 42, via Pellegrino Rossi – 20161 Milano - Italy

E. Pessa

Cattedra di Intelligenza Artificiale. I° Facoltà di Psicologia Università degli Studi di Roma  “La Sapienza”, Via dei Marsi  78, 00185  Rome, Italy. e-mail:

Dipartimento di Filosofia – Sezione Psicologia. Università degli Studi di Pavia. Piazza Botta 6 (Palazzo S. Felice) 27100 Pavia, Italy

Associazione Italiana per la Ricerca sui Sistemi (AIRS) Italian division of the International Society for the Systems Sciences (ISSS), USA – 42, via Pellegrino Rossi – 20161 Milano -  Italy

R. Petroni

Cattedra di Intelligenza Artificiale. I° Facoltà di Psicologia Università degli Studi di Roma  “La Sapienza”, Via dei Marsi  78, 00185  Rome, Italy. e-mail:


A systemic approach was introduced to implement a model for investigating the  propagation phenomena of electric activity in the cortical substance of the brain. The proposed model was based on a  cellular neural network. We  focussed our attention both on modality of  generation and on mechanisms of propagation of epileptic activity. Differently from other models, the one we proposed takes into account also the diffusion phenomena in the Extra Cellular Fluid (LEC), the glial transport and the dynamic interaction between three subsystems: Neurons, Glia and LEC.



Sabre Brahms, M.A.; 368 Museum Drive, Los Angeles, CA, 90065, USA ; Consulting Faculty, Phillips Graduate Institute, Los Angeles, CA, USA; Doctoral Student, Saybrook Graduate School, San Francisco, CA, USA


This analysis focuses on the impact of religious ideology as it affects the tensions between Palestinians and Israelis that are now occurring in the occupied territories.  The discussion revolves around the West Bank city of Hebron as a point of reference, and specifically pays attention to religious tensions as they manifest at the Tomb of Abraham, a holy site in Hebron that is sacred to both Jews and Muslims.

The paper provides a brief historical overview of the shared use of the Tomb of Abraham and then proceeds to define several systemic factors that have emerged in the past few years within both the Islamic and Jewish religious communities.  An analysis of the co-evolution of religious zealotry is provided, as is a discussion of the increase in religiously sanctioned violence.

An analysis of the factors that contribute to the current far-from-equilibrium status between Jews and Muslims is presented and the negative aspects of their mutual interdependence are discussed.  According to Jackson's definition (1988, 1995), aspects of the Israeli-Palestinian conflict that are coercive are defined.  Conclusions are drawn based on the analysis and suggestions are made to effectively attempt any future peace-building activities between Israelis and Palestinians.

KEYWORDS:  Systems Analysis, oercive systems, conflict resolution, Israeli-Palestinian conflict.



Larry A. Magliocca, Ph.D.

Associate Professor & Executive Director

The Ohio State University


Gianfranco Minati, Ph.D.

Member of the inter university Center E.Co.N.A., at the University La  Sapienza, Rome (Italy)

Distinguished consulting Faculty at the Saybrook Institute, San Francisco (USA)

President of the Italian Association for Research on Systems (AIRS)


Globalization seems inevitable to many in the international community with various promises and risks for the nation state participants.  One significant risk of globalization is inherent in the “default” approach of transnational business of their “ethics-in-practice”: competition of businesses in the international arena, regulated primarily by a free market philosophy, ignores the well-known limits to growth of the finite resources of Earth.  Although recent international conferences have “professed ethics” of sustainable development, these professed ethics seem to have little impact on the daily business of globalization.  The authors believe that a clear distinction must be drawn by ethics-in-practice and professed ethics to understand the confusing state of globalization.  Second, they propose that the transactional leadership process currently assumed by international partners of globalization attempts to implement technical solutions to fundamentally human problems.  This leadership approach creates a sense of alienation and cynicism among many international groups about the purposes and methods of globalization.  A more appropriate leadership approach, transforming leadership (proposed by James MacGregor Burns seminal work, Leadership) requires a foundation of shared needs, values, and aspirations to drive the real change of the future.  Transforming leadership, however, has been primarily a national phenomenon.  A real test of the transforming leadership concept is proposed here through the development of the emergent ethics of sustainable development through an international coalition of committed participants.  Substantial numbers of individuals and their associations believe in the values of sustainable development for the benefits of future stakeholders of the planet.  Their coalitions could reach a critical mass of impact if the basic tenets of transforming leadership were adopted as their guiding approach.



Arne Collen

Saybrook Graduate School, 450 Pacific, San Francisco, CA 94133 USA


Disciplinarity is defined and five forms are presented. They are compared in regard to their systemicity and complexity. Their integration is discussed in terms of relational dynamics, and as a conceptual system, that represents not only a way of thinking, but also a bridge to several advantageous applications.



Eliano Pessa

Dipartimento di Filosofia – Sezione Psicologia

Università di Pavia

Piazza Botta, 6

27100 Pavia, Italy ;

Germano Resconi

Dipartimento di Matematica e Fisica

Università Cattolica del Sacro Cuore

Via Musei 41

25121 Brescia, Italy


Compensation mechanisms are mental tools widely used in building models of complex systems. They enable the restoration of conservation laws and of invariances in systems where the inner interactions between the elements would seem to give rise, from the starting, to an unpredictable variability and to random behaviors. Between these mechanisms, whose logical structure is the object of General Systems Logical Theory (GSLT), we will take into consideration the special case of the ones based on gauge transformations, which deserve a paramount importance for the physical systems described by field equations. Namely, the requirement of invariance with respect to these transformations allows a precise definition of the concept itself of “interaction” between the elements of a system whatsoever, provided the latter be defined in terms of field quantities. Such an approach was so successful, within modern theoretical physics, as to offer a sound basis to virtually all theories of fundamental interactions and of macroscopic collective phenomena. Notwithstanding, its main shortcoming is that it can be applied only to fixed-structure systems, that is to systems in which the number of possible interactions is fixed in advance and the coupling parameters are to be considered as constants, determined once and for all, like the electron charge or the gravitational constant. On the other hand, in many physical and nonphysical complex systems we need a generalized description able to take into account the occurrence of a variable number of interactions and of space-time dependent coupling parameters. In this paper we will introduce a formalism, based on a generalization of gauge transformations, able to deal with such situations, typical of variable-structure systems. We will explore some dynamical consequences of such a generalization and its implications for a general systems theory, mainly in the case of description of cognitive, social and economic behaviors.



Maria Pietronilla Penna

Cattedra di Intelligenza Artificiale. I° Facoltà di Psicologia Università degli Studi di Roma  “La Sapienza”, Via dei Marsi  78, 00185  Rome, Italy. e-mail:

Dipartimento di Filosofia – Sezione Psicologia. Università degli Studi di Pavia. Piazza Botta 6 (Palazzo S. Felice) 27100 Pavia, Italy

Associazione Italiana per la Ricerca sui Sistemi (AIRS) Italian division of the International Society for the Systems Sciences (ISSS), USA – 42, via Pellegrino Rossi – 20161 Milano - Italy


The description of cognitive development can be done only within a systemic framework, because it appears as characterized by a sequence of structural “Phase transitions”, that is rearrangements of cognitive schemata and cognitive architectures associated to the emergence of new strategies and new behaviors. These transitions, in turn, are nothing by a by-product of the mutual interaction between the organism and its environment. A main question in cognitive science is related to the possible existence of a metacognitive system, acting as a controller in strategy selection and in cognitive change. Such a system should produce evaluations of strategies and behaviors, related to their feasibility, usefulness, pleasantness, and so on. In recent times, however, the existence of such a system was called into doubt, as suitable models of emergence of strategies in children performing arithmetic operations showed how strategy change could occur only as a consequence of the feedback received from the environment. In this paper we will deal with such a question, both from a model-theoretic and an experimental side. More precisely, by resorting to experiments on elementary school children learning multiplication, we will show how the temporal evolution of their behavior in solving multiplication problems can be simply explained through suitable reinforcement learning laws, without any introduction of a metacognitive system. Such a result is of importance as a constraint for the future models of the emergence of strategies in cognitive development.