Cybernetic Model of Intelligence

  

Figure 1: Circular processes of intelligence. 

1. Introduction

The hype around Artificial Intelligence raises the question “What is Artificial Intelligence?” and, consequently, “What is Intelligence?”. We need a coherent definition of the concept of intelligence, because otherwise the public discourse about AI will fail and degrade to an exchange of propaganda and to manipulation. But do we have a coherent definition of intelligence? 
Consider the following example, which is representative for many others: 

Human intelligence is the capability to learn, form concepts, understand, and apply logic and reason ... recognize patterns, plan, innovate, solve problems, make decisions, retain information, and use language to communicate. (adapted from Wikipedia).

Such definitions try to define the concept by means of a list of examples of intelligence, a list of possible uses of intelligence. But the question is not which types of intelligence exist, nor what intelligence can be used for. We need a definition which is valid for all the types and for all the uses: that is, we need a general definition. Such a definition can be obtained if we ask: what does intelligence consist of? And particularly: “what functions does intelligence consist of?”

2. Piaget’s theory of intelligence

An answer to these questions can be found in Jean Piaget’s studies of the development of intelligence. Here are some quotations from and adaptations of Piaget’s theory, that mention the main functions and characteristics of intelligence:

1. EIVIRONMENT. Intelligence develops (and knowledge is generated) through interaction with the environment.
2. ACTION. …all knowledge is tied to action, and knowing an object or an event is to use it by assimilating it to an action scheme… (Piaget, 1967:14-15)
3. SCHEME. A pattern consisting of (von Glasersfeld, 1995:66):
• Recognition of a certain situation;
• a specific activity associated with that situation; and
• the expectation that the activity produces a certain previously experienced result. 
4.  ASSIMILATION. “Intelligence is, in fact, assimilation to the extent that it incorporates into its schemes all the data of experience” (1). Piaget, 1936:12. 
5. ACCOMMODATION. « … mental life is also accommodation to the surrounding environment… intelligence constantly modifies [previous schemes] to adjust them to new data.” (2). Piaget, 1936:13.
6. EQUILIBRIUM. The interaction between the organism and the environment requires a balance between assimilation and accommodation. Adapted from Piaget, 1937:309.
7. KNOWING. "all knowledge is both accommodation to the object and assimilation to the subject" (3). Piaget, 1937:312.
8. CONSTRUCTION OF KNOWLEDGE. "The very notion of object is far from being innate and requires a construction that is both assimilative and accommodative." (4). Piaget, 1936:13.
9. COORDINATION. It is precisely the role of intelligence to coordinate assimilation and accommodation with each other. Adapted from Piaget, 1937:309.
10. COORDINATIONS OF SCHEMES. Assimilation establishes an ever closer network of coordinations between schemes and between the things to which these schemes apply. Adapted from Piaget, 1937:308.

Further, in his study of the development of intelligence, Piaget gives the following definition of intelligence (Piaget 1936: 12):

« Intelligence is an equilibrium between assimilation and accommodation” (original: «L’intelligence … est un équilibre entre l'assimilation et l'accommodation. »)

One year later, in his study of how intelligence constructs reality, Piaget suggested to view intelligence as follows (Piaget 1937: 311):

"Intelligence ... organises the world by organising itself" 
(original: "L’intelligence . . . organise le monde en s’organisant elle-même."

The idea of an equilibrium in the first definition and the repetition of “organize” in the second suggest a circularity such as we find in cybernetic (feedback) systems. So, based on Piaget’s developmental theory of intelligence and knowledge - as summarised in the previous statements - we developed the following cybernetic model of intelligence. 

3. Cybernetic Model of Intelligence

Intelligence is constituted by 3 main processes (functions) and 2 feedback loops that enable the organism to develop its own experiential world. 

Figure 2: Cybernetic model of intelligence.

Intelligence is not found ready-made; the functions which constitute it progressively develop thanks to the interaction of the organism with the environment. These interactions typically follow a tripartite pattern called a “scheme”. Experience happens through construction, organisation and application of schemes (knowing) thus contributing to the creation of the organism’s experiential world. 

The function of assimilation controls (regulates) this process of knowing by incorporating experience into a scheme (assimilation loop) whereas the function of accommodation controls it by adjusting schemes when they do not fit to experience (accommodation loop). In this way the three processes constituting intelligence (knowing, assimilation, accommodation) are coordinated and equilibrium between the organism and its environment is maintained.

4. Conclusion

What is intelligence? The cybernetic model presented here suggests that the essence of intelligence is a feedback system consisting of 3 main functions - knowing, assimilation and accommodation - connected by 2 feedback loops (assimilation loop, accommodation loop). This system demonstrates that intelligence is not ready-made: it develops by ensuring - thanks to its functional architecture - a balanced interaction of the organism with the environment.

References

Piaget, J. (1936) La naissance de l'intelligence chez l'enfant, Neuchâtel: Delachaux et Niestlé. 9th edition, 1977. Translated as: The Origins of Intelligence in Children (New York: International University Press, 1952) as well as The Origin of Intelligence in the Child (London: Routledge and Kegan Paul, 1953).

Piaget, J. (1937) La Construction du Réel Chez l’Enfant. Neuchâtel: Delachaux et Niestlé. 5th edition, 1973. Translated as: The construction of reality in the child (New York: Basic Books, 1954). Also translated as The Child's Construction of Reality (London: Routledge and Kegan Paul, 1955).

Piaget, J. (1967) Biologie et connaissance (Biology and knowledge), Paris, Gallimard.

von Glasersfeld, E. (1995). Radical Constructivism: A Way of Knowing and Learning. London: Falmer Press.

Notes

(1) L'intelligence est, en effet, assimilation dans la mesure où elle incorpore à ses cadres tout le donné de l'expérience. Piaget, 1936:12.

(2) … la vie mentale [est] aussi accommodation au milieu ambient … l'intelligence modifie sans cesse [les schèmes antérieurs] pour les ajuster aux nouvelles données. Piaget, 1936:13.

(3) toute connaissance est à la fois accommodation à l'objet et assimilation au sujet. Piaget, 1937:312.

(4) la notion même d'objet est loin d'être innée et nécessite une construction à la fois assimilatrice et accommodatrice. Piaget, 1936: 13.

 

Circular thought patterns: An autopoietic process model

This post has been inspired by reading what Zack wrote on November 17, 2012 as a comment to Myrko Thum's blog post here: http://www.myrkothum.com/finding-balance/ I think that my model of how we generate knowledge and experience can shed some light on the conception of "circular thought patterns" mentioned by Zack.

Here is my model, in a version recently published (Bettoni & Eggs, 2010, section "The Logic of Experience", note: in the article the Figure 5 reproduced here has an error, corrected here below).

-------------------------------------------------------------------------------

Circular organization of knowing in a feedback loop.

 First of all, in line with Piaget (1967) we suggest seeing a formative, organic principle at work in the generation of knowledge, too; secondly, as proposed by Freeman (2000: 9), we try to conceive of knowledge operands as “a kind of living structure” with constructive procedures or operational sequences organized according to an underlying organic principle. Finally, since the essence of a living system (organism) is autopoiesis, or in other words “self generation,” we suggest understanding knowing as an autopoietic process with its peculiar form of circular organization. Maturana, who developed the concept of autopoiesis, says: “The product of the functioning of the components is the same functioning organisation that produced them.” (Maturana 1980: 9).  In the domain of knowing, this requires that the interactions of the elements (= knowing) “bring forth elements of the same kind; that is crucial” (Maturana & Poerksen 2004: 107).

Accordingly, we conceive of knowledge as a result of cognitive processes in the dynamic form of a functional organization that extends or modifies the functional organization that produced it. In this conception, knowledge displays a “product/function duality” similar to the wave/particle duality in physics; as a product, the results of knowing can be used as building bricks of a knowledge edifice (a theory, an inquiry, a claim, a judgement, etc.); as a function, they become part of the same “knowing system” that produced them.

In Figure 5 the autopoietic process of knowing is represented as circular organization with two blocks (thinking and experience) connected by a feedback loop.

Thinking here has been distinguished into two sub-processes – perception and elaboration – with A (alteration) as input, P (percept) as an intermediate result and K (knowledge) as the final product.

For example, if we consider the knowing needed to knot a necktie in the morning, then A comes from the necktie, Perception and Conceptualization are relevant when learning to knot the tie and Elaboration is relevant when the knot is made without looking at, automatically.

The second block, Experience, where the final result K is fed back from Thinking, has been distinguished into three sub-systems: a system of attention (Ceccato 1964, Ceccato 1964/1966; Bettoni 2007), which on one side controls the constitutive part of Thinking (perception and conceptualization) through a Water Logic System, and on the other side also controls the regulative part of Thinking (elaboration) through a Rock Logic System. These three systems are the place where the knowledge, K, produced in thinking and fed back behaves as function and becomes part of the same “knowing system” that controlled its production.

Surprisingly, perception is far more important for knowledge than elaboration. But traditional thinking – according to Edward de Bono – is focused exclusively on elaboration and dislikes the vagueness, subjectivity and variability of perception. In our tradition, elaboration consists basically of the use of argument and reason with the goal of “falsification”: i.e., demonstrating the contradictions of a position or showing that something is false. Reality is proposed as the Universal Absolute that has to be used as the reference. “I am right – you are wrong” (de Bono 1992) condenses the essence of the “logic of elaboration” (rock logic, because, like a rock, it is permanent, hard, and has a definite shape).

Luckily perception has a different logic, the logic of pattern-building systems, but we ignore it. Why? Because we have never understood perception! Just as water fits in a bowl or bottle, the patterns that perception constructs are not right or wrong; they simply “fit” in the situations and circumstances that the person lives and experiences (water logic). Conceptualization (categorization) also works within the same “water logic”: this is the main reason why perception is more important for knowledge than elaboration. For example, this page can be conceived as a “part” (of the blog) or as a “whole” (in relation to the lines, words, etc. of this webpage), depending on what fits what the person lives, not depending on “Reality.” We, with our conceptual operations, can flexibly adapt our perception and conceptualization to “fit in the bowl.” This “operational” perspective is the pioneering contribution of Silvio Ceccato and his Italian Operational School (Glasersfeld 1995; Sowa 1983; Bettoni 2007).


REFERENCES:
Bettoni M. (2007) The Yerkish language – From operatio al methodology to chimpanzee
communication. In: Glanville R. & Riegler A. (eds.) The importance of being Ernst. edition echoraum, Vienna: 107–121.
Bettoni, M. & Eggs, C. (2010). "User-centred Knowledge Management: A Constructivist  and Socialized View". Constructivist Foundations, Vol. 5, number 3, 130-143. http://www.univie.ac.at/constructivism/journal
Ceccato S. (1964) A model of the mind. Methodos 16: 3–78.
Ceccato S. (1964/1966) Un tecnico fra i filosofi. Vol. 1 & 2. Marsilio, Padova.
de Bono E. (1992) I am right – you are wrong: From this to the new renaissance – From rock logic to water logic. Penguin, London.
Freeman W. J. (2000) How brains make up their minds. Columbia University Press, New York.
Glasersfeld E. von (1995) Radical constructivism. A way of knowing and learning. Falmer Press, London.
Maturana H. R. (1980) Biology of cognition. In:Maturana H. R. & Varela F. J., Autopoiesis and cognition. Reidel, Dordrecht.
Maturana H. R. & Poerksen B. (2004) From being to doing. The origins of the biology of cognition. Carl-Auer Verlag, Heidelberg.
Sowa J. F. (1983) Conceptual structures : Information processing in mind and machine. Addison-Wesley, Reading MA

What means "Cybernetics"?

I searched Wikipedia for a definition of the term "Cybernetics" and found a sentence, that in my view is rather wrong and misleading for various reasons (Cybernetics is the interdisciplinary study of the structure of regulatory systems; http://en.wikipedia.org/wiki/Cybernetics, 11.12.2011).

Then I searched on the website of the American Society for Cybernetics (http://www.asc-cybernetics.org/foundations/definitions.htm), but could not find a convincing solution.

Finally, looking in my archive I found a folder with notes of 1988 in which I had tried to define the term "Cybernetics"; I have completed them now, and here is the summary of what came out (an article will follow later):

• We call Cybernetics the science that considers dynamic systems in terms of OPERATIONS and CONTROL and aims at enabling artificial systems to perform like organisms at 3 levels: physical, biological and mental.

 This definition is based on my experience as engineer in the fileds of System Dynamics and Process Control as well as on my cooperation with Silvio Ceccato and Ernst von Glasersfeld. I have tried to integrate in one sentence the definitions given by three cybernetic pioneers: Norbert Wiener, Heinz von Förster and Silvio Ceccato.