New models for the learning process: beyond constructivism? - André Giordan
When we observe the teaching or mediation of science, three main traditions may be discerned. The first tradition, which is the most widespread and long established, is founded on the idea of frontal transmission of knowledge. Each stage in the process introduces specific subject matter, taken from a syllabus or table of objectives, that in sum makes up the knowledge to be acquired. In this type of teaching or mediation there is a linear relationship between the teacher, the repository of a body of knowledge, who delivers an increasingly often illustrated lecture, and the pupil on the receiving end. In museums, this tradition is reflected in a ‘bookish’ presentation or in the presentation of a ‘medium’. In every case a ‘person who knows’ pours out a predetermined package of knowledge to a passive listener. At school, this transmission of information is reinforced by a corresponding effort of memorization.
The second tradition, developed since the 1950s, is based on a training process upgraded to the rank of learning principle. The chosen propositions are of the stimulus-response type, and faith is pinned on ideas of ‘conditioning’ and ‘reinforcement’. The teacher, or most commonly the programme designer, analyses the behaviour, the chaining of which expresses the skills to be acquired. He or she then devises questions capable of bringing those skills into the open and couples the replies of the pupil with approving or disapproving reinforcement stimuli. In practice, this tradition takes the form of teaching through exercises. In museums, it is reflected in the widespread ‘push-button’ trend. Programmed teaching of this sort has been given a new lease of life with the development of computers.
Lastly, the third tradition, of more recent origin,’ corresponds to what is generally called ‘the discovery method’ or ‘learning by construction’. It responds to the spontaneous needs and interests of the pupils; it advocates their freedom of expression, creativity and life skills; it highlights independent discovery and the importance of proceeding by trial and error in a process of construction initiated by the pupil. In theoretical terms, the construction of knowledge is achieved by giving a major role to pupil activities, a somewhat debatable approach in practice.
It is interesting to find that each of these positions in general refers back to an emblematic psychological theory. Do there exist constants in human thought? The pedagogy of transmission has been based on empiricism throughout a long tradition dating back to Locke (1693). The second approach was founded by behaviourism (Holland & Skinner, 1961; Skinner, 1968). The third one developed as part of constructivist psychology. Actually, we should speak of constructivist theories in the plural, since constructivism had a number of variants: some writers stressed associations (Gagné, 1965, 1976; Bruner, 1966), others ‘cognitive bridges’ (Ausubel et al., 1968), and yet others ‘assimilation and accommodation’ (Piaget & Inhelder, 1969; Piaget, 1971), co-action (Doise et al., 1975, 1985; Perret-Clermont 1979, 1980) or interaction (Giordan, 1978).2
For Ausubel (1968), for example, everything is a question of linkage, which is facilitated by the existence of ‘cognitive bridges’ that render the information meaningful in relation to the existing overall structure. For him, new knowledge cannot be learned unless three conditions are met: first, the availability of more general concepts that are gradually differentiated during the learning process; second, a system of ‘consolidation’ to facilitate the understanding of ongoing lessons, since new information cannot be introduced as long as the preceding information has not been mastered. If that condition is not met, the entire learning process may founder. Lastly, the third condition - ’integrative conciliation’ - consists in identifying similarities and differences between the old and the new knowledge, in distinguishing them and in resolving any contradictions that arise. This necessarily leads to readjustments.
Piaget (1971) also assumes that the pupil deals with the new information in accordance with previously acquired skills and assimilates it. In return, some accommodation is often necessary. As a result, the patterns of thought already in place are transformed to take account of the new circumstances. For Piaget, the aim was to link the new information with what was already known and to graft it onto other concepts by taking account of the ‘patterns’ available to the pupil.
Contrasting with these three traditions, research is proceeding in a new direction, known as the ‘didactics of science’, from which new ideas about the learning process are emerging. This approach calls for teaching environments that facilitate at the same time the understanding process, the learning process and the mobilizing of knowledge. It has given birth to new trends aimed at improving the teaching and mediation of science.
The starting point for this new research direction was a highly pragmatic observation. Why did normal teaching methods, whether of the traditional or so-called ‘new pedagogy’ type, yield such poor results? Educational output, i.e. the amount of knowledge acquired and capable of being mobilized in relation to the time spent, seemed very low and even nil in some cases. Moreover, a number of ‘errors’ in reasoning or ‘wrong’ ideas appeared again and again in pupils with disconcerting repetitive-ness, even after one, two, three or sometimes ‘n’ lessons on the same subject. In an attempt to respond to this situation, a series of scientific investigations was begun to describe the processes used by learners and the conditions that impede or favour those processes. This research has resulted in models that are now challenging the cognitive sciences.
Contributions to the teaching of science
Constructivist models no longer see the learning process as the result of impressions left in the pupil’s mind by sensorial stimulations emanating from the teaching, like the effect of light on a photographic film. It is quite rare for the mental structure of a pupil to be spontaneously in tune with that of a teacher, even if the teacher has done his or her job well and the pupil has been listening carefully; in any event, it never happens immediately. It occurs only in very special cases, when the teacher and the learner ask themselves the same type of questions and possess the same type of references. In fact, such a situation arises mainly between peers or where the information in question is very commonplace.
The organization of a learning process or the structuring of knowledge stems at bottom from the activity of the subject. Learning thus becomes a capacity for effective or symbolic, physical or verbal action. This capacity depends on the existence of mental patterns resulting from the action. The latter are engendered by the active repetition of behaviour that, when concerned with picturing realities or abstractions, reconstructing them and combining them in thought processes, plays a fundamental role.
Unfortunately, the constructivist models appear rather primitive in actual teaching or mediation. Our pedagogical research shows that the learning process covers a number of multiple, polyfunctional, ‘contextualized’ activities. There is very little in common between learning the number of sepals and petals in a flower - a process that in practice only involves relating the two followed by straightforward memorization - and learning population genetics, which relies chiefly on a highly abstract deductive approach and on the concept of regulation, which necessitates a change or paradigm.
Similarly, this research reveals that the learning process mobilizes several levels of mental organization, apparently quite different at first sight, as well as a considerable number of self-regulation loops. To want to explain everything within a single theoretical framework is rather a tall order, especially as the various constructivist models have been developed in highly simplified contexts. If we take the learning of the concepts of energy, particle structures or genes, for example, the process does not depend on cognitive structures as defined by Ausubel or Piaget. Subjects who have attained high levels of abstraction may reason out new subject matter in the same way as young children. It is not only an operational level that is being questioned, but what we call a global conception of the situation, which embraces simultaneously a type of self-questioning, a frame of reference, signifiers and semantic fields, including metaknowledge on the context and the learning process, etc. All these elements influence the way of thinking and learning and are passed over in silence by the constructivist theories.
Similarly, the assimilation of scientific knowledge does not take place solely through ‘reflective’ abstraction. In learning to handle subjects such as systemic analysis or modelling, abstraction may distort the process. In most cases it causes a radical shift: a new factor seldom fits into the sequence of previous knowledge, and thus that knowledge often represents an obstacle to its integration. Wanting to explain everything in terms of ‘cognitive bridge’, ‘assimilation’ or ‘accommodation’ is over-ambitious. A process of deconstruction needs to take place at the same time as any new construction process; the latter then becomes preponderant.3
To enable learners to understand a new model or mobilize a concept, it is necessary to transform their whole mental structure. The framework of their questions has to be completely rearranged, and their web of references to a large extent reworked. These mechanisms never take place at once but need to go through phases of conflict or interference.
Lastly, the various constructivist models say nothing or almost nothing about the social or cultural context of learning processes. They do not allow us to deduce the consequences of situations, resources or environments that foster the act of learning - which is to be expected, since that is not their initial concern. At the most they put forward the idea of ‘maturation’ or ‘regulation’ without specifying the conditions for such activities in actual practice. In 1989 Vinh Bang noted with regret that ‘a psychology of the pupil was still lacking’.4 In reality, it is the whole psychology of learning that remains to be worked out. But is this still psychology in the traditional sense?
The conceptions of learners
Research on the conceptions of learners has shed new light on the question of cognitive learning processes. The specialist in teaching methodology began by characterizing the ‘representations’ as a gap between the learners’ thinking and scientific thought. The term ‘misconception’, widely used in English and American research (for example Novak, 1985, 1987) is significant. Since then studies on the subject have forged ahead and, among specialists, the term ‘conception’ has replaced ‘representation’ (Giordan & de Vecchi, 1987).
Nowadays it is considered that conceptions play a role in the identification of the situation, in the selection of pertinent information, in the processing of that information and in the production of meaning. According to the authors, they appear as ‘tools’, ‘levels of operation’ and ‘thinking strategies’ and are the only means available to the learner to apprehend reality, the subject matter of the teaching or the information content (Novak, 1984,1985; Host, 1977; Lucas, 1986).
Conceptions are interpreted less as components of a stock of information stored for subsequent use than as ‘a kind of decoder’ by means of which the learner is able to understand the world around him or her (Simpson et al., 1982; Osborne & Gilbert, 1980; Osborne & Wittrock, 1983; Osborne & Freyberg, 1985). In consequence, they are of considerable importance in teaching and mediation. They open the way, it seems, to the tackling of new issues, the interpretation of situations, the resolving of problems, the provision of explanations and the making of forecasts. It is by their means that the learner will select information, give that information a meaning - possibly in conformity with the scientific knowledge of reference - understand it, integrate it and hence ‘understand, learn’ (Giordan & de Vecchi, 1987; Driver, Guesne & Tiberghien, 1985) and mobilize knowledge (Giordan, 1994).
CHARACTERISTICS OF CONCEPTIONS
Conceptions are always anchored in questioning. They appear to exist only in relation to a problem which may be implicit. Once framed, the conception frequently leads to reformulation of the problem. In addition, the conception is determined by four other interacting parameters, namely the frame of reference, the operational constants, the semantic field and the signifiers.
The frame of reference comprises all the previous integrated knowledge, which, when assembled and activated, gives a shape and a meaning to the conception. This frame leads the learner to ask himself or herself direct questions and provides the context (information and other conceptions) that explains the production and presentation of the conception.
The operational constants comprise all the underlying mental operations that establish relationships between the elements of the frame of reference, make the conception function and if necessary transform it on the basis of newly recovered information. It is also these constants that regulate the conception by interacting with the frame of reference.
The signifiers are all the signs, traces, symbols and other forms of language (natural, mathematical, graphic, diagrammatic, modellized, etc.) used to produce and explain the conception.
Lastly, the semantic field is the web of meanings deduced from the preceding elements. Its nodes represent the frame of reference, and its linkages may be equated with mental operations. It is through that field that the meaning of the conception emerges.
HOW A CONCEPTION FUNCTIONS
There are several aspects to a conception: informational, operational, relational, dubitative and organizational. One discernible function is conservation of a piece or body of knowledge, including practical knowledge. Such memorization is not direct but shaped by integration into a structure. A conception organizes information and stands as the trace of previous activity.
This function, however, cannot be likened to a mere memory. The structured and conserved information is subsequently re-used in new situations. The conceptions are transformed by the situation that activates them, being continually reshaped in order to be ‘in phase’ with the new context.
Conception thus makes recall possible but is primarily concerned with identifying the situation and selecting relevant information. Events, context and perceived messages feed in external elements (new information) and activate internal ones (memorized knowledge). Their importance in knowledge construction mechanisms is apparent: the acquisition of a piece of knowledge signifies advance from a previous conception to another conception that is more relevant to the situation.
FIGURE 3: Different conceptions about fecundation and the making of a baby (children from 10 to 12 years old)
A second important function is the constitution of relationships and even systematic arrangement. The individual is constantly seeking, at least when he or she is personally concerned, to regroup all the elements of the knowledge at his or her command in a field or with respect to a particular question. However, the relationships observed are in most cases incomplete or disparate in comparison with those established in a scientific framework.5
Lastly, conceptions structure and organize reality. They are applied to situations so as to enable the learner to raise problems, carry out various activities, conceive new methods of procedure to be applied, and so forth. They are the signs of a model and of a comprehensive method of approach in response to a problem area. They are in fact cognitive strategies that the learner implements in order to select pertinent information to structure and organize reality. They refer back not only to the elements the learner will mobilize directly in order to explain, foresee or act but also to the individual’s personal history, including his or her ideology, social stereotypes and even fantasies.
FIGURE 5: Decoding obstacles by using the allosteric learning model grid
Conceptions must therefore not be interpreted as accumulations of past information or parts of a stock of information simply stored for subsequent consultation. They represent first of all a mobilization of known elements for the purpose of explanation, questioning, prediction or the carrying out of a simulated or real action.
In this mobilization, based on the pupil’s experience in the traditional sense, the learner builds up an ‘analytical grid’ of reality, a sort of decoder that will enable him or her to understand the surrounding world, tackle new problems, interpret new situations, reason through a difficulty and come up with an answer that he or she thinks provides an explanation. It is also on the basis of this ‘tool’ that the learner will select information from outside for possible inclusion and integration.
Models of the learning process
The learner’s conceptions thus lie at the heart of the learning process. They are involved in the various interrelations existing between the information, operations and processes available to an individual and those that individual will come across throughout life. On these elements he or she will build up new knowledge and in doing so shape his or her future modes of conduct.
At this point the research into teaching methods came up against a major problem: how can a teacher ‘utilize’ the conceptions of learners in his or her teaching practices? Should he or she help the learners to enrich their conceptions and/or displace them? Should he or she imperatively begin by refuting them? Can he or she transform them? And if so, by means of what teaching strategy and with what teaching aids?
THE ALLOSTERIC LEARNING MODEL
In the face of the marked shortcomings of constructivist models in this area, various teaching models have been suggested. One of them, known as the allosteric learning model, conceived by Giordan and de Vecchi (1987) and developed by Giordan (1989), has aroused international interest. It allows a series of conditions conducive to the generation of relevant learning processes to be inferred. Indeed, it is precisely that aspect, known as the learning environment, which is most frequently made use of.
When applied to specific bodies of knowledge, the allosteric learning model first of all makes it possible to decode the processes, grouped under the customary terms of understanding and learning, in the form of entities of a systemic and multistratified nature. The self-regulation loops and levels of integration are highlighted and, at the same time, the various obstacles are ironed out and explained.6
At the functional level, this model tends to reconcile the paradoxical and contradictory aspects inherent in any learning process. All knowledge that has been mastered is both an extension of previous knowledge, which provides the framework for questioning, references and meaning, and is at the same time a discontinuity with that knowledge, at least by diverting or transforming the questioning. One learns ‘thanks to’, as Gagné writes, ‘on the basis of (Ausubel) and ‘with’ (Piaget) but also ‘against’ (Bachelard) the functional knowledge in the ‘head’ of the learner. The model shows that learning is a matter of approximation, interest, confrontation, decontextualization, interconnection, discontinuity, alternation, emergence, threshold, detachment and above all mobilization.
In point of fact, the allosteric model reveals that the success of any learning process depends on a transformation of conceptions. All knowledge acquisition stems from complex formative activities: the learners match the new information with their present knowledge, now mobilized, and perceive new meanings that come closer to answering the questions or problems glimpsed. They then create for themselves what we call ‘active conceptual sites’; these are like interaction structures and play a leading role in organizing the new information and developing the new conceptual network.
Such a process is never simple; neither is it neutral where the learner is concerned. It could even be considered a disagreeable process. The conception mobilized by the learner provides him or her with an explanation, and any change is perceived as a threat because it changes the meaning of past experience. The conception serves as both an integrator of and a powerful resistance to any new fact that contradicts the established system of explanations. Furthermore, the learner has to exert deliberate control, at various levels itemized by him or her, over his or her activity and the processes that govern.
A TEACHING ENVIRONMENT
Apart from the description of cognitive strategies7, the main contribution of the allosteric learning model is of a pedagogical nature. It shows that only the learner is able to learn and that he or she cannot do it alone through his or her own mental structures. This approach can be considerably enhanced by an interactive series of parameters, called a teaching environment, placed at the disposal of the learner.
Between the learner and the knowledge-object, a system of multiple interrelations has to be established. This system never arises spontaneously; it is highly improbable that a learner will be able to ‘discover’ for himself or herself all the elements capable of transforming his or her questioning or of facilitating the establishment of networks.
FIGURE 7: Use of the allosteric model for the concept of circulation
At the beginning of any learning process, for example, one must be able to introduce one or more ‘discords’ that jog the cognitive network of mobilized conceptions. This dissonance creates a tension that breaks or displaces the fragile equilibrium achieved by the brain. Only such dissonance can allow progress to be made.8 Without it, learners have no reason to change their ideas or way of doing things. Similarly, it must also be possible to motivate them or interest them in the teaching situation proposed.
Learners should then find themselves faced with a number of significant elements, such as documents, experiments or arguments, that raise questions and lead them to stand back a little, reformulate their ideas or argue them through. Similarly, a number of simplified formal aids are required, such as symbols, graphs, diagrams or models, that can be integrated into their approach to help them think.
It should be added that a new formulation of what the learner knows will not displace a previous one unless the learner finds it interesting and learns how to make it work. Further exposure to appropriate situations and selected information is a profitable means of mobilizing knowledge in this process. Lastly, some knowledge about knowledge is also necessary. It enables learners to situate their approaches, see them with a certain detachment and clarify the practical relevance of that knowledge.
In each case, the allosteric learning model provides tools for decoding the constraints and predicting the situations, activities, interventions and resources that facilitate the learning process.
The allosteric learning model has confirmed a new relationship to knowledge and new functions for the teacher. The latter’s importance no longer lies in what he or she says or in his or her introductory demonstrations; the effectiveness of his or her action is always to be found in a context of interaction with the learner’s learning strategies. It lies primarily in the teacher’s regulation of the act of learning, his/her capacity to arouse interest, and his/her skill at providing pointers for learners’ guidance and aids to conceptualization.
1. This third tradition is always regarded as recent: people continue to call it the new pedagogy. And yet our studies of the history of education show it to be a very ancient approach. Attempts along these lines were already defended by Montaigne and Rousseau and widely explored in the nineteenth and early twentieth centuries.
2. Today it is necessary to add the recently discovered contributions of Vygotsky (1967, 1978) and those of the cognitive science movements.
3. Construction and deconstruction are interactive: the new knowledge takes its proper place when the old knowledge is seen to be outdated. There is a time when they cohabit in different problem areas.
4. Vinh Bang, Introduction, in: A. Giordan, A. Henriques and Vinh Bang, Psychologie génétique et didactique des sciences [Genetic psychology and science teaching], Bern, Peter Lang, 1989.
5. Children do not spontaneously relate ‘oviparous reproduction’ (as with hens) and ‘viviparous reproduction’ (as in human sexuality). For them, these are two types of mechanism that are strictly different, with no common denominator. In one case procreation is focused on the (highly visible) egg and in the other on the spermatozoon.
6. A learning process never takes place without a series of difficulties that are obstacles to be overcome. These obstacles may be extremely diverse in nature - insufficient information, inadequate mental training, difficulty in carrying out operations, or simply the learners’ lack of self-confidence in their ability to handle a problem or find a way forward.
7. The learning process is rather the prototype of a highly complex organized system that cannot be broken down into a few guidelines. It functions more like a jungle ecosystem than a computer, even of the most recent generation!
8. These dissonances are less and less easy to accept when the pupil has much more experience and the subject matter is well known. Any change represents a discontinuity and takes place in a sort of crisis, which may sometimes reach identity crisis proportions when an individual invests everything in what he or she knows and does. The change is smoother when there are signs of a fresh equilibrium in the offing.
Ausubel, D.P., et al. 1968. Educational psychology: a cognitive view. Orlando, FL, Holt, Rinehart & Winston. 238 p.
Bachelard, G. 1934. Le nouvel esprit scientifique [The new scientific approach]. Paris, PUF.
Bachelard, G. 1938. La formation de l’esprit scientifique [Training for the scientific approach]. Paris, Vrin.
Bruner, J.S. 1966. Toward a theory of instruction. Cambridge, MA, Belknap Press of Harvard University Press. 176 p.
Doise, W., et al. 1975. Social interaction and the development of cognitive operations. In: European journal of social psychology (The Hague), vol. 5, no. 3, p. 367-83.
Doise, W., et al. 1985. Le développement social de l’intelligence. Aperçu historique [The social development of intelligence]. In: Mugny, G., ed. Psychologie sociale du développement cognitif. Berne, Peter Lang, p. 39-55.
Driver, R.; Guesne, E.; Tiberghien, A., eds. 1985. Children’s ideas in science. Philadelphia, Open University Press. 208 p.
Gagné, R.M. 1965. The conditions of learning. Orlando, FL, Holt, Rinehart & Winston. 308 p.
Gagné, R.M. 1976. Les principles fondamentaux de l’apprentissage: application a l’enseignement [The basic principles of learning: applied to education]. Montreal, HRW. 148 p.
Giordan, A., et al. 1987. Une pédagogie pour les sciences expérimentales. Paris, Centurion. 280 p.
Giordan, A.; de Vecchi, G. 1987. Les origines du savoir [The origins of knowledge]. Paris, Delachaux.
Giordan, A.; de Vecchi, G. 1988. An allosteric learning model. Actes IUBS-CBE 1988, p. 12-32. Sydney. (Revised at Moscow meeting, Actes IUBS-CBE, 1989).
Giordan, A.; de Vecchi, G. 1994. Le modèle allostérique et les théories contemporaines sur l’apprentissage [The allosteric model and modern learning theories]. In: Giordan, A.; Girault, Y.; Clement, P., eds. Conceptions et connaissances. Bern, Peter Lang. 319 p.
Holland, J.G.; Skinner, B.F. 1961. The analysis of behavior. Toronto, McGraw Hill. 337 p.
Host, V. Place des procédures d’apprentissage ‘spontané’ dans la formation scientifique [The place of spontaneous learning procedures in scientific training]. Bulletin de liaison INRP -Section Sciences (Paris), vol. 17, 1977.
Locke, J. 1693. Some thoughts concerning education. Oxford, Clarendon.
Lucas, A.M. 1986. Tendencias en la investigación sobre la enseñanza/aprendizaje de la biología [Research trends in biology teaching/learning]. Enseñanza de las ciencias (Barcelona), no. 4, p. 189-98.
Novak J.D., ed. 1984. Learning how to learn. Cambridge, Cambridge University Press, 1984.
Novak J.D., ed. 1985. Metalearning and metaknowledge: strategies to help students to learn how to learn. In: West, L.H.T.; Leon Pines, A., eds. Cognitive structure and conceptual change. New York, Academic Press, p. 189-209.
Novak J.D., ed. 1987. Misconceptions and educational strategies in science and mathematics. In: Proceedings of the second international seminar ‘Misconceptions and education’, 26-29 July, 1987. Ithaca, Cornell University.
Osborne, R.J.; Gilbert, J.K. 1980. A method for investigating concept understanding in science. European journal of science education (The Hague), vol. 2, no. 3, p. 311-21.
Osborne, R.J.; Wittrock, M.C. 1983. Learning science: a generative process. In: Science education (London), vol. 67, no. 4, p. 489-508.
Osborne, R.J.; Freyberg, P. 1985. Learning in science: the implications of children’s science. Portsmouth, Heinemann. 258 p.
Piaget, J. 1950. The psychology of intelligence. London, Routledge & Kegan Paul.
Piaget, J. 1971. Science of education and the psychology of the child. London, Longmans.
Piaget, J.; Inhelder, B. 1969. The psychology of the child. New York, Basic Books.
Perret-Clermont, A.N. 1979. La construction de l’intelligence dans l’interaction sociale [The construction of intelligence through social interaction]. Bern, Peter Lang. 244 p. (Collection Exploration.)
Perret-Clermont, A.N. 1980. Social interaction and cognitive development in children. London, Academic Press. 206 p.
Simpson, M., et al. 1982. Availability of prerequisite concepts for learning biology at certificate level. Journal of biological education (London), vol. 16, no. 1. p. 65-72.
Skinner, B.F 1968. The technology of teaching. New York, Appleton Century Crofts. 271 p.
Vygotsky, L. 1967. Thought and language. Cambridge, MA, MIT Press. 168 p.
Vygotsky, L. 1978. Mind and society: the development of higher psychological processes. Cambridge, MA, Harvard University Press. 159 p.
Vinh Bang. 1989. Introduction. In: Giordan, A.; Henriquez, A.; Vinh Bang, eds. Psychologie génétique et didactique des sciences. Bern, Peter Lang. 297 p.
[Ukrainian] [English] [Russian]