![]()
It is a redoubtable honour to have been invited by our colleague Patrick Lemaire to write a chapter about the European reception of Bob Sieglerâs work. Bob Sieglerâs contribution is so impressive and universally acknowledged, his views have been so influential in our understanding of childrenâs thinking and development, that it would certainly be presumptuous to claim that we can provide an exact reflection of the impact such a contribution has had on European developmental psychology. More modestly, we shall propose in the following sections our own view of the uniqueness of Bob Sieglerâs approach and the way in which it questioned and went beyond European developmental psychology, necessarily strongly influenced by Piagetâs work, especially for French-speaking psychologists.
In the preface of his book Emerging Minds: The Process of Change in Childrenâs Thinking, published in 1996, Bob Siegler noted that âevery intellectual discipline is an attempt to answer a few basic questions. For developmental psychology, one of these questions, perhaps the question, is âHow does change occur?â (Siegler, 1996, v). Due to the extreme state of immaturity in which human infants come into the world, human development involves so many dramatic transformations from birth to adulthood that one can only agree with Siegler that the question of how these changes occur is of paramount importance for psychology. The most prominent and penetrating (the best ever?) attempt to answer this question in the European psychological tradition is undoubtedly the Piagetian genetic epistemology. How change occurs was a constant preoccupation for Piaget, who dedicated two books to this question, a first in 1957 entitled Logic and Equilibrium, and a second in 1975 entitled The Equilibration of Cognitive Structures: The Central Problem of Intellectual Development. This latter essay is the clearest explanation that Piaget gave of the mechanisms that generate changes in childrenâs intelligence.
The mechanism of equilibration in Piagetâs theory
Piagetâs main idea in studying the equilibration of cognitive structures was that development does not simply result from perception or experience with objects, nor from an innate and preformed programming within the subject, but from the interactions between the subject and the object. These interactions were assumed to lead to regulations that compensate for the disequilibrium that occurs in cognitive functioning due to the fact that a subjectâs actions produce observables in objects that cannot be entirely anticipated. Importantly, these regulations never lead to static forms of equilibrium, but to constant re-equilibrations that are the motor of development as they improve the existing structures through a process known as majoring equilibration. The diagram reproduced in Figure 1.1 illustrates this mechanism. Observables related to the objects (Obs. O) lead to a process of awareness of the subjectâs own actions by which they were produced, this awareness resulting in observables related to the subject himself and his actions (Obs. S). The new perspective provided by the Obs. O on the Obs. S allows for new coordinations of the subjectâs schemes (Coord. S). However, as long as the childâs comprehension is not sufficiently accurate, these new coordinations of schemas of action (structures) lead to the discovery of new relations and coordinations between objects (Coord. O), which in turn produce new observables (Obs. O). Thus, immature levels of understanding lead to perturbations, contradictions, and the discovery of new observables. The compromised equilibrium can be re-established through three levels of compensation labelled α, ÎČ, and Îł by Piaget. The first consists of a mere neglect of the perturbation leading to ignoring or minimizing the importance of the corresponding observable (level α, e.g., in the task of conservation of liquids, neglecting the width of the glasses to focus on the level of water). The second, level ÎČ, integrates the observable in a slightly modified system as a new but innocuous variable, whereas the third type of compensation (level Îł) makes the new observable no longer disruptive by its complete integration in a new structure allowing for a genuine anticipation of its possible variations. This restructuring of a developmental level n provides new observables related to both the subject and the object that will be integrated at a developmental level n + 1 (Figure 1.2).
Despite its complexity and technical nature, this account of how change and novelty occur in development remains rather vague. Apart from the fact that it is not always easy to identify what the different concepts of observables (Obs.) and coordinations (Coord.) refer to, the real mechanism by which development shifts from level n to level n + 1 is underspecified. How does the compensation mechanism move from level α to level Îł? If the disequilibrium produced by a discrepant and contradictory observable can be compensated for by ignoring the perturbation it creates (level α), why are other levels of compensation necessary? It seems that we are facing what Siegler (1996, p. 5) amusingly called âthe mystery of the immaculate transitionâ. Moreover, the equilibration process described by Piaget depends on noticing observables on objects, a process that might be dependent on contingencies and result in a domain-specific rather than a domain-general (i.e., structural) development, as Piaget favoured. In the same way, the recursive process of constant re-equilibration points toward continuous rather than stage-like development, with phases of abrupt change followed by protracted stable states.
Finally, it remains unclear why the coordinations of a subjectâs schemes and actions should take the form of logical structures. It could be assumed that only logical structures can lead to the rational organization of Coord. O from which no contradictory Obs. O could be drawn, resulting in more stable equilibrium. However, are logical structures necessary for producing genuine understanding and coherent predictions? This is what Bob Siegler denied in what seems to be his very first published article (Siegler & Liebert, 1972).
Rules instead of logical structures
This first study addressed the question of the conservation of liquids, and more precisely a problem that agitated developmental psychologists at that time (end of the 1960s, beginning of the 1970s), which was the efficiency of logical training on conservation tasks. Siegler and Liebert (1972) noted that while this training proved effective for conservation of number or length in children aged four or five, only minimal success was achieved with the conservation of liquids. Siegler and Liebert surmised that the problem was in conceptualizing the conservation of liquid quantity as a logic-based task. They reasoned that the fact that the quantity of liquid remains constant through some transformations is more a matter of experiential or empirical discovery rather than a matter of logic. After all, transforming a square piece of clay into a ball does not change the amount of clay or its weight, but certainly changes its surface area despite the reversibility of the operation and the fact that nothing has been added or removed. Thus, they suggested that conservation judgments reflect familiarity with, and adoption of, rules rather than logical manipulations.
They tested this hypothesis by designing a training program for conservation of liquids in which different rules were taught to six-year-old non-conserving children. These rules concerned how specific transformations result in specific outcomes: that just pouring all the water from a narrow to a wider glass results in the same amount of water, despite the decrease in water level, that pouring all the water plus some additional water results in more water, and that pouring some but not all of the water results in less water. Interestingly, the non-conservation situations could be made perceptually consistent or discrepant (e.g., pouring all the water plus some water could still result in a lower level of water in the wide glass than in the narrow one).
In a rule-only group, children were told the appropriate rule after each trial, whereas in a rule + feedback group, children were additionally provided with feedback about their responses, these groups being compared with a feedback group that received feedback on the correctness of their answers but no rules and a control group that performed the same problems but received neither rules nor feedback about their responses.
Presenting children with both the appropriate rule and feedback led them to correctly classify the transformations that do and do not affect the quantity of liquid in more than 70% of the trials, thus exhibiting conservation. The other conditions were less effective in producing conservation learning. This first study provided evidence that conceiving of intellectual development as the acquisition of appropriate rules refined through feedback and analysis was a fruitful avenue of research.
Bob Sieglerâs conceptualization of cognitive development in terms of rules was continued during the 1970s, probably culminating in the influential book, Childrenâs Thinking: What Develops?, that he edited in 1978 and in a monograph Developmental Sequences within and between Concepts that he published in 1981. He offered in these publications reviews of his studies about the balance scale, the projection of shadows, and the probability problems, as well as conservation of liquid quantity, solid quantity, and number, establishing that thinking about these problems develops in children through the successive adoption of more and more sophisticated rules. For example, for the balance scale, Rule I adopted by the youngest children considers only the numbers of weights in each side of the fulcrum, predicting that if they are the same, the scale will balance, otherwise the side with the greater number will go down. Rule II, adopted by older children, incorporates Rule I but stipulates that, when weights are the same on both sides, distance is taken into account. Rule III incorporates both the effects of weight and distance. However, when the two dimensions lead to contradictory predictions, for example when the lower number of weights is farther from the fulcrum, children âmuddle throughâ, as Siegler says, or simply guess. Finally, Rule IV, which reflects the mature comprehension of the physical system, involves computing the torques on each side by multiplying the number of weights by the distance from the fulcrum and comparing the two products. A similar progression was observed for the projection of shadows and for the probability tasks, though the progression for the conservation problems was somewhat different.
Siegler tested this model by presenting children aged 5, 9, 13, and 17 with a variety of problem types designed in such a way that the use of a given rule would lead to a specific pattern of predictions. The responses of 107 of the 120 children tested conformed to one of the four rules that developmentally followed each other in the predicted order.
Interestingly, Sieglerâs investigations went further and addressed the question of the origins of knowledge on this particular problem. For this purpose, he compared two groups of children aged five and eight who spontaneously used Rule I and analyzed their reactions when trained to solve conflict problems corresponding to two or more rules beyond their initial level (e.g., Rule III and Rule IV problems in which the side with the smaller number of weights goes down or in which the scale balances with different numbers of weights in each side). Surprisingly, although a majority of older children took advantage of the feedback and moved to Rule III, not one young child moved beyond Rule I. The explanation of this difference in sensitivity to experience was provided by testing the encoding hypothesis advanced by Siegler (1976), which assumes that the younger children disregard the effect of distance because they do not even encode it, whereas older children encode both weight and distance. This is exactly what was observed.
The same hypothesis shed light on the origins of scientific reasoning. The large majority of five-year-olds are Rule I users, but how do children get to that level? Siegler observed that although Rule I is used by virtually all the five-year-olds, only half of the four-year-olds and almost none of the three-year-olds used it. The other three- and four-year-olds did not appear to use any systematic rule. Siegler (1978, p. 146) concluded âsomething important happens between ages three and five. We might call it the development of systematic strategies or perhaps the development of rule-governednessâ.
What is the origin of these differences? It appeared that three- and four-year-old children differed in their capacity to benefit from feedback. When presenting three- and four-year-olds who did not use Rule I with appropriate feedback, use of the rule by the four-year-olds substantially increased, whereas three-year-oldsâ performance remained unchanged. This again appeared to reflect differences in encoding. When explicitly instructed to encode weights and exposed to appropriate feedbacks, even the younger children proved able to learn and use Rule I, as five-year-old children spontaneously do.
Overall, the rule-based approach allied with the encoding hypothesis proved effective in explaining childrenâs reasoning and its development. The research revealed that five-year-old children, who were described by Piaget as pre-operational, irrational, and caught up in egocentrism, have actually reached the culminating point of a first developmental phase, lasting from birth until age five, during which reasoning progressively becomes coherent and rule-governed. These studies raised the question: Is the rule-based approach that different from the equilibration model?
Piagetian schemes and information-processing rules
At a first glance, the information processing background of the rule-based approach makes it totally different from the Piagetian structural and logic-based constructivism, and rule acquisition seems to have nothing to do with the equilibration process. However, let us come back to the way Piaget formali...
