![]()
PART I
The nature of concepts and conceptual change
Editor: Bruce Sherin
Orientation
This section addresses the nature of concepts and conceptual change. As such, this section is very important to the larger endeavor undertaken in this book, which is to work toward a convergence in research on conceptual change. As a microcosm of the larger volume, the chapters here illustrate some of the tensions to be negotiated, as well as the prospects for synthesis.
The first three chapters in this section represent a trio of voices that have been very prominent in research on conceptual change in the context of school science learning. The first chapter, from Andrea diSessa, describes the knowledge-in-pieces (KiP) theoretical perspective. diSessa illustrates this theoretical perspective in a discussion of his models of two types of knowledge, phenomenological primitives (p-prims), and coordination classes. In the second chapter, Stella Vosniadou lays out her own theoretical perspective. She argues that, in response to phenomenal and cultural experience, people develop what she calls framework theories. Then, when students encounter formal instruction, the existence of the framework theory leads to the formation, by individuals of āsyntheticā conceptions. Vosniadou summarizes a range of empirical work that supports this theoretical perspective, and she also responds to a range of criticisms. In the third chapter, J. Bryan Henderson, Elon Langbeheim, and Michelene Chi ask the question: What makes some misconceptions robust? Their answer is that misconceptions tend to be robust when they involve ontological miscategorization. They argue, in particular, that many robust misconceptions result from the miscategorization, by students, of emergent processes as sequential processes. Furthermore, learning is complicated by the fact that learners generally lack the emergent process category.
The remaining three chapters broaden the discussion, each in a different manner. Cecilia Lundholm extends the discussion of conceptual change to school learning in the social sciences. She argues that conceptual change in the social sciences involves many of the same challenges as in disciplines that have been the traditional focus of research. But there are, in addition, challenges that are somewhat unique. Studentsā values and identities are relevant factors in conceptual change in the social sciences. Also, social science disciplines tend to include value-laden and contestable assumptions. In the next chapter, Ananda Marin, Douglas Medin, and bethany ojalehto argue that concepts should be treated as embedded within epistemological orientations, which can differ across cultural communities and contexts. As an illustration, they look at how an underlying view of the relationship between humans and nature influences conceptual organization. To support their points, they draw on research on the organization of folkbiological knowledge. The final chapter, by Geoffrey Saxe, broadens the discussion still further, to look at the conceptual change of communities. Saxe looks at the changing meaning of a word, āfu,ā used by the Oksapmin of Papua, New Guinea.
Clearly, the chapters here are divergent in many respects. The broad nature of the phenomena understood differs dramatically. Some of the chapters, for example, look at learning in response to formal science instruction, while others look at developmental and community-level phenomena. Furthermore, the theoretical perspectives employed are diverse. Nonetheless, as discussed in the synthesis by Bruce Sherin, it is possible to view the larger body of work as complementary; we can see the work conducted as small parts of the same larger endeavor. Furthermore, the chapters contain some of the seeds of a theoretical synthesis.
![]()
1
KNOWLEDGE IN PIECES
An evolving framework for understanding knowing and learning
Andrea A. diSessa
Empirical focus
Instead of providing a scholarly setting for the Knowledge in Pieces (KiP) perspective, it is briefer and as insightful to observe that my studies of conceptual change emerged from a passionate and sustained personal interest in how people (students, children, adults) naturally think about situations that might also be construed from the viewpoint of professional science. I have been enchanted by the richness, flexibility, great nuance, and often wonderful insightfulness of everyday thinking. I have systematically sought topics to discuss, situations that are accessible, but also somewhat problematic, so as to engage extended thinking and reflection. āExplanationā is at the center of this; āproblem-solvingā is peripheral. Problem-solving in school often entails students grasping at straws conceptually and just juggling the combinatorics of variables in equations. Understanding mathematics and its use in science is a worthy topic, but I believe it is secondary to deep qualitative, conceptual understanding.
Focusing on reasoning about less technical situations has proved immensely enlightening of schooled learning. Of course, there is a lot that is new in school, but the assumption that learning is substantially a recrafting of naĆÆve conceptual resources has been among the most robust and productive assumptions in the history of learning studies.
Early on, I learned that just talking with people was a superior way of āseeingā their thinking. I learned a few lessons from this, which I think still escape the attention of many conceptual change researchers: (1) People have a vast repertoire of ways to think about many scientific situations; it might be that requiring them to engage with the most inscrutable ones (often forced on them in school) is mainly beside the point. (2) Their thinking is subtly tied to circumstances and frames of mind; slight shifts of perspective can induce dramatic changes. (3) Every person with whom Iāve engaged seems to be one-of-a-kind. Hence, there is not only a variety in how individuals can consider a given situation, but thereās great variation across individuals.
The technical version of ājust talking with peopleā is clinical interviewing, which has been a mainstay of my work. Of course, as someone with a deep professional and personal commitment to education, instruction and learning are also important foci, even if overt instruction almost never enters the clinical context. I see the following connections between just ātalking with peopleā and instruction:
Input to conjectures and expectations: Knowing how people think within their zone of felt competence provides important, general and often very specific conjectures about how they can learn from instruction.
Input to design: Some of the insights from item 1, above, can instigate new avenues of instruction based on using powerful but underexplored and undervalued intellectual resources. This means that such research affects the very goals of instruction (what and when topics should be taught, and how they should be construed), in addition to instructional strategies.
Input to observation: Looking at learning-from-schooling shows indelible earmarks of both the productive and sometimes less productive roles of pre-instructional knowledge.
Recent work of our group exemplifies all these: (1) From studying studentsā ideas, we conjectured that it would be possible for middle school students to learn about an exotic topic, dynamical systems theoryāand with substantially increased enthusiasm compared to traditional topics. (2) We conjectured that teaching about equilibration could be based on a prominent but often maligned intuitive idea. (3) Results of the instructional design validated that conjecture. But with the help of a prior and parallel clinical study, we observed many details not evident in our conjecture, which would have been unrecoverable from classroom observation, student work, and test data alone (diSessa, 2014).
Theoretical orientation
KiP theory involves problematizing the concept of knowledge, gradually developing a modern, detailed, and empirically supported replacement for previous commonsensical, inexplicit, philosophical, or other views of knowledge that are not up to deeply engaging the specifics of learning (diSessa, 2016). In particular, the concept of āconceptā has too often evaded critical consideration, or versions of the idea have shown no purchase on designing or understanding instruction. The technical concepts of science and mathematics turn out to behave very differently from everyday concepts, such as ābirdā or āanimal,ā which have been a mainstay of psychological study. Traditional concepts of āconceptā also turn out to be completely ineffective in understanding the form and function of intuitive ideas and their powerful roles in learning.
The KiP approach involves two strategies. First, divide-and-conquerālooking to define, elaborate, and validate a variety of new knowledge terms that enfold different aspects of knowing, such as everyday intuitive understanding, or, in contrast, full-blown professional concepts. I will exemplify KiP approaches to both such kinds of knowledge.
The second strategy is an incremental modeling approach to developing a science of knowledge. We need to develop new theory slowly and with due respect for how much we do not know. This entails two characteristics. Explicit and consequentialāWe seek to articulate specific models of various kinds of knowledge, models that have clear consequences and can be thoroughly tested against the realities of learning and instruction. Continuous improvementāWe must simultaneously cultivate an understanding of the limits of these models, which constitute a frontier for future development.
KiP as a whole, then, aims to provide a coherent global empirical and theoretical framework in which to design and continuously improve a family of models of various kinds of knowledge, models with strong empirical tractability and powerful consequences.
Focal themes
I highlight two KiP themes in this chapter.
Integrated analyses at multiple time-scales: Conceptual change research (and KiP in particular) is distinguished by a strong focus on learning that can embrace an extended learning trajectory (years) with many difficulties for students and challenges for teachers. What is more distinctive of KiP is a focus on process data and analyses. We seek high-resolution accounts of thinking-and-learning-in-the-moment. Evident long-term changes in understanding must be happening sometime, and we embrace the task of saying exactly when something is being learned, how that is happening, and how such events accumulate over the long term. The complementary āmicroā focus is rare in conceptual change work, especially within the tradition of developmental psychology. In education, also, before-and-after studies of learning are rarely augmented with process data and analyses. In sum, KiP accepts the challenge to integrate short and long time-scale descriptions and explanations.
Encompassing Diversity in Learning: I mentioned that I felt that every subject in my clinical interviews thought differently. If trueāwhich I expectāthis has strong consequences for learning, especially given that many theories of conceptual change emphasize (1) generic views of before-and-after states (āthe naĆÆve theoryā vs. an assumed-to-be uniform ānormative scienceā), and (2) generic paths to understanding to the point of severely marginalizing individual differences.
Two models: illustrative data and analysis
The remainder of this chapter concentrates on concretizing and exemplifying the generalizations above, both with respect to theory development and with respect to phenomenological focus and empirical methods. I will use the two best-developed and best-known KiP models of knowledge types. Descriptions here, of course, are necessarily bare bones, giving only hints about model details and the breadth of empirical support.
Intuitive knowledge
P-prims are elements of intuitive knowledge that constitute peopleās āsense of mechanism,ā their sense of what happenings are obvious, which are plausible, which are implausible, and how one can explain or refute real or imagined possibilities. Example glosses of p-prims are as follows: (1) increased effort begets greater results; (2) the world is full of competing influences for which the greater āgets its way,ā even if accidental or natural ābalanceā sometimes exists; and (3) the shape of the situation determines the shape of action within it (e.g., orbits around a square planet are nearly square).
We must develop a new model for this kind of knowledge because, empirically, it violates presumptions of standard knowledge types, like beliefs or principles. First, classifying p-prims as true or false (like beliefs or principles) is a category error; p-prims are unclassifiable by standard scientific norms. They workāprescribe verifiable outcomesāin typical situations but always fail in others. Indeed, when they will even be brought to mind is a very delicate consequence of context (both internal: āframe of mindā; or external: the particular sensory presentation of the phenomenon). So, for example, it is inappropriate to say that a person ābelievesā a p-prim, as if it would universally be brought to mind, when relevant, and as if it would always dominate other ways of thinking. Furthermore, students simply cannot consider and reject p-prims (a commonly prescribed learning strategy for āmisconceptionsā). Blocks to āconsiderationā are severeāthere are no common words for p-prims, and people are in general not even aware that they have such ideas; ārejectionā does not make sense for ideas that usually work!
Data and analysis: J, a subject in an extended interview study (diSessa, 1996), was asked to explain what happens when you toss a ball into the air. J responded fluently with a completely normative response: There is only one force in the situation, gravity, which slows the ball down, eventually to reverse its motion and bring it back down. Then the interviewer asked a seemingly innocuous question, āWhat happens at the peak of the toss?ā Rather than responding directly, J completely reformulated her explanation of the toss. She first implicated āair resistanceā as a second force that is competing with gravity to influence the ballās motion, but quickly decided that it really is only gravity that is acting against the upward motion. Finally, restarting her explanation once again, she imputed āa force that you gave the ball with your hand,ā which gradually dies out, leaving gravity to pull the ball downward.
The key to understanding these events so far is that the interviewer ātemptedā J to apply an intuitive idea of balancing and overcoming; he asked about the peak because the change of direction there looks like one influen...
