Constructionism in Practice:

Designing, Learning and Thinking in a Digital World

 

Edited by Yasmin B. Kafai and Mitchel Resnick


Introduction

Nearly 30 years ago, Seymour Papert and colleagues initiated a research project at Massachusetts Institute of Technology (MIT), dedicated to the study of how children think and learn, and to the development of novel educational approaches and technological tools to help children learn new things in new ways.


During the past three decades, that research effort has evolved and grown. One of its technological offspring, the programming language Logo, has been used by tens of millions of schoolchildren all over the world. At the same time, its theoretical foundation, which has become known as "constructionism," has deeply influenced how educators and researchers think about directions for educational reform-and, within that context, about the roles for technology in learning.
Constructionism is both a theory of learning and a strategy for education. It builds on the "constructivist" theories of Jean Piaget, asserting that knowledge is not simply transmitted from teacher to student, but actively constructed by the mind of the learner. Children don't get ideas; they make ideas. Moreover, constructionism suggests that learners are particularly likely to make new ideas when they are actively engaged in making some type of external artifact-be it a robot, a poem, a sand castle, or a computer program-which they can reflect upon and share with others. Thus, constructionism involves two intertwined types of construction: the construction of knowledge in the context of building personally-meaningful artifacts.


In 1991, Papert's research group at the MIT Media Lab published a collection of papers under the title Constructionism, edited by Idit Harel and Seymour Papert. This current collection is, in some ways, a follow-on to the Constructionism book. Many of the ideas and projects presented here grew out of research described in Constructionism. But this collection also addresses new issues and steers into new directions. Constructionism is not a static set of ideas. Consistent with the theory we are writing about, we as researchers are continually reconstructing and elaborating what we mean by constructionism, and continually constructing new educational activities and tools to help us in that effort. For example, readers will notice a greater emphasis on ideas related to "community" in this volume. Of course, those ideas have always been a part of the Logo research effort, as evidenced by Papert's discussion of samba schools in Mindstorms. But they play a more prominent role in many of the papers here.


In some ways, constructionist research is challenged by its success. A decade ago, many constructionist ideas were viewed as radical and out of the mainstream. But today, at educational research conferences, the idea that children actively construct new knowledge is taken almost as gospel. And a growing number of research efforts are studying how children learn through design and invention activities. The challenge is to continue to refine constructionist ideas, and to make sure that these ideas spread not only to members of the educational research community, but also to the classrooms, homes, and community centers where children work, play, and learn.
This collection includes 14 papers, divided into four sections: Perspectives in Constructionism; Learning through Design; Learning in Communities; Learning about Systems. These sections are somewhat arbitrary and possibly misleading, since many of the papers cut across the boundaries. But the section themes provide a general framework for understanding some of the core issues in constructionist research today.

Perspectives in Constructionism

One of the main tenets of constructionism is that learners actively construct and reconstruct knowledge out of their experiences in the world. It places special emphasis on the knowledge construction that takes place when learners are engaged in building objects. Constructionism differs from other learning theories along several dimensions. Whereas most theories describe knowledge acquisition in purely cognitive terms, constructionism sees an important role for affect. It argues that learners are most likely to become intellectually engaged when they are working on personally-meaningful activities and projects. In constructionist learning, forming new relationships with knowledge is as important as forming new representations of knowledge. Constructionism also emphasizes diversity: it recognizes that learners can make connections with knowledge in many different ways. Constructionist learning environments encourage multiple learning styles and multiple representations of knowledge. The chapters in this section explore these core constructionist ideas, further developing the intellectual underpinnings of what we mean by constructionism.


In his paper, "A Word for Learning," Seymour Papert wonders why theories of teaching have received far more attention than theories of learning, and he addresses the fundamental issues of what we mean by the word "learning." He invites us to revisit our notions of learning through examining a personal experience. Using the example of learning flower names, Papert illustrates two important aspects of learning: "The simple moral is that learning explodes when you stay with it: A full year had passed before the effect in my mind reached a critical level for an exponential explosion of growth. The more complex moral is that some domains of knowledge, such as plants, are especially rich in connections and particularly prone to give rise to explosions of knowledge." The time to build personal connections is an essential ingredient strikingly absent from most school learning situations.


The next chapter, "Perspective-Taking and Object Construction" by Edith Ackermann, also addresses the issue of relationships with knowledge. Ackermann brings together Piagetian and situated-cognition approaches, illustrating how the alternation between assimilative and accommodative attitudes punctuates an individual's interactions with the world. Drawing on results from the study of perspective-taking, Ackermann argues that the ability to decenter from one's own standpoint and to take another person's point of view requires the construction of cognitive invariants: a recasting of the world's stabilities that transcends any given viewpoint. According to Ackermann, this separation is a necessary step toward the construction of a deeper understanding; adopting a god's eye view is by no means contrary to situating one's own stance in the world. Ackermann notes that the process of learning includes opposites: forging relationships and creating separations at the same time. One cannot relate without, at times, separating.


In the last chapter, "Elementary School Children's Images of Science," Aaron Brandes introduces the framework of image of science as a tool for understanding and enhancing children's science learning. Science-education research has established that children's science learning depends critically on their ideas about science, scientists, and experimentation. Brandes extends this research to include affective components such as the child's identification with science or alienation from it. He describes three studies with elementary-school students, designed to probe children's images of science. The studies show that children's excitement about science decreases with age, even as their ideas about science become, in many ways, more sophisticated. Furthermore, the process by which new knowledge is generated remains mysterious, leaving most children on the "outside" of science. In his conclusions, Brandes proposes criteria for evaluating science activities, based on his images-of-science framework.

Learning through Design

Constructionist theory suggests a strong connection between design and learning: it asserts that activities involving making, building, or programming-in short, designing-provide a rich context for learning. But design and learning have not always been viewed so closely connected. Theories of design and learning have very different origins. Traditionally, design theorists have been interested primarily in the final product, and in the ways that constraints influence the design of the product. By contrast, learning theorists were interested primarily in process, not product.
Recently, however, theories of learning and design have begun to move toward one another. Both design theorists and learning theorists now view "construction of meaning" as a core process. In this view, design involves building a relationship between designer and object. Designers sort out what objects mean to them or others, and they selectively connect features of an object and features of a context into a coherent unity. The relationships that designers build with objects or situations constitute the core focus of design theory; the final artifact itself is secondary. Design is now viewed as the process through which a designer comes to understand not only objective constraints but also subjective meanings.


In this way, design theories overlap with constructivist learning theories: both focus on the construction of meaning. At the same time, learning theorists have begun to pay more attention to the role of product and artifact. Constructionist theory goes beyond Piaget's constructivism in its emphasis on artifacts, asserting that meaning-construction happens particularly well when learners are engaged in building external and sharable artifacts. Thus, there is a convergence in the fields of design and learning, with a natural intersection in the study of learning-through-design. The chapters in this section aim to contribute to the study of design and learning, exploring the nature of learning through design, its cultural contexts, activities, and tools.


In the first chapter in this section, "Learning Design by Making Games," Yasmin Kafai examines learning through design in a rather unusual context: video games. These games are a central part of late 20th-century children's culture. In the playing of video games, children mobilize energies which many educators, parents, and researchers wish would be dedicated to learning. Whereas most conventional efforts have sought to harness these energies in the form of playing educational games, Kafai presents a different perspective: in children making their own video games instead of playing games made by others. In a six-month project, a class of fourth graders was asked to design computer games to teach fractions to younger students. In the analysis of the students' learning experiences, Kafai focuses on students' development of project management skills-how students approached and managed this complex task. The paper analyzes students' first steps in beginning the project and framing the design task, and their development of programming strategies. The results provide insights into the project's evolution, the different approaches chosen by students, and potential support structures for young designers.


In her second paper, "Electronic Play Worlds," Kafai takes a close look at the video game artifacts designed by children to address the issue of gender differences. In studying the popularity of video games among children, many researchers have focused on explaining why boys in particular love playing these games. In this paper, Kafai uses the process of game design to elicit children's ideas and interests in video games. The students' choices are analyzed and discussed in regard to their game themes, worlds, characters, interactions, feedback, and narrative. The design of the games reveal clear gender-related differences and provide strong support for design activities as a way for children to express their personal preferences. Conclusions address the potential of game-making environments.


In his chapter, "The Art of Design," Greg Gargarian examines the relationships between design and interpretation, reasoning in dynamic problem contexts, and the role of concept design in artifact design. He sees design (and learning) as a process in which designers customize their working environment around the artifact under design. Furthermore, designers are engaged in the process of situated planning as a means of reducing design problems to conventional ones. To illustrate the intertwined nature of designing and learning, he presents the Textile Designer, a computer-based environment that preserves the qualities found in design environments predating computers. In his final section, he reflects on the nature of future learning environments that allow for the creation of a commonwealth of skills. The discovery villages, as he calls these design-based learning environments, bring together the arts and the sciences, school and play, and work and life.


In the last chapter of this section, "Building and Learning with Programmable Bricks," Randy Sargent, Mitchel Resnick, Fred Martin, and Brian Silverman discuss the applications and implications of the Programmable Brick-a tiny, portable computer embedded inside a LEGO brick, capable of interacting with the physical world in a large variety of ways. Programmable Bricks make possible a wide range of new design activities for students in robotics and ubiquitous computing, a new research field that aims to spread computation throughout the environment, embedding computation in all types of objects and artifacts. The authors describe initial results from three types of applications: building active environments in which children use Programmable Bricks as sensors and regulators; making autonomous creatures that display various behaviors; and personal science experiments to monitor and control everyday phenomena. The authors conclude with a list of 20 things to do with Programmable Bricks.

Learning in Communities

In the minds of many, Rodin's famous sculpture The Thinker provides the prototypical image of thinking: it shows a person, alone, in deep concentration. This image of the "individual thinker" plays a strong role in our culture. In most school classrooms, children must work on worksheets and tests on their own. In the scientific world, Nobel prizes are given to individuals, not teams.
So it is not surprising that computers, when they first entered school classrooms, were used primarily for individualized learning. There were debates about "computer as tutor" vs. "computer as tool." But the unspoken consensus was that the computer was a tutor or a tool for an individual learner.


In recent years, there has been a surge of interest in the social nature of learning. There is a growing appreciation for the role that communities play in the learning process: community members act as collaborators, coaches, audience, and co-constructors of knowledge. In the educational-research literature, there is new attention to communities of practice, knowledge-building communities, and computer support for collaborative learning.
The idea of community has always been present in the constructionist vision. In Mindstorms, written in 1980, Papert discusses the Brazilian samba school as an example of a community of learners. But many of the early constructionist studies focused on the development of the individual learner. It is only in recent years that ideas of community have emerged as a major theme in constructionist research. The chapters in this section examine the nature of learning in many types of communities: classroom communities, urban communities, and virtual communities.


In the first chapter, "Social Constructionism and the Inner City," Alan Shaw focuses on the challenges facing underprivileged urban communities. He describes a set of new tools and activities that he created to support growth and development in a Boston inner-city neighborhood known as Four Corners. Shaw developed a computer networking system called MUSIC (Multi-User Sessions in Community) that neighborhood residents used to discuss community issues and organize community activities. Using a theoretical framework that he calls "social constructionism," Shaw examines how such tools and activities support not only the development of individuals within the community but the development of the community itself.
Whereas Shaw explores constructionist ideas in the context of a proximal (geographically connected) community, Amy Bruckman and Mitchel Resnick explore them in the context of a virtual community on the Internet. In "The MediaMOO Project: Constructionism and Professional Community," Bruckman and Resnick describe a networked virtual-reality environment known as MediaMOO, designed to enhance community among media researchers. In MediaMOO, participants not only interact in a virtual world, they help construct (and continually reconstruct) the world in which they interact. Bruckman and Resnick explore how this combination of construction and community could lead to new possibilities for learning.


Michele Evard examines a different type of network-based community in her chapter: "A Community of Designers: Learning through Exchanging Questions and Answers." Evard examines how a computer-based network allowed elementary-school children to share ideas with one another during an extended design project. Her study focuses on two classes. In one class, children were designing their own video games; another class of children, who had already designed games, acted as consultants on the network. The children exchanged questions and answers with one another through a Usenet-style "newsgroup." Evard analyzes usage patterns of the network, examines the nature of interactions among children on the network, and discusses the evolving sense of community among the participants.


In the final chapter of the section, "They Have Their Own Thoughts," Paula Hooper describes her research at an African-centered community school. The school, Paige Academy, aims to create a community and culture rooted in African-American ways of knowing, values, and interactions. Hooper presents a "learning story" of an 8-year-old girl working on a Logo project. The story examines how children, in the process of making new computational artifacts, can take control of their own learning, developing and extending their own ideas. It also highlights the need for school environments to support and legitimize children's own personal ways of thinking and mental and physical constructs.

Learning about Systems

In discussions about computers in education, the focus is usually on how children learn. And, indeed, some uses of computers can bring about profound changes in how children learn. But often overlooked are the parallel opportunities for transforming what children learn. Current school curricula were created in the era of paper and pencil. The current mathematics curriculum, for example, is not necessarily the best or most appropriate mathematics for children to learn; rather, it is the mathematics that is most accessible with paper and pencil. As new tools and new media become available, we must rethink the content of mathematics and other subject domains.
One area ripe for rethinking is the study of systems. The world is full of systems (biological, technological, and social) consisting of many interacting parts. Such systems are very difficult to understand, and often impossible to analyze with traditional pre-college mathematics. But new computational tools can open up new approaches for studying and understanding the behaviors of systems. The chapters in this section discuss the barriers to understanding systems concepts (such as feedback and self-organization), and they describe computational tools and activities designed to support new ways of thinking about such concepts.


In the first chapter, "New Paradigms for Computing, New Paradigms for Thinking," Mitchel Resnick argues that new computational paradigms (such as object-oriented programming and parallelism) can significantly influence not only what people do with computers, but how they make sense of the world. In particular, he describes a parallel-programming language, called StarLogo, that he designed to help students explore decentralized systems, such as bird flocks, traffic jams, and market economies. By building models with StarLogo, students can move beyond the "centralized mindset" and develop a deeper understanding of decentralized systems.


Uri Wilensky uses StarLogo towards a different goal: to help learners develop their intuitive conceptions of probabilistic ideas. In Wilensky's chapter, "Learning Probability through Paradox and Programming," he presents a case study of a learner who uses programming to help resolve a probability paradox, and in the process develops stronger intuitions about randomness and distribution-and the connections between them. Wilensky's study illustrates that the primary obstacles to learning probability are conceptual and epistemological, and it shows how programming can play a powerful role in learning mathematics by making hidden assumptions explicit and concrete.


In the final chapter, "Ideal and Real Systems," Fred Martin probes students' thinking about systems in the context of a robot-design competition that he helped designed for MIT undergraduates. Martin analyzes how and why undergraduates have trouble developing effective strategies for controlling their robots. He shows that students tend to build robots that will perform properly only under ideal conditions, not in the "messiness" of the real world. Martin calls for a change in undergraduate engineering education, arguing that the standard engineering curriculum encourages design strategies that are not appropriate for many real-world technological systems.


This book would not have been possible without the help of many people. All of the authors made major contributions of time, thought, and energy. Our current work benefits from the past and continuing contributions of the extended Logo community, including many former members of the Epistemology and Learning Group at the MIT Media Lab. We are grateful to Wanda Gleason and Florence Williams for helping with many organizational and editing tasks, and to Jacqueline Karaaslanian and Mai Cleary for providing general administrative support. The National Science Foundation, the LEGO Group, IBM, Nintendo, and the News in the Future Consortium have been generous in their support of this research. Finally, we would like to thank Seymour Papert for inspiring all of us to think in new ways about learning, children, minds, and computers.



Yasmin Kafai
Mitchel Resnick

May 1995



Return to Main Book Page