Ph.D. Thesis

mooseSculpting Behavior
A tangible language for hands-on play and learning

Public defense of Ph.D. thesis

December 11, 2007 10 a.m. MIT Media Lab, Weisner room, 2nd floor (map)
Watch a webcast of the oral presentation



For over a century, educators and constructivist theorists have argued that children learn by actively forming and testing – constructing – theories about how the world works. Recent efforts in the design of “tangible user interfaces” (TUIs) for learning have sought to bring together interaction models like direct manipulation and pedagogical frameworks like constructivism to make new, often complex, ideas salient for young children. Tangible interfaces attempt to eliminate the distance between the computational and physical world by making behavior directly manipulable with one’s hands. In the past, systems for children to model behavior have been either intuitive-but-simple, e.g. curlybot or complex-but-abstract, e.g. LEGO Mindstorms. In order to develop a system that supports a user’s transition from intuitive-but-simple constructions to constructions that are complex-but-abstract, I draw upon constructivist educational theories, particularly Bruner’s theories of how learning progresses through enactive then iconic and then symbolic representations.

I present the Topobo system, a class of tools that helps people transition from simple-but-intuitive exploration to abstract-and-flexible exploration. The system is designed to facilitate mental transitions between different representations of ideas, and between different tools. A modular design approach, with an inherent grammar, helps people make such transitions. With Topobo, children use enactive knowledge, e.g. knowing how to walk, as the intellectual basis to understand a scientific domain, e.g. engineering and robot locomotion. Queens, backpacks, Remix and Robo add various abstractions to the system, and extend the tangible interface. Children use Topobo to transition from hands-on knowledge to theories that can be tested and reformulated, employing a combination of enactive, iconic and symbolic representations of ideas.


Download Sculpting Behavior (high res) or (low res)


Thesis Committee

Thesis Advisor
Hiroshi Ishii
Associate Professor of Media Arts and Sciences
Massachusetts Institute of Technology

Thesis Reader
Mitchel Resnick
LEGO Papert Professor of Learning Research
Massachusetts Institute of Technology

Thesis Reader
John Maeda
Associate Director of Research
Massachusetts Institute of Technology


Hayes Raffle | Biography

Hayes Raffle is a Ph.D. candidate in the Tangible Media Group a the MIT Media Lab. He is a practicing artist and designer researching how interactive toys, systems and new devices can support hands-on communication, education, and creative expression.

Before attending the Media Lab, Hayes received a B.A. in fine arts (sculpture) at Yale, helped design and develop the award-winning
ZOOB® building system and ran his own art and design studio in California. He is the winner of several internationally recognized design awards and has shown his art in Europe and the United States.

hayes (at)


Teaching at MIT – Toy Studio


I teach an Interactive Toy Design Studio to MIT grad and undergrad students.

Have a look at images of students’ work, at the course website.

This studio introduces students to fundamentals of interactive toy design and introduce basic design techniques and principles. Students build several toy prototypes in this class. We review related theory from the fields of design, fine art, education and cognitive science, and focus on fast ideation and implementation of ideas. Some readings related to design and learning are required, but the majority of the course is spent doing hands-on building. The course compresses a full semester into three weeks of intensive, daily design studios.

Topics Include

  • toys in modern art, and toys invented by modern artists. Alice Aycock, Alexander Calder, Dennis Oppenheim, Picasso, Paul Klee, Jean Tinguely, James Seawright, Michael Grey, Allan Kapprow, Cindy Sherman, Claes Oldenberg, Frank Gehery.
  • toys as educational tools: manipulatives. overview of constructivist theories of learning, piaget. froebel, montessori, building toys, digital manipulatives.
  • kinetic art, mechanical toys and automata. overview of mechanism design.
  • fantasy, storytelling, character and story. dolls, action figures, stage, performance.
  • toys made from trash, toys made from reappropriated materials, and toys invented by kids.
  • subversive toys. and dystopic visions of play and invention. movies: brazil, toy story, blade runner, etc.
  • toy manufacturing
  • large scale toys, jungle gyms, architecture for play and body-scale learning toys (a la exploratorium)
  • electronic toys, computers for play
  • toy marketing: what sells and why?
  • toy design process from concept -> the toy store shelf

    MIT Application Essay

    Sophisticated decision-making systems have evolved within our bodies and in other natural systems around us. DNA, protein interactions and chemical modeling are all examples of natural “spatial languages” that use physical form to communicate and make decisions. I believe it is possible to create a problem solving system that captures the intelligence of these natural decision-making schemas, and MIT’s Tangible Media Group is the ideal environment in which to create such a system. It would be an incredibly powerful and unique approach to problem solving because its analogs are evolutionarily proven in the natural world.

    The summer between my junior and senior years at Yale, I worked with conceptual artist Michael Joaquin Grey as his artist intern. Michael sought to bring dynamic modeling, which was only possible using computers, into physical space with a hands-on tool called Zoob®. ZOOB is an acronym for Zoology, Ontology, Ontogeny and Botany, and was an idea to create a haptic interface that had the complexity and dynamics of information behavior or living system behavior. I loved the idea and I approached the problem determined to find a solution. Over the next two months, I helped develop the abstract idea of Zoob into specific conceptual models with functional engineering solutions. A revolutionary system emerged that embodies dynamic relationships found in micro and macro systems such as DNA, bones, and the cosmos, and makes their complex interactions accessible and fun.

    After I finished my undergraduate studies, I moved to San Francisco and helped form Primordial, LLC, a manufacturing and design company that produced Zoob for the toy market. My duties at Primordial were vast; I began by inventing a way to communicate the system’s behavior on the printed page, and over the next three years I built and managed the company’s creative department and oversaw manufacturing quality control. In 1999, I left Primordial to return to my art practice. Zoob taught me about system behavior, and I have sought to understand the natures of other systems through making art.

    How is a whole greater than the sum of its parts? In trying to communicate complex, overlapping ideas, categorization is often necessary to break information into bite-sized pieces. However, this act of categorization destroys the relationships that may be the very essence of what makes the whole greater than its parts. In recent years this has become my intellectual focus, and I have sought answers to this question through my art practice. My creative inspiration comes from making objects, playing with them, and learning how they behave. In November 2000, my visit to the Tangible Media Group inspired me to learn to integrate electronics with physical media. I began comparing electronic and living systems, and I created several “Electronic Organisms” that evoke organic behavior using simple, recursive electronic circuits. The following July, while creating interactive science exhibits at the San Francisco Exploratorium, I severely injured my left hand in an industrial accident. This injury left me disabled. During my recovery I have been studying electromechanical control systems, which led to my creation of the “Solargarten” of self-sufficient electronic organisms. My healing process has given me a unique opportunity to contrast synthetic and natural systems while I watch my body grow with the aid of prosthetics. It has also led me back to the Media Lab.

    I returned to the Tangible Media Group in November 2001 and was very excited to meet Professor Hiroshi Ishii and learn more about his group’s focus; Ph.D. candidate Brygg Ullmer introduced me to his work creating Tangible User Interfaces for Manipulating Abstract Digital Information. In speaking with Mr. Ishii and Mr. Ullmer, I realized that it was possible for Zoob to model information. This is very exciting to me, because in its origin Zoob was an attempt to create a system that would reflexively bring people closer to understanding both real and information behavior. If Zoob successfully modeled information, it could harness decision-making strategies inherent in natural spatial languages such as DNA and proteins! Such a “Zoob-User-Interface,” or “ZUI,” could create an entirely new way of understanding complex problem solving.

    The potential for organic information-modeling systems is great, and TMG provides the ideal environment to develop such a system. TMG tackles tough problems with rigorous enthusiasm. The program has a strong history solving many of the important problems that a ZUI presents; this includes the definition of “meaning” in the system, hardware and software solutions to interface a modular, tangible system with computers, and the identification and application of specific complex problems. Furthermore, TMG’s work evinces clean design solutions that are integral to successful tools and interfaces.

    Further work is necessary to establish a clear direction for a ZUI, and I have the right sensibility for the project. A strength of mine, perhaps because of my experience as an artist, is my ability to synthesize different media to communicate complicated ideas. My work in business as a project manager and art director trained me to direct creative and technical teams to deliver results on time. I am confident in my ability to lead projects that demand precise design solutions within required time frames, while contributing an artist’s sensitivity to the aesthetics, meaning, and social ramifications of those solutions.

    A primary professional goal of mine is to redefine cultural awareness through meaningful integrations of design and technology. The successful implementation of a ZUI could redefine how people form ideas and solve problems. In 1848, the English high school math teacher George Boole developed a binary algebra and asserted that numbers can represent ideas. This spawned modern philosophical logic and the binary processes used by every computer today. If the Zoob system intelligently represents ideas, it too could be a powerful philosophical and computational tool. ZUI could move us from Boole’s binary, mathematical problem-solving paradigm to a tangible, organic one. On a large scale, society reflects the technology it creates, and we are seeing computers become so prevalent that people are starting to mimic them. ZUI could shift the computing medium to more closely match natural structures—structures that we have both evolved to use, and that we have used to evolve.

    ©2001 Hayes Raffle. ZOOB® is a registered trademark of Z Co.