Serendipitous Design for Play

I so clearly remember the first play sculpture that I did in a school. I had a very clear idea about how the kids would use each feature of the piece. As is generally the case, there was a ceremonial “opening” in which all the students were released all at once to mob the structure. Within five minutes, they did everything I had imagined and then went on to invent so many more that I lost count.

One of the play events I introduced on that day, and nearly all subsequent designs was the banister slide. Adults would often ask me, “How are kids supposed to use that?” My response was, “Exactly the point, it’s up to them to figure it out.” Indeed, when I would join a class for P.E. I would have a group of kids join me at the banister slide and give them the following instructions, “You can do anything you want to do, except something that has already been done.” We typically spent the whole period on that one event and never ran out of innovation.

Throughout my five decades of play design, I have followed this simple rule, the less I can anticipate how kids will use the designs, the better they are for play. Of course, there is generally some push back from adults who have definite ideas about what playgrounds and toys should do and look like, and I’ve had to learn how to design so that those expectations don’t create too much rulemaking on the part of adults who expect their children to behave “appropriately.”  The balance is to appear to be normal and safe to adults while including as much oddball open-ended stuff as possible. I’ve learned to add serendipity to my designs.

This image of the little girl hanging by her hands is an example. The arch at the top rail was added to provide an entrance-like detail and head clearance for older kids climbing onto the fourth and highest platform. I anticipated that it would become a spot for children to hang by their hands and was delighted that not only does this occur regularly, but parents are generally permissive of this very beneficial activity.

hang by hands

This image of two climbers side by side illustrates the idea of serendipity perfectly. The climber on the left uses shapes that mimic rock climber holds and is part of my original design 20 years ago. We replicated them for the new system to honor the success of that design and maintain our traditional appeal. The Peg Climber in the center is something else again. I had no idea how exactly this design would be used. It turns out that each age approaches climbing it differently. Toddlers simply ignore the pegs all together and just swarm up. A year later, and they go up trying to avoid using the pegs to assist them. By three they will try to go up using just the pegs. The most advanced children will go up, stepping only on the pegs.


The Play and Music system has a huge advantage when it comes to providing serendipitous play. All of the elements can be moved and the experience reconfigured. That means what was once a climber can become a bridge. What was used as a slide can become a climber. Perhaps the best illustration of this is the Net Climber that can be removed for the frames and set on the floor where it can be a rocker for five babies or a spinner for two preschoolers.


As we noted in another blog a few days ago, toys and play equipment currently available for children are predominately one dimensional and support very specific and limited play opportunities for creative and unexpected play. We can do much better than this as designers and as consumers. For example, parents and teachers can help by selecting playthings that have some ambiguity about them. Look for loose part systems that are complex. Allow kids to mix systems. And if at all possible, find ways to introduce magic.


STEM and Serendipity

STEM Part 2

Any sufficiently advanced technology is indistinguishable from magic.  – Arthur C Clarke

A surprising amount of modern technology is the result of happy accidents. Penicillin, X-ray imaging, microwave ovens, and many more were not the goal of the lab worker but came about because the researcher was highly observant and open-minded.

While STEM knowledge is valuable, the capacity to recognize an unusual pattern and explore it is incalculably precious. Even in the science and tech communities, people vary in their encounters with serendipity. Apparently, this is a learned skill. The question is when and how does this skill arise?

We recognize happy accidents in adults when they result in significant breakthroughs. These tend to happen when the research is in highly complex environments such as pharmaceuticals and quantum mechanics. Thus, we can speculate that an information-rich setting is fertile ground. Second, interviews of “super-encounterers” who find happy surprises everywhere tend to seek out novel information just for the joy of it, the odder, the better. What others see as a waste of time they see as data gathering.

I suggest that most young children are also super-encounterers by nature. They come into this very complicated world where everything is new, and they are driven to discover all they can about it as quickly as possible.

Recent findings in neuroscience tell us that at birth, children already “know” a tremendous amount. They come pre-loaded, if you will, with many behavioral templates that they deploy and refine through direct experience with their environment. For example, it is well known that neonates are wired for language and over the first year their babbling increasingly sounds like the dialect of their parents. Parents are also programmed for this interaction and use “baby talk” and eye gaze to enhance this learning.

We have identified 20 of these programs that we have labeled Play Patterns. These templates reside primarily in the cerebellum and as they are acted upon messages are sent to the right cerebral cortex where they increasingly become under voluntary control.

Most of the current neuroscience suggests that for the most part, the left cortex, which tends to be the seat of numbers and letters, is largely dormant until the second year. I suspect that this is not entirely true. Children exhibit a unique behavior that is not accounted for in our current model of brain development, that is, serendipity. This behavior can be observed when children encounter phenomena for which they have no predefined play pattern. The telltale signs of serendipity are best characterized by rapture and wonder, in other words, “magic.”


At Gymboree Play and Music, we conjure up this magic at the end of every class. Our parachute and bubbles are now ubiquitous and for a good reason, as they evoke the wonder of childhood. Put yourself in the mind of the child. She sees the teacher dip a wire circle into what appears to be water. Pulling the loop out of the bottle, the teacher blows on the loop from which emerges a magic bubble that detaches and floats free into the air. When she explores this with her index finger, which is the child’s discovery tool, the bubble suddenly disappears entirely. Pure magic. This experience is then followed by parachute time during which a colorful “cloud” is made to rise up into the heavens under which all the children can gather. Suddenly it too magically disappears.

Whether this sort of non-pattern matching phenomena becomes, as I suspect, the early awakening of the left cerebral cortex or not, serendipitous experiences are one of the great joys of life for parents and children alike. I also contend that having such experiences helps children seek out other wondrous experiences and welcome the unique event that does not fit into a preprogrammed pattern.

My concern is that with the increasing emphasis on STEM, we may ignore this higher level of awareness. We may be educating perfectly functional people who function adequately in a technological world but at the same time conditioning them in a way that makes them blind to the magic in this world and the next reality transforming breakthrough.

For further reading:

Too Young to STEAM?

STEAM Part 1

Back in 1974, Buckminster Fuller wrote in the introduction to my first book, Build Your Own Playground, the following:

I think that playgrounds should be renamed “research environments.” This is what the children are doing so vigorously. They are not playing. They are finding out how the universe works. This is spontaneous research which is inherently gratifying, often joyously gratifying. How wonderful to find out how to use gravity as an accelerator or a brake. Nobody is around to tell you or to give you the name gravity, but you learn quickly that the greater the drop, the more it hurts your legs. That is what Galileo’s work with falling bodies was all about. You want to understand that invisible power that is working around for you; you wish to check out your theory on a slide.

Children learn about tension. They have got to tear a great many things apart to find something that won’t tear, that they can spontaneously grab for to arrest the falling and anticipate leg shock or break. They don’t have to know the names tension, compression, gravity, or acceleration, but they have to get very familiar with such phenomena before a sound emanating from somebody’s mouth can develop a word meaning experience. City-born and -matured children have almost no access to operative research environments as have had the billions of humans in the millions of years of their occupancy on planet Earth’s pre-city eons. 

Playgrounds provide children with experience-fortified gratification of physical research. Thus, their intuitive assumptions of “can do” are proven; they are thereafter confident of their own capabilities for sensing and employing the principles operative in nature, such as gravity, flotation, wind resistance, tension, and compression. Teen-agers and adults then may successfully deploy into wilderness for such activities as mountain skiing, surfboarding, cross-country motorcycling, and flying kites. 

Of course, we all know that Bucky was WWWAAAY ahead of his time. What I especially love about Bucky’s statement was that it presented the vision of full-body exploration. Far too many STEM products are desktop and don’t account for the fact that little kid CAN’T sit still. They learn faster and better when they are totally immersed in playful learning. Many of the toys presenting themselves being STEM have a single “right” solution which, for me, is the antithesis of scientific inquiry. This is especially true when looking at toys for children seven years-of-age and younger.

Screen Shot 2019-05-27 at 10.39.40 AM

I just ran across an excellent white paper by the Toy Association, STEM/STEAM – Formula for Success. The paper does a good, if very superficial, job of identifying right and left-brain functions and clearly supports the proposition that a good STEM product needs to be both fun and intellectually challenging. What they don’t point out is that, for young children, the left cortex is in a very primitive state and is not ready for intellectual growth as such. I liked nearly everything they have to say, especially this point:

“OPEN-ENDED. This refers to a toy that encourages the child to find his or her own individual way to play. After the mandatory characteristics, this was the most predominant attribute mentioned for a good STEM/STEAM toy – the ability for a product to be used in multiple manners where there is no one right way to play. This includes toys that offer various pathways to solving a problem, building a structure, creating a design, or accomplishing a task.

I will recommend this paper to you if the object of your interest in STEM is for kids eight years and older. However, if you have, or are teaching, younger children stay tuned to this blog as we will be exploring what the neuroscience has to say about how the developing brain is preparing the young child to take on the rigors of STEM.

One Dimensional Play


The folks who have been creating children’s educational toys have been shortchanging us for decades. Kids will play with just about anything. That kids play with the materials in most early childhood programs is no evidence that these materials are as beneficial as possible. A great example of this is the ever-popular hollow blocks. These are excellent examples of loose-part play toys that have been around for decades and are in nearly every early childhood education center.

There is no question that kids will play with them, often for extended periods. But ask yourself, after the first few play sessions what more can they learn besides how the shapes fit together? I assert that block play devolves to social play in short order, not that this is a bad thing, but could we do something more?

Our recently departed and much-loved play guru, Bernie DeKoven, used the term “complexification” for the idea of maximizing playful learning by making the environment multi-dimensional … more stuff means more complex play and more learning.

floor blocks

Visit nearly every preschool, and you will find the environment laid out into functional spaces; the kitchen area, the reading nook, and the block play space. The materials in each of these areas are siloed and rarely mixed. Teachers are taught to program these spaces. School supply catalog merchandise their products along these same categories. And so it has been for decades. And let’s not even talk about the lack of travel between indoors and outdoors. This “traditional” ECE format seems set in stone. The result of all this stagnation is that designers of educational materials find themselves trapped into these narrow and shop-worn classifications.

What will happen when we begin to look at the physical plant of an ECE program as a system of interoperating elements? I’m not suggesting a rigid requirement that everything has to fit with everything else, but that the apparatus and space are thought of, and planned for, dynamic and fluid mobility. How are the boundaries between functions presented to allow for flow? How is the presentation of materials transformed? What changes need to happen to storage?

The good news is that there are many examples of precisely this thinking that creative teachers have pioneered. Puddle Jumpers Nature Preschool is an excellent example. The way that Teacher Tom organizes space is also worth checking out. The rapid rise of AnjiPlay and Regio Emilia centers shows that this approach is rapidly becoming a trend.


We need to think about how much more we can get out of our school equipment if we allow for the mixing of functions. As Bucky Fuller taught, one plus one doesn’t equal two. These synergistic combinations can be transformative.

For many teachers and administrators, taking this new path will take commitment and patience. The good news is that there are great exemplary programs and organizations like NAEYC that are there to help make the transition.


The Ideal Learning Environment

BH yard
An OK place to play but not an ideal learning environment

Most early childhood education centers do an adequate job of providing an outdoor play space. That said, these environments are not ideal learning environments. This is somewhat strange because teacher generally learning during their education process the basic principle of how children learn but then don’t fully put this knowledge into practice when developing their outdoor play space. If they did, what would that look like?

Until the ‘70s was the consensus of childhood researchers like Piaget was that children’s brains were “tabula rasa”, a blank slate. Ten years ago, Alison Gopnik and her colleagues Andrew Meltzoff and Patricia Kuhl published The Scientist in the CribMinds, Brains, and How Children Learn. The main thesis of the book is that children are born with very powerful brains and do a lot of thinking. They are like scientists who are constantly creating predictions about their world and how it works and refining those predictions based on experience. Her analogy was that they are like a computer with tremendous computational power and loaded with sophisticated programs but until a person sits down and enters information, they are not functional. Another way of saying this is that kids are born with complex templates and these are adapted and filled in as the child gains experience. For example, from the moment of birth children are listening for words and they learn the specific language that they are born into, they have a template for language which they fill with the local dialect.

Another point that is brought out in the book, which I don’t think has gotten enough attention, is that for the most part learning is promoted by two key components, action, and social engagement. Babies give rapped attention first to the parents, and as they mature, to other people they encounter. They are also moving almost constantly when they are awake. It is through motion in a social context that the child’s intrinsic templates get adapted to their environment, i.e., this active play is the optimal condition for learning.

In the decade following the publication of The Scientist in the Crib researchers have been able to actually peer inside children’s brains and can now verify that the book’s contentions are correct. They can see the parts of the brain that light up in response to specific stimuli. They have shown that for the first two years children are primarily learning how to operate their bodies. The term we often hear used for this process is sensory integration. The main player in this process is the cerebellum. The interesting finding has been that the cerebellum had been thought to be essentially a movement computer like the self-driving computer in a Tesla. It turns out that the cerebellum is constantly creating a model of the whole world of the child and anticipating what will happen next. It then adjusts this model based on the accuracy of those predictions. To do this it talks to the right cerebral cortex to assess how best to make adaptations, i.e. the right cortex is the diver in this analogy. So, far from just learning how to move, for the first seven years, the cerebellum and its partner the right cortex are the main areas of learning about everything in the child’s world including emotions.

Let’s make a list of what the current research has established the ideal learning environment for children from 2 to 7 years of age:

  • There are other players in the setting, preferably with a mix of ages
  • The space allows for lots of movement, especially large gross motor activities
  • Children in the space are able to experiment, test limits and to fail often
  • Children will have essentially unlimited ability to change the elements within the space
  • The elements in the space have more than one function, preferably they can be used in many ways
  • The optimum learning space will be primarily outdoors
  • The space promotes immersive and emergent learning that is indicated by very long play episodes

While still rare, there are schools that embody all seven of these criteria. For example, AnjiPlay schools in China, the increasingly popular “Forest” schools, and many Reggio Emilia schools.


anji yard2
The many AnjiPlay sites have ideal playspaces

It is fair to say that than most schools in the USA fail at providing the ideal learning environment. There are many reasons for this, the push for academics, the need to provide a “safe” environment, the lack of teacher training for operating in such a learning space, and parent expectations of what a “proper” school should look like and teach.

The fact is that for the majority of programs being able to have an ideal learning environment is hampered by the lack of well-designed equipment. Indeed Cheng Xueqin, the Director of the Office of Pre-Primary Education had to invent from scratch the apparatus they use in her program. Most other schools that meet these criteria have access to naturalistic spaces and hand-make whatever else they feel they need to support the children’s learning.

It is no wonder that few schools can implement an ideal learning environment. For example, one need only look at what outdoor equipment is currently available for early childhood educators to see that large motor apparatus is invariably fixed in place, has a single function and cannot be changed by the children.

In my next blog, I will explore ideas that can offer new options for creating the ideal learning environment.

In the meantime check out this great article by our friend Peter Grey – Children Educate Themselves

Dogs, Neuroscience and STEM Education

In my lifetime I have been the human for six wonderful dogs. I was just six years old when I got my first one. I wanted my pet to be the best, so I enrolled in an obedience class for Coalie, named for his coat color, as well as several of my subsequent companions. One of the most important things I learned in those classes was that the better-trained dogs required the fewest words. Indeed, if you attend sheepherding or agility trials, you will rarely hear a command spoken, and yet such animals display a large repertoire of learned skills. These days when I see someone verbally instructing their pet, I laugh, usually not out loud, because I know that dogs respond to gestures and body language and not so much to words.

As a play advocate, I’ve recently become aware of the breakthroughs happening in the neuroscience and developmental evolution. I’ve studied how intelligence progresses from fish to primates and have learned how the smarts of my dear Coalie are not that far off from humans. Indeed, for the first few years, kids and dogs are relatively closely matched. That means that their learning is primarily through gesture, body language, and movement.

The use of fMRI has given us the ability to see living brains in action and allowed a much more actuate view of learning. For example, it was a cannon of psychology that the role of the cerebellum was the center of motor control. While that is still mostly true, the cerebellum is far more complex and important. Take this fact for example. The cerebellum contains 69 billion neurons while the cerebral cortex, the area of the brain that we tend to think of as where all of our smarts resides, only contains 16 billion neurons.

It is also interesting to note that initially, the cerebellum communicates primarily with only the right half of the cerebral cortex. That’s the side that deals mostly with imagination, empathy, and intuition. The left half deals with facts, numbers, and letters are ignored during the early years. What do these findings tell us about Science, Technology, Engineering, and Math education for young children?

The genius of Gary Larson captured this idea perfectly in this Far Side cartoon. Notice that we laugh at the truth of this when the subject is dogs, but the situation would be much the same if we were talking about children. In a very real way, trying to have kids learn STEM ideas verbally is a fool’s errand. Knowing this the developers of most STEM education focus on hands-on projects — all well and good. But wait, what does neuroscience say about the efficacy of that approach?

First, hands-on is good, but body-on is many times better. Early childhood learning progress best during full body engagement, i.e., play. For it is during play that the feel-good chemicals like dopamine and endorphin flood the brain and significantly increase the rate of neuron myelination which marks the structural changes in the brain that results in learning.

Second, both dogs and kids already know many of the basic principles of STEM. There are interesting studies that show that babies act surprised when they see something that violates fundamental physics. Or take the fact that if you load one glass with 5 M&M’s and another with eight, kids will invariably select the glass with the most candy showing that they understand the notion of quantity. So, what does this tell us about “teaching” STEM to young children?

To start with they are smarter and know more than we assume. Kids also “understand” intuitively and not intellectually. Maria Montessori understood this, and it is the basis of her educational system. Unfortunately, the teaching methods that embody her insights have become viewed by many as sacrosanct and held to dogmatically rather than being a wellspring of creativity.

The other issue with Montessori and much of STEM education is that there is a single known outcome to the materials presented to children. Whereas, what is far more critical is fostering curiosity, creativity, and experimentation. Kids are very quick to figure out that adults have provided a lesson to be learned and that real play is not on the agenda. Soon kids just look for the embedded lesson rather than being free to explore.

What dogs, kids and the new findings in science teach us is that learning is best when it is full-body, active, fun, and open-ended. Children at very young ages can learn the underlying STEM information best when it is presented in a form that integrates well with those areas of the brain that are in the process of development.

Here’s a taste of the science:

Why Young Kids Learn Through Movement

The Association Between Childhood Motor and Cognitive Development

From Movement to Thought: The Development of Executive Function

Optimizing Early Brain and Motor Development Through Movement