The four-month concept phase focused on developing detailed media ideas around specific learning zones, in tandem with the exhibition's overall look and feel. It culminated in a grant-required (and infinitely helpful) user testing session, where we piloted three of our concepts.

At the outset, we conducted a goal-setting workshop, looking at sketches and precedents to build a collective vocabulary around "what should this space feel like?" We then dove into research with our content partners, and worked to create media concepts that felt best suited to each story.

For "Be a River Character" in particular, we had developed a mural-size graphic that introduced the cast of the Delaware (which is in the final exhibit). We wanted to give kids a chance to dive one step further, and proposed the idea of using a depth sensor to transform them into one of these players. We wanted users to build empathy for the diverse creatures of the watershed—by becoming one!

As we developed the concept, we started to ask, "what's the best way to user test this?" From a technology standpoint, we knew roughly how to make it: using depth camera-based skeletal tracking. However, from an experience standpoint, we weren’t sure it would provide the right educational balance. It felt at first that in order to prototype this concept, we would have to more or less build the real thing (i.e., not a prototype, really!). The possibility of a dysfunctional prototype would mean users fighting the technology, rather than thinking about the experience; on the other hand, making it work really well would mean us making a slew of design decisions outside of visitor feedback.

Enter: the Marx Brothers. Inspired by a scene in Duck Soup, we decided to use only people and handmade costumes. Similar to the proposed experience, the characters would "appear" (walk out from behind a curtain) when kids walked up; from there, they would mimic every move. After some initial awkwardness, children and adults alike were able to interact with our “digital characters.”

The prototype was weirdly successful (more on that below), and led to a big pivot. If you visit the exhibition today, rather than a projected magic mirror, you'll see (you guessed it)... costumes!

This prototype, in concert with the cohort of lo-fi sketches that rounded out our testing session, encouraged us to get even scrappier with our user-testing efforts. We found that kids and adults are able to suspend their disbelief if the experience is compelling enough. (Admittedly, the adults needed a little more coaxing.)

There were a lot of learning moments during prototyping that challenged our assumptions: fact-based speech bubbles were ignored, and the audience quickly grew bored with simple one-to-one mirroring. Since no technology was involved, we could easily improvise. After a few minutes of mirroring the audience, our fiddler crab and eelgrass began to randomly wink, make funny gestures, and dance. This made participants pause and laugh, and created an opportunity to sneak in a learning moment with those same printed speech bubbles.

The rest of the session was more traditional and structured, which also really helped. After interacting with a paper touchscreen or our costume "mirror," participants were interviewed in a group setting—allowing them to think deeply about what might have at first seemed like a silly experience. Thoughtful prompts (many as simple as "what did you learn") as well as a group to riff ideas off of encouraged participants to give meaningful and, for us, actionable feedback.

River Alive! has gone on to win an AAM Excellence in Exhibition Award, for Special Achievement in Interactivity, as well as awards from Communication Arts Magazine. It is also a regular feature in "must-visit" lists around Philadelphia.

The recipe for success for this prototyping session was:

• 1 part planning: the careful orchestration of manageable groups of kids (out of our total of 100) was key.
• 1 part trust: our clients and our team got on board early with us going loose.
• 1 part mechanics: having a seasoned evaluator with a specific place in our flow was crucial during and after (we got good, actionable feedback!).
• 1 part scrappiness: our team made stuff out of paper, internet-sourced spandex, and cheap craft supplies.
• 100 parts dedication to the role: our crab, Lilly, and our eelgrass, Joanie, both went all-in, not only playing the part but also evaluating and adjusting as they went. Not just anyone can become a river creature, and while it seemed crazy at first to have our design team dress up in costumes, it was in fact invaluable.

Users will interact with the loosest of prototypes, and often the lowest-tech ideas often result in the most transformative takeaways. Always let story drive concept, and find ways to allow real-world tests and prototypes to hone and refine your experience—and try to put those ideas in front of your audience earlier than you feel comfortable. It'll be worth it!

River Alive! opened in 2019, and continues to be a successful core draw at the Independence Seaport Museum. Moving forward, we've continued to challenge ourselves to mockup more loosely, and earlier, in the process. We've yet to find another opportunity to don red and green spandex, but, rest assured, we will.