Science Behind the Fiction: How science is bringing Digimon to life

The '90s were a wild time. The internet had arrived but hadn’t quite come into its own just yet. Personal computers were creeping into homes, but it wasn’t unusual not to have one. People who did own a computer had to log on to the web with their phone lines, carried into the great digital sea by the siren song of the dial-up screech. Some of us still hear it in our nightmares.

It was a time of transition, of change between the world as it was and the world of today, a world wherein vast swaths of our lives take place in a digital space.

This dichotomy is, perhaps, best represented by the toys that captured the imaginations of the children of the day. Video games exploded in popularity, offering kids a portal to fantastical digital worlds, but we also made efforts to upload the more mundane aspects of our real lives into pixels, ones, and zeroes. Enter the digital pet and its many iterations.

Tamagotchis came first. They were docile pets that did little more than eat, sleep, defecate, and die (oh, God, they died so much). They were a full-blown phenomenon, and seeing their success, Bandai went to work expanding the market, specifically targeting boys. Inspired by the aesthetic of American comic books, designer Kenji Watanabe set out to design a line of digital pets that would appeal to male children. Thus, Digimon was born.

Digimon were aggressive, they did battle, they were edgy. At least as edgy as a black and white smattering of pixels could be. The image, and the brand, were sold mostly on the strength of associated video games and television series, which introduced a complex story involving a battle for the sanctity of a digital world. These toys took something familiar, the bond between a child and their pet, and transformed it into something wholly different. Digimon, both the series and the hand-held toys that inspired it, were the perfect marriage of old and new, the perfect crystallization of the Pog-Digipet Boundary.

While the golden age of virtual pets is long since over — despite Tamagotchi enjoying a renaissance among nostalgia-hungry older millennials — it might only have been a precursor for what’s to come.

BRINGING VIRTUAL PETS TO LIFE

The first step in bringing Digimon into reality would be to create a working digital brain.

The brain — how it works, and how all the neurons and connections in that few pounds of gelatinous material result in consciousness — is a matter of fierce inquiry and debate in the scientific community. Of all the varied and wonderful things we have come to understand since the birth of the scientific method, it seems that the thing that's still hardest to grapple with is ourselves. But that’s not for lack of trying.

Neuroscience is a field dedicated to the unraveling of that mystery, and one strategy is to map the connectome, creating a map of every connection in the brain. That process is slow-going, painstaking work, but the journey of a hundred billion neurons must start with a single step.

Others are starting smaller, with animal brains, just to prove it can be done, and to expand our collective knowledge about how brains, even very tiny ones, work. To date, only two complete animal connectomes have been mapped: those for the nematode (Caenorhabditis elegans) and the sea squirt (Ciona intestinalis).

The nematode in question was a reasonable choice, owing to its relatively simple nervous system. With only 302 neurons, the connectome of C. elegans is orders of magnitude simpler than ours; it's the equivalent of jogging down your driveway as a first training exercise in running a triathlon. The ultimate goal is much more difficult and complex, but this at least proves you can move in the right ways.

Mapping the mind of C. elegans was only the first step, however, in understanding its behavior. Having that map was like having a diagram of all the parts that make up a car. It’s a lot of really good information, especially if your goal is to build a car from scratch, but it won’t get you very far if you don’t know what each of the parts does and how they interact with one another.

Scientists have spent decades trying to fill in those gaps in our knowledge, studying the nematode’s behavior to determine which bits of the connectome are responsible for what.

The method by which that knowledge was attained gets a little messy. It was achieved by destroying certain connections to see what happened, thereby figuring which parts of the connectome were responsible for specific behaviors. To return to the car analogy, it’s a little like figuring out what headlights do by noticing it got dark after you smashed them with a hammer. Sometimes, in order to find out how a thing works, you have to break it.

One particular experiment was carried out in 1985 by Martin Chalfie. He figured out which parts of the nematode’s connectome were responsible for moving in response to external stimuli by destroying neural connections with a laser and noting when the behavior stopped.

Experiments like this have been carried out over the past few decades, since the completion of the C. elegans connectome in the 1970s. This combination of a neural map along with an understanding of what behaviors these pathways represent has resulted in a more complete understanding of how nematodes' minds work.

And once you know precisely how to construct an artificial mind, there’s really only one thing to do: Build an android.

In 2014, that’s exactly what we did.

The OpenWorm project is an international effort to take the information gained about C. elegans and transform it into digital space. They’ve created software modeling the connectome and related behaviors in order to simulate said behavior. In 2014 they put that simulation into a simple robot made of Lego parts. The result was a robot that moved of its own accord, reacting to external stimuli with no programming to determine its behavior other than that which came from the nematode’s simulated mind.

If you’re feeling existentially terrified, you’re not alone. But never fear: We’re a long way off from putting you inside a Legobot.

More recently, in April of this year, scientists at the Allen Institute for Brain Science in Seattle celebrated a major milestone in mapping the connectome of a mouse.

The team spent five months measuring 25,000 slices of mouse brain, each 40 nanometers thick, and a further three months compiling the data into a 3D model.

All of that work represents a portion of the mouse brain a cubic millimeter in size. For reference, the mouse brain is roughly 500 cubic millimeters in total, but the project was a massive undertaking totaling 2 million gigabytes of data.

While completing the connectome of the mouse is still a ways off, some researchers think it might be accomplished in the next decade.

Of course, then begins the difficult and horrifying work of figuring out what all those connections do and how they behave in concert with one another.

The trouble with gaining a clear picture of a brain is that it doesn’t scale easily. But understanding how increasingly large brains result in consciousness and behavior is not simply a matter of mapping a larger landscape. As the connectomes in question grow in size, they also grow in complexity.

C. elegans was relatively simple, with only a few hundred neurons to isolate. The total number of possible relationships was relatively limited, and still, it took decades.

Unraveling the inner workings of a mouse would likely take significantly more time, lasers, and cute little martyrs to science.

It’s easy to see, then, how the prospect of understanding the human mind represents a seemingly insurmountable obstacle. Yet history is marked by horizons previously deemed uncrossable.

What might not be far off, however, is a simulated animal inside your computer, your phone, or living in a dongle on your keychain, one that would be indistinguishable from the real thing. Virtual pets could make a return to popularity in incredible new ways, no longer the shiftless amorphous creature of your youth, eating pixelated candies and pooping in your pocket, but something that behaves in precisely the way its real-world counterpart does. Because, for all intents and purposes, it will be real.

While that might seem like a frivolous use of the hard-won knowledge we’ve attained about how brains work, it might be worthwhile.

The road ahead is long and difficult, and every step forward makes the next step easier to take. We might start along that road by shambling sloppily down the driveway, but then, before long, we’ll be sprinting full-tilt from the worm-Lego hybrids worming our way.