It seems simple enough: Your task is to trace lines with your computer mouse while listening to soothing music, drawing the branches of a neuron. You can rotate the block where the spidery neuron is embedded, and zoom in to see the details. It’s fascinating stuff, if you think about how you’re piecing together the parts and wires of your brain.
But as you follow faint signals consisting of blurry white dots, you realize that this game is less connect-the-dots, more hide-and-seek — it’s often about guessing where the branches lead and erasing mistakes in the process, wondering if your work is even remotely correct.
Even if you feel like you’re failing, though, you keep trying for one heartening reason: you’re helping advance brain science. And you’re at the forefront of a 21st century trend: “citizen science” initiatives that use data from game players to further ongoing research, including brain research.
This neuron-tracing game is called “Mozak,” the Serbo-Croatian word for brain, and is among the latest entries in this category. Created by the Allen Institute for Brain Science and the Center for Game Science, the free online game has attracted around 2,500 players since its release last November.
They’re helping to fill a major scientific gap: We still don’t really understand how neuron circuits in our brain are structured or how they work. From images of 3-D neurons inside living brain tissue, players can trace and reconstruct shapes of human and mouse neurons, which can then be classified and studied. This information may eventually help scientists understand and develop cures for brain diseases like Alzheimer’s.
“Already, after just six to eight months, ‘Mozak’ has accelerated the neuroscience reconstruction of neurons by 3.6 times compared to what scientists were doing up to now,” Center for Game Science director Zoran Popović said. “Now we’re looking into helping not just the Allen Institute for Brain Science but seven or eight different international labs across the world. So we’re hoping to make an impact on a global scale.”
Other successful examples of such “citizen science” include ” Eyewire,” from Sebastian Seung’s lab at Princeton, and ” Foldit,” a protein-folding game that Popović previously worked on. Boston-based “Eyewire,” which is similar to “Mozak” in that players try to reconstruct 3-D neurons from stacks of slices made by a diamond knife, is the granddaddy of citizen neuroscience games to date: More than 150,000 people from 145 countries have contributed to its attempts to map neurons in the retina, according to their website, and the data set has resulted in three published papers.
“We have a very committed community, where some of these people have been playing for years and are really invested,” Eyewire developer William Silversmith said. “We also had really nicely made renderings of some of our cells by our designers that even made it into textbooks.”
The Eyewire team is working on a new game called ” Neo,” to be released this year. It aims to have players find missing links in the brain and fix mistakes made by an AI companion called “MSTY” as it automatically builds neurons. To make the game more engaging, “Neo” will add other superhero characters and storylines as well.
“As important as the work is, [“Neo”] is a little bit repetitive,” Silversmith said. “So we wanted to give people a fun story with “Eyewire” heroes and characters to relate to.”
After creating “Mozak,” the first thing Popović noticed was that neuroscientists themselves were working significantly faster with “Mozak’s” new tools because they were so easy to use, he said. And since “Mozak’s” scoring is based on a consensus, he said, the science coming out of it is more accurate and robust than what the experts were initially doing.
“Even though we designed for complete novices, [researchers] decided to ditch their state of the art tools and use our mechanisms instead,” Popović said. “And then we rolled it out to people who, over time, became better and better at this reconstruction to the point that now they’re finding mistakes in what experts originally labeled.”
“Mozak”community manager Saira Mortier calls these proficient people the “power players” — they spend hours on “Mozak” almost every day, and are usually at the top of the leaderboard.
“These people are the true game experts,” she said. “They know when something is amiss, and they provide crucial feedback on both existing and new features.”
But why are these people, many of them non-scientists, volunteering hours of their time to play such a slow-paced game? One common misconception is that these games are addictive or give you an adrenaline rush, like “Candy Crush” or “Minecraft.” But according to Mortier, most people who play “Mozak” are interested in science, not necessarily video games.
“Everyday people are making headway on understanding the most fundamental part of being human – it’s incredible, and that’s a real motivator,” she said.
One thing that makes “Mozak” different from other citizen science games is that it takes more time for players to start making meaningful contributions, Popović said. It’s designed in a way that you’ll need more than just 10 minutes to become a decent player, but you can reconstruct neurons at a faster speed than in “Eyewire.” On the other hand, “Eyewire” is more detailed, and works only with neurons in the mouse retina.
Another unique feature of “Mozak” is that players can interact with scientists in real time, so they get direct feedback that helps them understand what they’re doing correctly or incorrectly. If you’re stuck or confused about a particular puzzle, for example, you can look at hints that scientists have posted or ask other community members.
“So there’s kind of a mass-scale apprenticeship process where one expert is training hundreds of people to become much better,” Popović said.
Through this online “apprenticeship,” the neuroscience community is growing and reaching more people.
“It’s not just scientists working with the outcomes of ‘Mozak,’ but really, people are slowly pulled into deeper and deeper processes of neuroscience,” Popović said. “It’s really kind of an avenue towards increasing the population of neuroscientists working on this problem by a factor of 10.”
Bob Bondi, a retired software developer and active bass guitar player in a local band in Bend, Oregon, is one of those people who have been sucked into the world of neuroscience through “Mozak.”
“It’s made me pay attention to the news that has to do with neurons and the brain more,” he said. “Some of the discoveries have been absolutely amazing.”
Bondi has been playing “Mozak” since it came out, normally logging around 10 to 20 hours a week. While he doesn’t enjoy other video games, he plays “Mozak” because he finds it both calming and mentally stimulating, he said. He sometimes likes to think about the neurons on the screen as pearls or streams of stars.
“I’m pushing 69 now, so I just need something to keep my mind active, not video games that require a lot of energy to play,” he said.
When players become really advanced, Popović argues, they should be hired and paid for their work.
“Some of these people have become so good that there’s a really strong rationale for a lot of these labs saying, ‘Hey, we want to pay you to do this because you’re highly skilled and we desperately need that,'” he said.
While he doesn’t know any labs that have actually hired players, Popović thinks that it will happen in the near future. He’s even suggested that labs provide incentive structures for players by including funds to pay them in grant proposals.
“It would pay off for researchers significantly, and it’s also a novel way of creating academic professions,” he said.
Along with the traditional route of going to grad school and working in a lab, there could be an “evidence-based fast track where you join the apprenticeship process, and then you become better at the task to the point where you become an irreplaceable part of the scientific process,” he said.
Maybe eventually, games like “Mozak” will lead to new jobs and professionals, the way traditional video games have created eSports.
But even if he’s not getting paid, Bob Bondi said, he plans to continue playing “Mozak.” He likes being part of a community of other players and scientists, in which he gives and receives help. After consensus is reached, he likes that he can check his work and improve. But most of all, he plays “Mozak” because of the knowledge that he’s contributing to neuroscience as a citizen scientist, he said.
“Maybe,” he said, “somehow, part of my tracing will be something that makes these guys sit up in the lab and discover something they didn’t know before.”