An online game makes people contribute to neuroscience research.

As a kid, one of my dreams was to be able to look at the brain with a microscope and find every connection between each neuron. When I started neuroscience research, I realized this was an unrealistic dream – the number of connections is too big. To find all the connections between every neuron in a chunk of brain smaller than one cubic millimeter, I would have to spend years, if not decades, under the electron microscope, clicking on a computer. Completing the entire brain would be impossible in a lifetime, even for a large group of scientists. Recently, a professor at the MIT has developed a tool that might change this.

Sebastian Seung is interested in the connectome – that is the entire collection of every single connection between every single neuron in an organ like the brain or the retina. He’s got pretty much the same dream I had when I was a kid, but apparently he just never gave up. With a large group of scientists from his laboratory and undergraduate students, he developed an online game called, where anyone can browse microscopic photos of neural tissue and reconstruct the branches of actual neurons. The tool is currently used for specific retina cells but it makes no doubt to me that a similar tool could potentially be used to reconstruct neurons in the brain. How many neurons and whether it becomes possible to reconstruct the complete brain of, say, a small invertebrate, will depend on the involvement of the community. We don’t know enough at the moment about the manpower that will be available. But that’s the exciting part.

A ganglion neuron reconstructed from EyeWire. It takes from 15 to 80 hours for a trained scientist to reconstruct this. How much time can it take to a worldwide community of volunteers ?Photo from the EyeWire wiki released under the Creative Commonslicense. 

The problem that scientists are facing when reconstructing the anatomy of a neuron is that neurons have many branches that are intermingled. Imagine a football stadium filled with very long spaghetti – then slice that spaghetti bowl in thin slices, and try to find all the pieces of one of the spaghetti and stick them back together. That’s pretty much what the challenge of reconstructing neurons from anatomical slices looks like.

If you like puzzles, it’s very similar. However, the data that comes out is actually useful to further our understanding of the brain as it allows scientists to reconstruct the complete anatomy of some neurons. For years, this job has been done by graduate and undergraduate students, at a very slow pace. I myself have spent a couple of months on the reconstruction of this neuron (Figure 3). This new tool makes it possible for anyone – no scientific background required – to bring a small contribution to the collective effort. This might allow us to understand brain connectivity at a level that was impossible to reach through the efforts of individual scientists. For now, the goal is to map the connections of the J cell during what they call the J-Day (December 10th, next monday). The J cell is one of the types of neurons that is found in the retina. Until J-Day, you can already subscribe, follow the tutorials and get some practice so we’re all ready to do a sprint reconstruction of that cell on Monday!

The success of this platform will ultimately rely on people – is brain research interesting enough to attract a thousand, a million, a billion contributors ? Who knows at this time. Similarly, who could have predicted in 1995 that all we needed to create a complete encyclopedia like Wikipedia was about 300,000 volunteer editors ? The success of EyeWire will also rely on their ability to use the data provided by the public efficiently and to communicate quickly about the scientific outcomes to users. The peer-reviewed process for scientific publications is way too long – it might take months, even years before this data gets actually published in a peer-reviewed journal. People will want a system of feedback that lets them know to what extent their contribution is useful before the final publication.

The integration with Facebook looks like a good idea. A ranking system also allows people to “compete” to be the one who most contributed on a given day, week, or month. The first few days reconstructing the J cell will provide a better idea of how much can be accomplished by the community. So for now, if you have someone who likes puzzles in your family, invite them to participate to this great project! There seems to be a big hype among people who want to improve their memory in buying those Brain Games. Well, here you have an alternative free-to-play game that will require visual acuity, 3-dimensional thinking, and on top of that it will actually contribute to science. So in the event that these games don’t improve your brain it will at least improve brain science.

Let’s see how much time it takes to a large community of volunteers to reconstruct a complete neuron!

Update: We had a virtual hangout hosted by the EyeWire team at noon on December 8th 2012, you can consult the video to learn more about the project. I asked a couple of questions to the team and their responses shed some light on their future plans with EyeWire.

0 thoughts on “An online game makes people contribute to neuroscience research.

  1. Pingback: An online game makes people contribute to neuroscience research. | Blog | sweet escape ;) |

Leave a Comment