After we identified SAC cells, then what?

So … our current goal is to follow possible synapses between the previously mapped J cell and potential SAC cells. All work on cells that were identified as NOT being SAC cells is discontinued and moved to ‘Completed’.


What happens with the ones that do turn out to be SAC cells? How long will we continue mapping them? Do we need them in their entirety or are we just looking for all synapse connections to SAC cells?

I too think these are interesting  questions and legitimate for those who provide some time for it.

I see a certain dilemma: at the one hand you want to win as many participants as possible and the game seems to help for that, but a bunch of players are mainly interested in the scientific gain of the work. So the organizers have some responsibility to keep both groups updated about the results and teach them - probably at different levels - what we may be learning from the work.

This touches another novel aspect of this approach: usually scientific work is done based on some ideas and concepts that are then studied and elaborated with sweat (data collection) and creativity. After some time you feel ready to present the results in a meeting and publish the results and your conclusions in an appropriate science journal. Until then things are pretty much kept “under cover” so not to disclose it untimely to eventual competitors (for grant money, positions).
Here data are created real time in public but they appear as little isolated puzzle pieces.  If one has some retina experience one may recognize a certain cell type
after a while but usually one does not see its spatial position and relation to previously reconstructed cells. So the organizers - deciding which pieces to hand out and how they eventually fit together - keep the “intellectual” aspects to themselves. This is the nature of the BIGGER SCIENCE game the whole project is part of.
The question is, how much of it can and should be visible at the players horizon? Currently I see no more specific comments on the type of task that is assigned to the teams or associated with a particular cellular cluster.  If the players are “blind” to the detailed goals they are not biased by any a priori assumptions and this is considered good for data collection, but at least questions as the ones above should get thorough feedback.
Some kind of regular summary reports on the work done, the particular progress and the next steps considered should be given, rewarding those who look for more than winning points. This may interfere with the usual course of science work, sketched above but novel approaches ask for novel solutions for the issues arising from them.

As far as I know from a post by blakamm here http://forum.eyewire.org/discussion/328/still-no-eyewire-games, they were originally going to trace the cells which connect to the J-cell all of the way, but since that would take too long, for now we just need to trace enough of the cells to determine what cell type they are.


I’m guessing that we will get updates as things are discovered, you have to remember that there haven’t been THAT many cells mapped by us yet when you look at the retina as a whole. I’m sure that when something interesting is found that they will let the eyewire community know as much as they can about it as soon as they can.

Hey all,


Great questions!  I’ve moved this discussion to the ‘Science’ category, since it’s more on the science side of things.  The short answer is that we want to be as open with you as we can in appreciation for the great contribution that you are making.  Most of the information about what we are studying and what our scientific goals are is disseminated through the wiki and the forums.  The wiki info is more static and stale.  The information on the forums is fresher, but it’s diffuse and the older stuff is probably out of date.  Basically, we don’t feel any need to keep this stuff secret from the world till we publish because no one else is doing anything like this!

On that note, we are trying to publish soon, so hopefully you’ll be seeing the results of that in the literature soon.  We’ll also share whatever we publish directly with the community.  Another thing worth noting is that we aren’t just studying neurons.  We are also studying all of you!  Doing science is one thing.  Helping a community of amateur scientists contribute is a whole other thing entirely.  Sometimes when you don’t hear much about neuroscience from us, it’s because we are working on ways of helping magnify your contributions, which is a not-insubstantial challenge in and of itself.

None of this actually answers your questions, but are all by way of explaining what you are seeing.  The moral of the story is that if you look through the forums and wiki, you can find a lot of information about the systems that we are studying and the questions we are trying to answer.  Also, keep asking questions, and we’ll keep answering them!

As a side note, updating the wiki is important for disseminating this information, but we don’t really have enough hands to keep it up to date.  If anyone would like to volunteer to help clean up and update the wiki, we’d be very grateful.

In answer to the original question, the primary scientific goal within this competition is to go through all the cells which synapse onto the J-Cell in question.  For each, we’d like to reconstruct enough of it to identify it as being either an SAC cell or some other cell.  For the SAC cells, we’d like to reconstruct enough of it to tell which direction the dendrite is coming from.  With that data, we’ll be able to provide a fairly accurate map of where the J cell receives all these SAC inputs from, and ideally that layout will correspond to a simple model of why these J cells behave the way they do.

Everything else is icing on the cake.  If we could reconstruct each of the SAC cells in their entirety, that would be great, and we could look at what other types of cells they synapse one.  We recently added some new J Cells to each team.  If we could reconstruct those we can show that the J cell we are focusing on is not unique, that there is some similarity across cells in the same region.

Think of it like kickstarter.  You have some baseline goal, and then you have reach goals.  At a minimum, we need to reconstruct the incoming branches of SACs synapsing on the J.  Everything else after that is a bonus!