We're a little crazy, about science!

Changing the experiment

For the past few days I’ve talked about the importance of experimental design. Well sometimes midway through you realize a better way to do things. That was yesterday when I realized the thing we were looking for in our experiment could be found a better way. I’m not thrilled about this, but sometimes it needs to happen and I think we will have a better chance of finding what we’re after if we do it this way.

A few posts back (here) I made a breakthrough in a type of analysis I was doing. Basically I was too close to the problem to see the solution that was right there the whole time. This happens a lot, or at least it feels like it. For my summer experiment (paper number 5, since I keep track of them that way now) I had some troubles and came up with a solution in a matter of moments after discussing it with my Co-PI (here). It was a good feeling to be able to figure out a workaround and advocate for it in a reasonable manner instead of waiting for the solution to be handed to me, progress I guess.

It was a solution and a good one. I still stand by the solution I came up with and hopefully in a few months I can go into detail about all this, but I realized while it was a good solution, there was a better one. For those who don’t recall we were doing a fairly basic study and since I can’t go into details, it involved electrical stimulation and EEG. That was the problem.

EEG records microvolt changes, very tiny, tiny changes in electrical charge on the scalp because the changes in that charge correspond to activity in the brain. We’ve shown this, we know this is real information, and we use it pretty regularly. It’s kind of amazing that it works, but here we are nature working in our favor. That seems to happen pretty regularly around here actually now that I think about it and I’ll probably talk about another way our biology works in our benefit some other post.

Since EEG is incredibly sensitive, it picks up just about everything, from the electrical current on your skin to the electrical currents in the walls (seriously we need to filter that stuff out, it’s wild). Well we were applying an electrical stimulation to the skin and no matter how little or how far away we went with it, we saw the artifact from the stimulation in our data. There was no hope of doing the experiment as planned since there is no way to clean the stupid artifact from the data. Trust me, I’ve tried, nothing works well.

So I devised a solution that involved a few tricks to work around it. I again can’t go into detail and that is incredibly frustrating since I really, REALLY, want to share! Point being the solution worked and I have two newly collected datasets to show for it.

Then yesterday I realized I was thinking about it all backwards, by combining the original experiment with the solution I had come up with we could get closer to the behavior we’re trying to test while still avoided the artifact all together. Again I hate having to talk in generalities, but I want to share and this is as good as it gets until something, EVENTUALLY… gets published. I’m hoping soon I can talk about some of my research, this is killing me and I feel like once we get one paper published the rest should go smoother (maybe…?), probably not.

Yesterday I went over my realization with my Co-PI and he approved. So we’re changing it up and collecting data using the new method I came up with. This will get us closer to testing the thing we really want to find out and it will hopefully give us a better shot of seeing what we’re looking to find.

I’m really happy with this new solution, so hopefully it works out well. I mean if we don’t find anything I guess that’s a paper too, but it would be nice if we did find something. From the data we collected, it looks like something is happening, what is going on is hard to say, but so far things look promising. By the end of the month I hope to have everything analyzed and I’ll feel more confident saying we found something, but for now all I can really say is we have something weird in the data. Maybe it’s something, maybe it’s nothing, but we won’t know until I collect the rest of it.

Story of my life, yet more work ahead.

But enough about us, what about you?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.