Experiment prep

Tomorrow is my first experiment! To get ready I have a few loose ends to tie up before the big day. Mostly I just need to write everything out, make a list of items I need to set up, organize a few things, the stuff no one thinks about when they head off to an experiment. Tomorrow is going to be bumpy, but the first time always is, so best to be prepared.

This is basically an update from the previous post where I talked about the big day (here). I’ve been incredibly busy this week with other experiments and work that needs to be done. I’m not 100% ready yet, but I’m close. I also realized that tomorrow is friday the 13th and my experiment starts at 1300 hours, oh and apparently there’s a blood moon this weekend, so you know all signs point to I’m doing the right thing.

I’m not superstitious, so I don’t really care, but it is funny and one hell of a coincidence. Still, a lot of good things happened this week so maybe my bad luck collided with the day and boom, a string of good luck came out.

So tomorrow I’ll be arriving about five hours prior to the start of the experiment to set up, finish putting the equipment together, and to grab everything I’ll need from our lab to transport to the lab across the campus that I will be doing the experiment in. It’s a mess, but again first experiment, don’t have my equipment, and there are a lot of moving parts at the moment. So, in short, busy.

Now for those just starting their own human experiments, I should probably use this time to pass on a few things I’ve learned along the way. Mostly to write things down. When you look at data after the fact, you lose all context for that data. We add “flags” or “markers” to the data to tell us when an event occured. Say we’re testing walking, we could add a marker when the person started to walk and another when it ended, the stuff in the middle can be segmented by step if we know where in the huge amount of data the start and stop points are no matter how large the data set is.

However, we lose all context of what was going on when it happened, so writing down notes about the experiment as you’re doing it to remind yourself of what happened that didn’t go according to plan. For instance if I were testing walking, I would make a note if my participent tripped or if I had to start the task over. Or maybe I put two markers instead of one into the data (that happens a LOT) I would make a note so I could go in, take out that marker, and properly segment my data (chop it up for analysis).

The second piece of advice, take pictures of all the stuff you’re doing. I do this semi-religiously and I still on occasion will be annoyed with myself that I didn’t take a photo of something in particular. Since I work with a lot of cables and sensors, taking photos of the way I connected things, the way they are attached to the person, the way the person is performing tasks (video here is good too), etc. can be a lifesaver if something goes wrong later. With a photo you can track down the source of the problem or at least eliminate possible sources because you can inspect the way you set things up or attached them to the person.

Third, check lists are your friend. Seriously, you have no idea how many times you can miss something when doing an experiment. The best way to avoid forgetting is to have it written down so you don’t have to remember. I don’t want to think that I’m forgetful, but the stress you put yourself under trying to remember a million things while doing an experiment is enough to make anyone forget. Plus there’s the added stress of working with people, who need breaks, food, water, to use the bathroom, etc. so you need to be quick with your experiments. That is a lot of pressure, so write out your experimental protocol and check it off along the way. You would be surprised how many times that has saved me over the years.

Lastly, don’t beat yourself up if things go bad, especially the first time. I’m a perfectionist, so it annoys me when things are not done to a high standard. I beat myself up a lot for stupid things. So this is easier said than done and maybe this is more for my benefit than yours, but mistakes happen. Things will happen outside of your control that you will not be able to account for or fix. Off the top of my head, I’ve had equipment break the day of the experiment, I’ve had a computer crash and lost all the data from the experiment (this happened multiple times, so save often if you can), I’ve even had datasets where nothing was recorded and we couldn’t figure out why. Things happen, that’s part of experimenting.

So yeah, tomorrow will be hard and future experiments until I get my equipment will be difficult. Mostly because we have four different labs I’m borrowing from at the moment, but that will (hopefully) be addressed soon. I’m hopeful that tomorrow will be okay, but I’m also realistic. I know that what I can do in the timeframe I was given will be limited because setup and teardown will take twice as long because it’s the first time, equipment will do weird things we didn’t plan for, and there will just be so very many mistakes. But we learn and move forward. The only downside is that once you’ve gotten into your rhythm, the protocol is done and it’s time to move onto something different.

Life as a researcher isn’t always easy, but it can be very rewarding. At least that’s what I keep telling myself.