We finished up the studies today, getting 5 more people, bringing our total to 22 people. Since we were aiming for 20, this seems reasonable. Today’s experience did not differ too much from our previous ones, although this time I did make sure to specifically let people know that speeding up, slowing down, going backwards and going forwards were all equally acceptable. Due to this, we saw some people exaggerate their movements more, which brings us a little closer to capturing the subconscious movements of the eye we want to be tracking.
Beyond this, I spent most of my time looking up tests that we might want to do on our statistics. It seems to me that there are a number of tests that we could use to compare distributions. I read this document on paired T-tests, and if I recall my Math 35 correctly, that’s pretty much exactly what we need. At a more basic level, I also read these lessons on comparing two populations and started making a list of steps we would have to go through when comparing our data. Basically, it boils down to the fact that we have very few data points, meaning the assumption that the distribution is Gaussian will have to be checked before we can go ahead with computing t-statistics. Tomorrow I was planning on reading some more about tests, and then going back to work on the paper and the presentation.