search

jensroes / nonadjacent-sequence-learning

Learning nonadjacent sequences
Apache License 2.0
0 stars 0 forks source link

Booking the SR EyeLink 1000 lab #1

Open jensroes opened 2 years ago

jensroes commented 2 years ago

The lab can be booked here: http://socscibookings.ntu.ac.uk/

However it is important to be collegial about this meaning don't block entire weeks because a participant may come around but book it when you actually need it and cancel your booking if you don't need it. I think communicating this on the eyetracker Teams page might help a lot. Ben is probably the only other person that is using the EyeLink 1000 right now but that might change.

jensroes commented 2 years ago

@karlnicholls just so you know my understanding is that the eyelab is not going to be super busy over the next weeks. Obviously do use the system above but I think there isn't going to be a lot of competition.

karlnicholls commented 2 years ago

@jensroes thanks for letting me know. I figured that was the case, but I like to book it just to cover myself. I've had two participants today with a third one at 15:00.

jensroes commented 2 years ago

Yes, I'm not saying you should skip booking the lab. Just saying there are less problems with potential overlap. Good to hear you're making progress with the data collection. How did it go?

karlnicholls commented 2 years ago

Today went very well. I had a participant on Friday, but had issues capturing data from her whenever the green circle was to the top right of the screen. Wasn't sure if it was a participant issue or an eye tracker issue, so I let her go. Even Ben was stumped by it. She did offer to come back and try again another time as she's doing her own SPUR project in the Taylor building anyway.

To clarify, she couldn't get past the second circle in the sequence.

jensroes commented 2 years ago

There are many reasons for why this could be happening and we definitely need to have a look if it is happening systematically with a lot of participants. There might be a technical issue behind this. Other reasons might be that the participant had changed their position after calibration, or the calibration was great for that corner of the screen, or something to do with the eyes of the participant. If this happens very early it might indeed be that the participant changed position or the calibration wasn't great for that part of the screen. Definitely keep an eye on that (pun partly intended).

jensroes commented 2 years ago

Not really the right issue for this question: every participant should be doing two blocks which are separated by a recalibration opportunity. Is this what you saw in the sessions you recorded? Was there an option to recalibrate?

karlnicholls commented 2 years ago

Yes, there were two blocks and I recalibrated in between the two.

karlnicholls commented 2 years ago

There are many reasons for why this could be happening and we definitely need to have a look if it is happening systematically with a lot of participants. There might be a technical issue behind this. Other reasons might be that the participant had changed their position after calibration, or the calibration was great for that corner of the screen, or something to do with the eyes of the participant. If this happens very early it might indeed be that the participant changed position or the calibration wasn't great for that part of the screen. Definitely keep an eye on that (pun partly intended).

Because the experiments today went off without a hitch, I'm presuming the issue was with the participant. I will ask the participant I had issues with to try again (as she offered anyway), just to make sure it was an issue with her and not the eye tracker. It definitely wasn't an issue with movement. We tried for almost an hour to get it to work properly and I had Ben present for at least 20 minutes of it.

jensroes commented 2 years ago

I will ask the participant I had issues with to try again (as she offered anyway), just to make sure it was an issue with her and not the eye tracker. It definitely wasn't an issue with movement. We tried for almost an hour to get it to work properly and I had Ben present for at least 20 minutes of it.

That's odd. My experience is, if you can't make it work in 5-10 mins, there isn't much you can do. I'd definitely not want to keep a participant for an hour. Even if you'd get a successful calibration in the end, you don't want to torture participants for 20 more mins after they already spent an hour saccading across the screen. As I said, there are many reasons why this could happen and that you haven't had issues with other participants suggests it is not a technical problem.