baseline vs. anticipation

Zahra

Dear Sebastiaan,

I am working on a data set of pupil diameter, recorded while the participants had been touched on their arm, either based on their choice (of the touch characteristics) or randomly. We are looking into the effects of the touching on the pupil size.

My question is, do you know of any papers about whether the few seconds before a stimulation (here the touching), which is normally used as baseline, could be thought of as an anticipation phase?

I need to know if this anticipation phase needs to be studied separately, and if this anticipation makes the phase not suitable to be used as baseline?

Thanks in advance for your kind guidance.

Bests,

Zahra

sebastiaan

Hi @Zahra ,

In any experiment, participants learn what a trial looks like (e.g. they know approximately when the stimulus will occur), and so there's always some form of anticipation. This is by itself not a problem for baseline correction, as long as the anticipation is the same for all conditions, and therefore affects the baseline for all conditions similarly.

The key question therefore is: during the baseline period in your experiment, can the participant already predict what kind of stimulation will occur? If no, then (without knowing the details) I suspect that you can safely apply baseline correction. But if yes, then baseline correction is risky.

Do you see what I mean?

— Sebastiaan

Zahra

Hi @sebastiaan,

Yes I understood the whole concept, thanks a lot!

In such an experiment, I am interested in how different the arousal is affected in conditions where the participants had a choice and the ones where they didn't. I am going to look at the peak of the pupil diameter in each condition, but area under the curve and latency could also be my other dependent variables.

Can you please guide me to papers where they might have looked at these dependent variables in pupil data? I'm looking for ways of extracting them out of my data, whether it is first some routine pupil preprocessing and a simple calculation of maximums, etc. or there's more to it.

Thanks!

Zahra

Hey @sebastiaan

also related to this question, as you have worked on touch-induced changes in pupil size;

I am removing pupil size samples from my data that pass a threshold of dilation speed in comparison to their neighboring samples. Since I'm looking into the changes induced by touch, I am curious if I have to rethink this threshold as maybe the dilation would not change with the same speed while being touch, as compared to normal dilation speeds.

Do you maybe have any references in mind you can refer me to in this regard?

Thanks a lot,

Zahra

eduard

Hi,

Can you please guide me to papers where they might have looked at these dependent variables in pupil data?

Not specific to pupil data, but if you want to analyse temporal information of time series data (like pupil dilation is), you can check out this paper: https://www.frontiersin.org/articles/10.3389/fnins.2018.00765/full

It has been written for M/EEG data, but it translates pretty much directly to pupil data.

Hope this helps,

Eduard