BIDS structure for longitudinal dyadic data

1

Hi Neurostars,

I am trying to come up with a BIDS structure for data that is both longitudinal and dyadic. Importantly, there is some behavioral data that is shared between members of a dyad, as discussed here. This approach uses the session entity to track dyads, but since my data is also longitudinal, I am already using the session entity to denote different time points. My current thought for hacking this is to use the acquisition entity for dyads as follows:

rawdata/
    sub-S001/
        ses-1/
            sub-S001_ses-1_task-video_acq-dyad001_bold.nii.gz
        ses-2/
    sub-S002/
        ses-1/
            sub-S002_ses-1_task-video_acq-dyad001_bold.nii.gz
        ses-2/
    sub-S003/
        ses-1/
            sub-S003_ses-1_task-video_acq-dyad002_bold.nii.gz
        ses-2/
sourcedata/
    beh/
        acq-dyad001/
            ses-1_task-conversation_acq-dyad001.m4a
            ses-2_task-conversation_acq-dyad001.m4a
        acq-dyad002/
            ses-1_task-conversation_acq-dyad002.m4a
            ses-2_task-conversation_acq-dyad002.m4a

But if this is too flawed or there’s a better way to do it, I’d appreciate any feedback!

Thanks,
JD

2

Given that the acq entity is the usual fall back solution, that kind of makes sense to me.

As usual make sure you clarify this in the README of your dataset so that others (or you in 12 months when coming back to the data to reply to reviewers) can understand the logic of the naming

1 Like
3

@Remi-Gau thanks so much for the feedback. That’s very interesting to learn that the acq entity is a common fall back solution! Makes me feel better about doing it this way.

And I will indeed make clear notes about this in the README.

Thanks again!