Make NiStats terms compatible with BIDS terms

[kchawla-pi]

PR #246

[kchawla-pi]

Any suggestions where/how I should start finding the nistats terms which need replacement?

[bthirion]

key concepts are: onsets, confounds, fMRI file, anatomical files. Try and check those, e.g. in first_level_model API.

[kchawla-pi]

@effigies @tyarkoni @satra @chrisfilo
Feel free to chime in and add any term that you have encountered that is contrary to BIDS spec, where you see it, etc

[kchawla-pi]

@bthirion I made my attempt and I think someone with more experience with fMRI would be better suited for this task. @GaelVaroquaux suggested Juan or Anna Louisa, if not you.

Or maybe I can work with them, they can point and I can shoot.

[kchawla-pi]

I may replace paradigm with event during the course of working on issue #223
Don't let me delay your contribution to this, I will give precedence to your work on this issue.

[bthirion]

Note that i don't have time to work muwh on nistats till next week :- (
Please go ahead.

[tyarkoni]

@kchawla-pi I've meant to contribute to the discussion and look at the code, but unfortunately I'm completely swamped for the next 2 weeks. Keep in mind that the core BIDS-Stat-Model spec will probably be finalized after the Stanford meeting next month, and at that point it'll be safer to lock in any changes that attempt to make nistats more BIDS compliant. I'll try and look at this before then though, if you haven't already finished the work.

[bthirion]

I may replace paradigm with event during the course of working on issue #223
Don't let me delay your contribution to this, I will give precedence to your work on this issue.

@kchawla-pi Have you done / are you doing this ?

[kchawla-pi]

Soon as I prepare #234 for review, which should be today. Then I will start replacing the phrase paradigm with events in code and docs.

[kchawla-pi]

@tyarkoni Thanks Tal!

[kchawla-pi]

In some places in the documentation, the word paradigm is not literally interchangeable with events. I was thinking of using "sequence of events" or "events sequence" there.
Concerns?

[bthirion]

No concern on my side.
I think we can use "experimental paradigm", that is not ambiguous. "paradigm" alone is ambiguous.

[kchawla-pi]

I am closing this issue, please file new issues for other term non-compliance.