The Core

Mobile first indexing is a separate issue from mobile friendly site design / layout.

Mobile friendly = ranking sites with viewport settings & readable text on mobile, etc. slightly higher on cell phones.
Mobile first = entire search index driven off mobile version of a page.

For now if the content is viewable on your desktop version of your site that can drive the indexed content on mobile. But when Google goes mobile first indexing they'll (allegedly?) count content hidden behind extensible panels like normal on-page textual content across mobile & desktop versions of the site while (allegedly?) not counting desktop content which is absent from the associated mobile version of a page.

I think mobile user engagement metrics have an outweighted impact on Google's relevancy algorithms. Google has more control over their behavior (owning about 95% marketshare on mobile & tracking everything Android users do) & navigational issues on mobile are far more likely to lead to a quick bounce than similar issues on desktop.

Yes, i know all that.

What I'm saying is that I may be seeing signs of mobile first beeing testes.

Ah, I must have glossed over this part "After showing that content on mobile many pages have jumped 2-3 positions in the SERP."

My (meatheaded reply was based on "I have a responsive site"

sorry about that!

Have you tested showing some content on the mobile version of a page while not showing it on desktop & seeing if there is any shift in rank?

Even with that it could perhaps be hard to isolate things, because sometimes it seems there is more rank variation between Google & Google mobile SERPs than there is between say Bing & Yahoo! SERPs.

Quite often I see a site for which nothing has changed where on parallel keywords it goes up on one keyword and down on the other, or up in rank on say the mobile version of Google's index while staying flat or even going down on the desktop version. There are so many moving parts it is hard to be certain of almost anything.