When I switch to a kindle or a physical book I can feel those impulses to divert my attention for the first minute, but they dissipate quickly when my brain realizes I'm not on a computer. Then my brain stops trying to interrupt itself, hunting for some dopamine shot.
But to stay on topic, if I were to give you a wild guess, I think it might have to do with the sort of feedback you get form reading physically compared to digital. The mere act of turning a page might trigger positive feedback loops in your brain that you don't really get with the sort of instant-page turning from e-readers.
It might also have to do with people just not being used to the medium, I got my mom an e-reader and for a while it worked for her but eventually she went back to physical books, she's just much more used to them.
I can understand in backlit screens were the light can cause eye strain and that's why I cannot read nor comprehend as much.
But on e-paper? My experience is that I read as well on e-paper as on real paper.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6384527/
says comprehension is about the same, but that a physical book makes it easier to remember if something happened earlier or later in the book.
On a completely unscientific annecdote that seems to match my experience, where with paper I subconciously remember positions in the book, while on a kobo every page feels the same. That said, full text search is sorely missed on dead tree.
I even find with a digital device a distraction aspect to having 100 books I want to read at my fingertips. It is the paradox of choice. This book I have open is good but maybe one of the others would be even better.
Of course, I doubt there would be any difference if the digital device only had one book on it at a time and could not do anything but read but that would largely defeat the purpose of a digital device.
The kindle is unique in that you can't do anything but read on it. Same thing with a physical book.
I also believe that the physical traits such as how far you are in a book and the smells / sights can also have plenty to do with it. You don't get that with a digital equivalent.
With all that said, I don't see any difference in comprehension. Some books are simply better on physical and some are simply better on digital.
Using Clayton Christensen's Jobs To Be Done Framework, by giving your devices a job, the more successful you'll be in the outcome you seek. For example I give my iPad Mini a job to be done of being my e-reader at night. And it works wonderfully as I don't swipe through random apps if it's already my intention to read.
Even for exercise, a 30mins gym workout is often much better than an hr with digital platforms.
That is, smaller screens are bad for cognition. But bigger ones are supposed to be better.
I like paper more with no other thing taken into consideration. But a Kindle is really nice for reading thick novels and non-fictions.
For reading technical books, research papers, books with lots of graphs, figures, etc. I have a 10" android tablet.
I am very satisfied with those.
I still buy physical books regularly.
Jumping to 50 pages back intuitively with your hand feels something different from jumping to a page in a PDF.
While reading a physical book, your peripheral vision also comes into play. Although you read a particular line, your eye "sees" the graph on the other page.
When you navigate to places inside a book, you are also quickly reminded old things with a flash of the materials.
Holding a particular book, sitting and opening it is also an elaborate associative ritual. With ebooks, you are just holding the same piece of silicon brick for all of them.
The speed at which you can turn over to previous page compared to scrolling up is also something to be considered.
These are just some things that I think can explain the gap if there is one.
Everybody notes distraction. But that can be learned to avoid through self-control. I have done that.
Besides self-control, a Kindle has abundance in choice, but its distractions are nothing compared to a much more common smartphone.
My first reading-purpose tablet was a Lenovo 512 MB RAM tablet. Couldn't do anything with it besides running a few ebook reader apps.
And as I said, the choice abundance problem can be solved with self-control.
And I do not trust the conclusion drawn by the papers. Were they done with people who all grew up reading traditional books? If so, the "grouping" would be futile. All people grew up reading paper books. Now you give screens to some of them. Of course these people will have trouble reaching the same level of comfort as compared to the others.
These studies have a general reproduction problem as well.
If we draw n=20,000 1 yos, and give only paper to 10,000 of them, and only screens to the other 10,000, and monitor them for 20 years, will we notive cognitive difference? I might be wrong, but I don't think so.
Aside from the technicalities I mentioned, I don't think paper is something special. Because paper is modern enough to not matter from an evolutionary standpoint.
Because of my insatiable hunger of books and money constraints, I have been reading digitally for a long time. I read 300-700 pagers regularly on a 4.5" smartphone when I was little.
As an adult, I have very little troubles with screens now.
A Kindle is an English novel reading nirvana, and a 10" tablet is a research paper reading nirvana. The experience can't be better for me.
I read on bed, I read on commute, even toilet. Books are impractical in many situations. And you can pack only so many of them.
I also suggest reading on a big tablet rather than a smartphone or an immobile monitor. You will notice the difference.