The idea that scrolling may be jarring enough to hamper reading fits my experience. When I scroll, it doesn't feel like just my browser has to repaint the whole webpage, but it also feels like my <i>brain</i> has to reconstitute the structure of the page via a kind of inverse-repainting, just so that I can reorient my attention, before I can resume it.<p>In other words, if I<p>- have a (semantic) pointer to, say, the last word on a line<p>- am maintaining just the single last word I read in my short-term memory/register<p>- scroll and then have to look for the line I was just on before I have reoriented myself<p>then it feels like I have to do a kind of mechanistic attention-interrupt/syscall that locks my conscious interpretation of the text's meaning until I have returned to the index of the text that I was just at. I guess that also explains why sometimes, when I am simultaneously trying to reflect on the text <i>while</i> scrolling, I am significantly less able to do so fluidly, as if there were some underlying deadlock, and more often than not have to repeatedly attempt finding the next line..<p>But if you hold a book in your hands, there is much less variation in the 'streamed/online/', structural form of the text. More or less, all that my brain knows it needs to anticipate is page turning. It can figure out how to cancel out my hand movements, background visual information, surroundings, etc. from my conscious experience because that's what we've evolved to be able to suppress from our attention.<p>Maybe, then, computer file viewing UIs that have page-flipping skeuomorphisms are less attention interrupting, because they would avoid these interruptions being done more than one time per page/pair of pages?<p>Link to the mentioned paper: <a href="http://www.co.twosides.info/download/To_Scroll_or_Not_to_Scroll_Scrolling_Working_Memory_Capacity_and_Comprehending_Complex_Texts.pdf" rel="nofollow">http://www.co.twosides.info/download/To_Scroll_or_Not_to_Scr...</a>