Reading, paper and screens

There’s an interesting article in the Washington Post that concerns readers, but should also concern us as authors. It’s titled ‘Researchers say computer screens change how you think about what you read‘. Here’s an excerpt.

Reading something on a screen — as opposed to a printout — causes people to home in on details and but not broader ideas, according to a new article by Geoff Kaufman. a professor at Carnegie Mellon, and Mary Flanagan, a professor at Dartmouth.

“Digital screens almost seem to create a sort of tunnel vision where you’re focusing on just the information you’re getting this moment, not the broader context,” Kaufman said.

. . .

“Over time, it might lead to an evolution of thought that’s less inclined to look at the bigger picture,” he said.

. . .

The studies covered in the latest article were prompted by earlier research from Kaufman and Flanagan that found players using the iPad version of a disease prevention strategy game struggled with long-term strategy much more than those playing a physical copy of the game.

“On the iPad, they seemed not to focus or show consideration for the long-term effects of their decisions,” Kaufman said. “And they just lost the game much more often.”

There’s more at the link. Recommended reading.

This poses an interesting conundrum for authors. Are we writing primarily for the print market, or for the e-book market, or for both? Many would probably shrug and say, “I’m writing for whoever wants to buy my book, and I don’t care about the format!” However, that may be short-sighted. If most of our readers are buying in e-book format, wouldn’t it be a good idea to structure our books in such a way as to appeal to the reading pattern such customers are likely to adopt? If most of our readers are buying paper copies of our books, shouldn’t that affect how we write them?

I can see both sides. I’d say well over 90% of my sales (to date, at any rate) are in e-book format; yet I have a solid core of fans who prefer paper, and in fact won’t read an e-book for preference. They say they just can’t ‘get into’ the book until they’re holding it in their hands and can turn the pages. Therefore, if I ‘optimize’ my writing towards an e-book mode of reading and comprehension, I risk putting off (and perhaps losing) those who prefer a paper copy. What’s more, if my market grows and I start getting my books into bookstores, it’ll be very important to make the paper edition as readable and attractive to potential customers as possible.

What say you, Mad Genius Club readers? Do you agree with the research cited in the Washington Post? What’s your preference, e-book or hard copy? Do you find that your approach to either format has changed over the years, given greater exposure to reading on screen instead of on paper?



Filed under Uncategorized

36 responses to “Reading, paper and screens

  1. Draven

    Short answer: no, not really.

  2. I’m wondering about the exact parameters of the survey. I can’t find the link so (salt), but I remember one survey where they failed to account for age and a few other factors and drew the same conclusion. You dug deeper and discovered that they’d polled on college campuses, and there was a stronger correlation between ‘traditional’ students (straight out of highschool) reading even their fiction for details. I’m trying to remember if they had a demographics chart in the back, or if it was another grad student that had looked at the numbers and done his own rebuttal paper on them.

    The article doesn’t report on what factors they considered, what they discounted nor why. Neither does it seem to correlate by age. I am especially suspicious of the initial work that lead up to this work. On the digital game vs. the paper and pencil game, what resources were given to actually project trends in the digital version? I can draw a graph on paper, but a lot of games have me diving for Excel if I want to project any kind of trend, and if it’s some level of ‘real time’ that’s not possible, where as the table top game may be less real time. If they didn’t take that sort of thing into account (don’t know, don’t have links to the survey) what might they have not taken into account in this particular study?

    From a reader stand point. I have discovered that a screen does not make me more detail oriented. On the other hand most of the books I am getting for my reader are old friends. Which means when I read them on the reader I have read them repeatedly before. Which does bring out nit picky details because my brain knows the story well enough to actually focus on those things. Was this kind of phenomenon taken into consideration in the above study? What kinds of texts were being presented?

    For usage, I think, in the realm of fiction things are getting less leasurely and tight consistency is important. Yet I have heard casual readers as well as dedicated ones, digital and paper, complain about issues in the broad spectrum of stories. “It is glaringly obvious that X plot point taken to its logical conclusion will result in Y Very Bad Thing, but that is never addressed nor hinted and and when Very Bad Thing should happen, it is handwaviumed away.” Is that a detail orient thing or a broad picture sort of thing? To me it seems a broad thing. People don’t comment as much on themes, but they still seem to react to them. So… maybe they were talking about people conciously noticing themes? I think we should keep on as we have, see what sells consistently. That’s the best indicator of ‘doing it right’ that we currently have. I’m seeing that the core of what sells is the same as what always has. Stories of hope, courage, and honor. People becoming more than they were. Maybe what has been lost is not the ability to see these things… but the willingness to discuss them?

  3. Yeah, without seeing the original research paper, the methodology and the stats I would take this with a pinch of salt.

    First question would be can you replicate the experiment? If not then what’s changed?

    Of note is the reference to the game: a port of a board game to an iPad. I can imagine several answers as to why the players of the board-game were able to achieve better outcomes over the players using the iPad, which would boil down to implementation or more precisely taking a 3d game and making it a 2d game: in short info loss.

  4. Robin Munn

    My first thought, before I found out they were talking about a board game, was that they were mistaking responses for attention. Specifically, that they were mistaking “Oh, he only responded to that one section of the blog post” for “Oh, he only paid attention to that one section.” When in fact, the commenter may have agreed with most of the post, but only replied to the one section where he disagreed and therefore had something to say. (Because “Yep, that’s exactly right” doesn’t usually lead to much discussion.)

  5. Like the others, I am suspicious of the research. All we have is the claimed result–which fits nicely into an attention-grabbing headline–and none of the details of the studies.

    For example, there’s the study about playing a board game vs. playing the same game on an iPad. Among my many hobbies is playing board games, both the physical games and iPad implementations of them. I readily admit I make more mistakes playing on the iPad, but it has nothing to do with digital vs. physical. When I’m gathered around a table playing a game, I can study the board while other players are taking their turn. It’s easier to keep my focus on my strategy because everything is in front of me the whole time. When playing on the iPad, I’m either using the pass-and-play method, where the iPad is passed around and players can only study the screen while it’s in their hands, or I’m playing asynchronously with players all around the world. In the latter case, I only look at the screen when it’s my turn and, many times I take those turns during a brief lull in some other activity (usually at work). I tend to rush the turn in those cases, both so I can get back to what I’m supposed to be doing and so other players aren’t held up for too long. Simply put, my focus on the iPad game is different because the way the game is played is different. That’s an effect of playing on the iPad–at least the way I play on the iPad–but it has nothing to do with playing on a digital screen.

    I find the studies about reading equally puzzling. Why would a story presented on a Kindle be perceived any differently than a story presented on paper? How much of the result is selection bias on the part of the researchers? How much of the result is due to the selected participants? Did they test the same people with both physical and digital books or was there a paper group and a digital group? Were the two groups otherwise identical, such as randomly divided from a single pool of participants who all met specific criteria?

    We’re missing too many details to know just how reasonable these studies are. I also note that the only two studies we’ve heard of which investigate this issue have at least one researcher in common. Is anyone else investigating it? If so, what are their results?

    Too many questions, not enough answers.

  6. Bjorn Hasseler


  7. In addition to the valid points that others have made above, the PDF format is just a simulation of paper, and thus has all of the disadvantages of paper while adding a few of its own on top.

    It doesn’t make full use of the capabilities of the digital environment (in fact, it’s more or less explicitly designed to not show anything on the screen that can’t be shown on paper).

    There’s a reason why PDF isn’t the default format for Kindle books. 🙂

  8. Margaret Ball

    Granted, PDF is a terrible format. However, I have yet to find an ebook format that makes it easy to read nonfiction. When I’m flipping back and forth between “here’s what the author asserts in chapter 5” and “this is his supporting citation,” and “wait a minute, doesn’t this contradict with what he said a couple of chapters ago”, a dead-tree edition is much easier to use than an ebook.

    Fiction, on the other hand, works well in ebook format – at least if, like me, you favor fiction that tells a compelling story, starts at the beginning and finishes at the end! I normally read entertaining fiction straight through without backtracking, so the paper/ebook distinction isn’t a big detail. (And I strongly prefer to read long complicated novels in kindle editions; then when the author brings on a character who previously made only a cameo appearance over a hundred pages ago, I can search on the (by now) unfamiliar name, locate the earlier appearance and refresh my doddering memory.)

    • This fits my experience. Fiction and short abstracts work find on e-readers (have not tried on an i-Pad) but non-fiction is still awkward at best and very problematic at worst. Even if there were hot-links to illustrations, it would still require more work to read and absorb the material.

      • Aye, I find I make many, many references to the Index at the end to look up things as I read, and flip between a few things. Multiple bookmarks happen – and did long before Post-It made it easier to do.

        As for the study… e-format vs. “real” game? So, simulation versus “really” doing, in a way? Of course the “real” is going to win. But is reading one format really different from another? Using a Kindle is superficially, not fundamentally, different from using dead trees.

    • Civilis

      In college, I had no problems reading fiction as a text (.txt) file for the same reasons you cited… there’s not much to it if all you are doing is reading the text straight through. Same goes for Microsoft Word documents. (There’s a trick where if you scroll the document without moving the cursor, most programs will take you back to the cursor if you then move it, say, with the arrow keys).

      For non-fiction, if I only need to check something in the same book, I agree about preferring the dead tree version; I can bookmark if necessary with my other hand much more readily than fumbling with a keyboard. For more complicated references, I actually like web HTML format in a modern web browser. When I want to check and reference information, I can open a new tab and not lose my place. If necessary, I can open a new window and see both pieces of text side-by-side.

      As a board game person, I also question the use of an iPad for a game the players are not familiar with. When playing games on a touchscreen, I’ve found I have a reluctance to try to navigate the interface that isn’t there with games on a conventional computer due to the inherent clunkiness of a touchscreen-operated window-based GUI, especially with my tendency to fat-finger things. For me, the mouse / keyboard / window-based GUI combination works much better than touchscreen based interfaces for ‘where’s the manual? I need to look up what this option does’. If I know the game well, I don’t need to consult the manual, and usually know right where the options I need are. For games, the interface method is important; there are definite times that each of keyboard & mouse, gamepad, and touchscreen interfaces work best, and times when each is completely inappropriate.

      • If you can remember wording, you can run a search for previous comments about a topic. But of course, a really good index with links, or a really good table of contents, provides an easier way to find stuff in either an ebook or a paper one.

  9. “Do you agree with the research cited in the Washington Post?”

    No. I know that in my case, I play Sudoku on paper, computer and ipod. My times are roughly the same. I play Free Cell on the computer, Ipod and with physical cards. I win on the computer and Ipod nearly 100% of the time. Physical cards I win around 90% of the time. I play Mah-jongg on the computer and have played with a physical set. I do about the same in either format. (So-so, I’m still trying to get the hang of the strategy) I’ve played Chess on the computer and on a traditional board. I am much better on the computer where I can see the board from above.

    I think wyrdbard hit on it. They are not taking other factors into consideration. For instance, is the subject inclined to look two or three moves in advance. Is there a competitive drive that doesn’t come into play unless there is a physical opponent? My mother was like that. She could play Black Jack and do marginally well, but if something was at stake, she became competitive and started winning more.

    As to physical books over e-books. E-books are nice to get an idea if I like something, but I prefer the physical book. I like flipping back and forth in books, looking at previous scenes to make sure I read it correctly, seeing what I missed when a character suddenly sprouts wings or something. But that doesn’t change how much attention I am giving to what I am reading.

  10. _If_ there’s a difference in how fiction is mentally handled depending on paper or electronic . . . The we’ll all be trying to figure out how to best please the reader in each format, hopefully without having to completely rewrite the story.

    I suspect things like extra white space to circumvent the experienced fast reader’s tendency to grab first sentence-last sentence of paragraphs might be more important in ebooks, but that a minor change. If more repeated references and foreshadowing to keep the “big picture” in the corner of the readers mind is needed, that’ll take more work.

    How much we do will depend also on the balance of sales of our work between paper and ebook. I sell so few paper copies that “The best format for ereaders” will be my default, and I won’t spend much time reformatting, let alone rewriting, for print. This will quickly change if a route to getting my books into book stores opens.

    • I’m in the same boat (my print-to-ebook ratio is in the dozens to thousands range), so I don’t put a lot of effort on the former.

      I admit I haven’t actually sat down and read the print version of any of my books from beginning to end, other than flipping through them to spot any formatting errors, and maybe I should start, just to see if the experience is any different. I have read quite a few books on both formats (rereading a bunch of long-lost books after getting them cheap in ebook format, for example), and haven’t noticed any differences, though.

  11. I’m just writing… c4c on the ‘questions’…

  12. BobtheRegisterredFool

    1) Study involving human beings 2) via mainstream media. I’m inclined to suspect screw ups, and do not care enough to check.

  13. I too am skeptical. While the printed word has a magic all it’s own, the media it’s printed on (digital or dead tree) isn’t. The letters on the page or screen get translated into words which get translated into thoughts. That process doesn’t change when you move everything to the electronic arena.

  14. The entire study is based on the flawed premise that comparing reading a PDF on a computer screen with a printout is comparing reading on a computer screen with reading paper. No. What they’ve done is measure the effect of reading a PDF on a computer screen verses a printout. We don’t know if this was the only app on the tablet or if there were others and possible distraction. To accurately compare electronic vs hard copy, the study would have to compare e-ink readers with paper (since they are dedicated devices and present a different mind set), and both screen and print out need to be the same size.

    Next, they would have to evaluate size of the reader screen compared to the size of the print out. This can be easily done by printing the same document in both statement and letter size, and comparing that with e-ink. E-ink screens are approximately statement size, so if there is a difference based on the size of the media, statement size print outs should yield the same results as e-ink readers.

    The type of screen should be evaluated, in case the flicker from refresh rate is doing something odd. This should be done by comparing tablets to e-ink. If it seems that something is going on, the tests should be done in incandescent and fluorescent lighting to see if it’s a combination of refresh rate and light flicker (since LED lights may or may not flicker depending on the internal power supply, incandescents are the simplest way to practically eliminate lighting flicker).

    They should have thought about this from the beginning. A quick check of the study (the PDF can be found at this site )
    shows that Mr. Kaufman and Ms. Flanagan apparently thought that evaluating PDFs on iPads (the device is never stated, and is assumed to be the iPads used in the first test) is the same as evaluating computer screens. It’s not. Unless a study addresses these issues, it’s really not worth anything.

  15. Until said research is duplicated (several times) I’m skeptical of their findings. While I do find that I do much better with non-fiction (such as text books) as Dead Tree Books, simply because it’s easier for me to go back and forth, check notes, etc. For fiction I typically don’t have to check anything else, so eBooks are now my preferred format.

    As for porting a board game to a small computer screen, if it’s a large scale game I can easily see why people would tend to focus on near term details. Out of site-out of mind is a pretty universal phenomena in our culture. If you can’t see it on the screen, you’re likely to forget about it.

  16. I think that e-books have not gotten to their full potential yet. Part of this is the crossover with physical books—one does not want to privilege one set of readers over another, after all. But even the basics of “book” are not as good with an e-reader—it’s much too easy to lose one’s place (seriously, I hate it when my e-book decides I’m back at the beginning), and it’s hard to go back and check details later on that you swore you saw ten pages back… but was it ten? Maybe it was twenty as the e-reader counts. And maps should have some zoom capability…

    • Alan

      Maps need zoom (sometimes provided), with better detail at full zoom than a print book usually offers, and a convenient bookmark to flip between reading-point and reference-point.

  17. Phil Ryan

    Initially I had a hard time with ebooks mostly because of the readers. I found the Kindle reader was excellent in sunlight while I was deployed to Iraq, but the Kindle app on iPad is what I usually read on now. The exception is when reading pdf which is usually technical stuff – I found Apple’s iBooks actually works best for that. Anymore I prefer ebooks for their easy access and portability. Formatting issues are still the biggest distraction with ebook, though.

  18. Alan

    I don’t think there’s much similarity in mental processing between playing a game, especially derived from a board game (alternating between gestalt views and logically-navigated details needed for the next move) and reading a story (more-or-less linear plot and narrative flow).
    Study does not apply, whether valid in its own context or not.

  19. Unfortunately, at this point whenever I hear the words “researchers say” I put one hand on my wallet and my skeptic hand on with the other, because I’m positive I’m about to be sold a bill of goods. When you see yet other studies (and yes, that makes “*them* suspects) claiming that as many as 75% of social psychology studies could not be duplicated (, it’s going to take a lot to convince me of anything any “researchers” claim.

    Anecdotally, my reading habits (as opposed to my book buying habits) haven’t changed very much at all since I switched to ebooks, other than not needing a night light to read before going to bed. Oh, and I’m reading a lot more new/indie authors thanks to my KU subscription (and a lot less of my previous must-read writers thanks to the insane ebook pricing).

  20. Never agree with research published in the WaPo. It only encourages the bastards.

  21. Related data:
    My general impression from nearly 40 years of teaching university students is that if you turn on a video the students immerse in it uncritically, do not think about it any more than they would think about a soap opera, and soon forget what they heard. Many years ago, colleagues of mine at Michigan did a test of using then-current computer graphics to teach how to do a titration. You actually saw the burette, the solution being tested, and the change in the solution as you crossed the key point, either gray to white or red to blue. After getting unexpected results, they tried a number of other items and concluded that, by a large margin, using color in educational computer graphics was a major negative factor.

    A more dramatic related outcome was tests of reading primers for lower grade students, in which it was conclusively demonstrated very many years ago that inserting pictures markedly reduced student learning of reading.

  22. amiegibbons15

    No. Until they do a study, with reproducible results, comparing reading on ebook screens to reading paper, it’s just more unprovable social science.

    • BobtheRegisterredFool

      It might be more correct to say ‘unprovable social speculation’.

      I just performed a survey that indicates a majority of scientists think that social speculators might be fed into a plastic shredder without greatly harming the pursuit of science. Which impresses no one but those whose memories are so bad that they forgot their methodological error bingo cards.

  23. julieapascal

    Did they control for the *size* of the screen? A paper game spread out all over the table is different than a game on an iPad. But my husband has three 27 inch (or 20-something) inch screens in a wrap around display with a flat screen television above. Play the game on that.

  24. julieapascal

    I find it easier to edit on paper, to notice errors. This seems like the opposite of “seeing the big picture” and seeing only what is in front of me. On the screen I have a harder time noticing the stuff that isn’t the story.

    • Aye. One of the best debugging tools is a printout so you can not only things, but mark them up and see how they really relate rather than how you believe they relate. Text/code editors and programming environments have improved, but sometimes you only truly see by really looking.