Although electronic texts have been with us for many decades, in the past few years electronic reading has become increasingly popular. The ready availability of mobile, connected devices like smartphones and tablets, along with dedicated ereaders like the Kindle and Nook, have moved electronic reading out from behind a desk into the environment. This change has brought increasing attention to the differences between reading in print and reading via digital devices.
In a recent article in Scientific American, Ferris Jabr argues that “paper still has advantages over screens as a reading medium,” claiming that “most studies published since the early 1990s” support this claim (The Jabr article is behind a paywall, but this piece on Salon a few months ago covers much of the same ground, with some slight differences).
If true, this claim deserves serious attention. Companies like Apple see the education market as an important growth area, and textbook publishers are pushing electronic books—either rentals or individual books that can’t be resold—as a way to eliminate the used textbook marketand increase sales. When powerful companies such as these have a compelling business interest in promoting certain educational technologies, it is important for academics to question if these technologies serve students and learning, and, if reading on paper truly is superior to electronic reading, educators, parents, and students should be concerned about this trend.
Enter Jabr’s article. He argues that paper is superior to screens for four reasons:
Together laboratory experiments, polls and consumer reports indicate that digital devices prevent people from efficiently navigating long texts, which may subtly inhibit reading comprehension. Compared with paper, screens may also drain more of our mental resources while we are reading and make it a little harder to remember what we read when we are done. Whether they realize it or not, people often approach computers and tablets with a state of mind less conducive to learning than the one they bring to paper. And e-readers fail to re-create certain tactile experiences of reading on paper, the absence of which some find unsettling.
These four claims underlie the argument of the article: as the title puts it, “the brain prefers paper.” Indeed, these claims would suggest that any serious reading of any kind should make a point to avoid digital texts.
However, on closer reading there are a number of problems with the evidence Jabr uses to support his claims. Throughout the article, Jabr draws broad conclusions from very limited studies and he frequently fails to acknowledge that there are many different kinds of reading technologies for both paper and screens, and these technologies produce significantly different effects. In this post, I’m going to address his first claim: digital reading inhibits the navigation and comprehension of long texts.
Jabr argues that two features of “digital devices” make it difficult for readers to “navigate a text.” Where books have fixed pages and volumes—their width, height, and depth—electronic texts do not. This feature of a printed book, Jabr argues, allows readers to navigate them more efficiently. Readers tend to fashion what they read into “a kind of physical landscape,” a “mental representation” of the text that, when combined with the fixity of print books helps readers to find what they read. For instance, if I recall that a particular passage appeared in the upper half of the left page of a book and that it came before the book’s midpoint, and this can help me find this passage more effectively.
As anyone who has read a Kindle book or other electronic book format such as EPUB knows, these books do not have fixed pages: rather, they present texts as a continuous scroll, and on subsequent readings, the same portion of text may appear in a different position on the screen, depending on features such as how the reader navigated to the passage or the size of the text.
Of course, digital texts of all kinds represent the location of a reader in the text, generally as scroll bars, and they enable readers to find individual passages by searching for single words or phrases. Although Jabr acknowledges this, he argues that these scroll bars make it “difficult to see any one passage in the context of the entire text.” He expands on this point, arguing that in electronic reading
glancing at a progress bar gives a far more vague sense of place than feeling the weight of read and unread pages. And although e-readers and tablets replicate pagination, the displayed pages are ephemeral. Once read, those pages vanish.
It is worth pointing out here that Jabr is not comparing “paper” to “screens,” the supposed point of the article. Rather, he is comparing two particular technologies: the bound book, what book scholars call a codex, with electronic texts like Kindle books or Web pages that scroll text. Here we have an example of Jabr attempting to compare a particular print technology—the book—with a particular form of digital text—scrolling text—then using these two to represent all forms of “paper” and “screen” reading.
Consider how Jabr’s arguments fail when addressing other forms of print technology. The ability to locate a selection of text spatially in a two-page layout is irrelevant for pages that aren’t bound. Further, the fixity of pages isn’t an exclusive property of print. One of the most popular formats of digital texts, Adobe’s PDF format, does fix the content of pages, affording readers the ability to find text just as they would in a book, by its location on a page. Finally, the need to find a selection of text via spatial memory seems unimportant when reading digital text, as one can simply search for the selection.
Jabr addresses this last point by suggesting that the issue at stake isn’t simply the retrieval of information, but comprehension. As you recall, he argues that “digital devices prevent people from efficiently navigating long texts” and this “may subtly inhibit reading comprehension.” However, the article only cites two studies in which participants showed worse comprehension after reading digital texts compared with those who read print texts, and it is difficult to see how either study supports Jabr’s conclusions. Neither study compares the reading a book to digital scrolling text (in fact, Jabr cites a third study, in which participants reading scrolling text performed equally well on comprehension as those in other groups), and, in each case, the participants read relatively short texts—not long texts—with the exact same spatial layout in print and on screen, a result that suggests the differences in tested comprehension have nothing to do with the fixity of the texts or their physical volume and says nothing about comprehension related to reading long texts, print or otherwise.
In one study of Norwegian 10th graders, participants were asked to answer a series of questions on a computer after reading either a PDF or a printout of a four page document. Both groups were able to consult the document while answering questions, but the students in the PDF group were unable to search their digital texts for answers, and when answering the questions had to switch between the PDF and quiz windows on their computer screens. Students with paper handouts, in contrast, were able to access their handouts while answering the quiz on the computer, glancing between the two. Here, the inability to search the text limited one of the main navigational features of digital texts, and the researchers themselves suggest that having to use the same screen to scan the text and answer questions may have impaired the PDF group.
In the second study, from 2005, university students were given comprehension tests after reading either a printed document or a PDF on a low-resolution (800×600) monitor. In this study, the PDF group scored lower on reading comprehension while also reporting greater stress and tiredness. As with the previous study, this study did not test for reading comprehension of spatially fixed paper texts versus scrolling digital text or for comprehension of long texts. It is likely, the low resolution monitors—which were likely unable to show an entire page of the reading at a time—may have interfered with the students comprehension. Indeed, participants in the paper group in that study reported greater difficulties in concentration and time pressure, while the PDF group reported problems with reading and following the text itself, which supports this conclusion.
In short, the studies Jabr cites simply do not back up his claims about the nature of paper versus screen reading for comprehension, and his other claims in the article similarly suffer. He attempts to conflate multiple reading technologies—paper and screen—with each other to build an argument that is easily challenged by considering other reading technologies and that his own data does not support. This is unfortunate, as the question of screen reading versus print reading is an important one, and he fails to approach it with the nuance it deserves. In my next post, I will take a closer look at these studies and suggest a more complicated picture of the virtues and drawbacks of both print and screen reading.
Banner image credit: sparkle glowplug http://www.flickr.com/photos/sparkleglowplug/6977328362/