As posted on Speed-reading apps: can you really read a novel in your lunch hour?
This article contains 1,993 words. If you were to read it to the end, without being distracted by your email or your dog or your children or the contents of the fridge or the bills you have to pay, it would take you, on average, a little over six minutes. But what if you were able to imbibe all of its (undoubted) nuance and richness in half of that time? Or a quarter? What if you could glance at the text and know everything it said just by running your eyes down the page?
The idea of speed reading was invented by an American schoolteacher named Evelyn Wood, whose search for a way to improve the lives of troubled teenagers in Salt Lake County, Utah, by teaching them to read effortlessly, led her to the belief that she herself could read at the rate of 2,700 words a minute, 10 times faster than the average educated reader. And further, that the techniques that allowed her to do so could be taught and sold.
With Doug, her husband, Wood opened her Reading Dynamics institutes across the US and beyond in the 1950s and 60s, and her methods became a self-help craze. The way in which we read, she professed, in the managerial spirit of the moment, was inefficient in terms of time and motion. We had to stop “subvocalising” – “saying” words out loud in our heads as our eyes moved across the page – as well as learning to outlaw the pauses and detours that led to us reread phrases when our minds drifted or our understanding snagged. Print should be consumed in blocks rather than words and sentences. To achieve this, Wood promoted a technique of running a finger down the middle of a page to “activate peripheral vision”. By the end of a course in Reading Dynamics, breathless students were “reading” Orwell’s Animal Farm at the rate of 1,400 words a minute, and telling tales of revolution.
President Kennedy, who believed himself to be a gifted speed reader (and who colleagues observed “reading” the New York Times and the Washington Post each morning in 10 minutes flat, scanning and turning the pages), sent a dozen of his staff to the Evelyn Wood Reading Dynamics Institute in Washington. Presidents Nixon and Carter, under mountains of briefings, followed suit. The science of Wood’s method was never remotely proven, however, and by the time of her death in 1995, her ideas had fallen out of fashion.
Recently, the attractions of speed reading have been revived and promoted, for a couple of reasons. The first is the persuasive perception that we are living in times of information overload, that we are daily presented with more words than we can possibly cope with, and that new tactics are called for to enable us to make sense of it all. The second factor is the belief that since text can now be presented more dynamically on screens we are not “restricted” by the rigidity of printed sentences on a page: surely there is a better way?
These twin perceptions have led to a wave of businesses and apps that once again aim to “revolutionise your reading speed” (at the cost of $4.99, or whatever, a month). For the past couple of weeks I’ve been experimenting with a few of the best known, mostly on my smartphone. The apps generally use a technology called Rapid Serial Visual Presentation (RSVP), in which individual words, or blocks of two or three words, appear one after the other in the centre of your screen. The rate at which they do so can be set to 300 or 500 or 1,000 words a minute, enabling you to feed in text and books to be “read” faster and faster.
Two of the more popular platforms offer a slightly different approach. The Spreeder app allows you to choose the number of words you see at each moment, and to vary the rate at which these words come at you. I found that I could just about take in three-word chunks of Animal Farm for sense at 800wpm, but that in doing so I not only had a slight feeling of panic in trying to keep up, I lost any sense of the rhythm of language, and with it any of the tone of what was being said.
Spritz technology, meanwhile, developed by a company in Boston, is based on the idea that much of the time “wasted” in reading is spent in the fractions of seconds as the eye’s focus moves between words and across the page. Spritz – which drives the app ReadMe! – offers successive individual words in which one letter, just before the midpoint of each word, is highlighted in red, keeping your focus on that precise point on the screen (the “Optimum Recognition Point”). With this technology I found I could just about read simple passages for sense at 700wpm, an ability I imagine would become more natural, if not necessarily more comfortable, the longer you practised it.
Both of the apps – and there are dozens of others to choose from – come with tutorials and exercises to help you “master” the system. In most cases you start, as Evelyn Wood used to, with an assessment of your current (bad) reading habits. It’s the nature of my job as a journalist to often assimilate a lot of information under time pressure, so I like to think – no doubt along with pretty much everyone else – that I have developed quite fast comprehension skills. An app called Acceleread was mildly impressed with my ability to read a passage about deep sea creatures and then answer a series of questions about it.
The assessment began positively enough: “385wpm Fantastic! You already demonstrate some advanced techniques such as reading words in groups rather than individually.” But the assessment had caveats: “You may still find that you often say words silently and get easily distracted.” (You’re not kidding.) “Your program will focus on reducing subvocalisation, strengthening your eye muscles and increasing your capacity to absorb more information at once. You should see rapid and dramatic results…”
Before embarking on this body-building course for my eyes and brain, I read through some of the quite complex science of reading (generally at no more than 200wpm, and with plenty of distractions). There have been many studies of the claims made by speed reading courses, going back to the early promises of Evelyn Wood. As well as arguing that it was possible to utilise peripheral vision, she claimed that our eyes were lazy, unless yoked into rigorous training. The studies – most definitively a large-scale research project, “So Much to Read, So Little Time: How Do We Read, and Can Speed Reading Help?”, led by scientists at the University of California, San Diego and published last year – concluded that in general such training is “neither biologically nor psychologically possible”.
The mechanics of reading have only recently been fully understood. They depend on a brief “fixation” of the focal point of the eye, which lasts about 0.25 of a second on each word. The transition of that focus to the next word is allowed by saccades – fine, ballistic eye movements, which last for about 0.1 of a second. The eye then either keeps moving forward or momentarily and subconsciously flicks back to confirm the sense of what has been read so far. All the experiments suggested that short-circuiting any part of this process led to a loss of comprehension and retention. The genius of normal reading is that it can minutely vary those fractions of seconds depending on how much of the sense of what is being read has been grasped. In a dense sentence, with sub-clauses and unfamiliar language, fixations and saccades are adjusted accordingly, so there is no break in reading flow. In easier passages the eye dances along swiftly. About 30% of the time it automatically shrinks the saccade over a familiar run of words, skipping past those it can predict.
How does this understanding bear on the apps such as Spreeder and Spritz? The acceleration they promise tends to depend on three issues: sub-vocalisation, looping backwards, and the time lag between words. The “So Little Time” study examined each of these in turn. When scientists tried to get people to eliminate sounding words subliminally in their heads – by having them constantly hum while reading, for example – comprehension dropped precipitously. The evidence suggested that when people saw words, they instantaneously accessed the sounds of those words to help understand them. The two processes worked seamlessly; speed dislocated them.
The problem with the second promise is perhaps more obvious – you don’t have to use the apps on fast speed for very long to realise that without the ability to go back and reread a phrase or a sentence, you can quickly lose the thread of what is being said. (Some of the apps have recognised this and added a rewind button.) The issue with the third claim has to do with rhythm. While it is true that you don’t receive any fresh information in the spaces between words, the research suggests that the millisecond pauses are crucial for cognition: they are our brain’s tiny spaces for reflection.
One of the things the studies don’t dwell too much on is the nature of what is being read. I can’t imagine ever wanting to read a novel at more than the normal 300wpm (by comparison, a speaking voice is roughly 150wpm – and even cattle auctioneers can only rattle at 250wpm), but the virtue of reading short articles or emails on RSVP at double that speed seems more plausible. Chances are, however, that most of us already use various intuitive skimming techniques to extract information from such documents when time is short.
You don’t really need studies to prove (though they do) that the more familiar we are with a subject, the more likely we are to be able to extract important information from it at pace. It is for this reason that JFK was able to “read” the New York Times so quickly – presumably he “knew” most of the stories first hand, anyhow, and was just letting his eye flick across headlines and first sentences for a sense of argument. Most of us do something like this with material with which we are familiar – although we are all probably less adept at it than we imagine.
Ronald Carver, a professor of education and psychology at the University of Missouri, proved in a landmark study of “brainiacs” in 1985 that, even for very practised speed readers, attempting to read above 600 words a minute meant that comprehension of any text fell below 75%, and went down dramatically as the reading speed increased beyond that. There is some evidence to show that we can, however, develop the ability to “fillet” a book quite quickly if we use adaptive techniques. In another study of the various techniques of “skimming”, two researchers at the University of Bath showed that skimmers who were most successful at extracting and retaining meaning were able to focus on critical sections of an argument and to jump forward as soon as the “rate at which they are gaining new information drops below a threshold”. They were particularly alive to bullshit or repetition.
Much of the buzz of our so-called digital overload comes from those latter growth industries. It has been argued that the subconscious mind can process 20,000,000 bits of information per second; but of those, the conscious mind holds on to only about 40 bits at any moment. Rather than trying to read more quickly we might be better advised to read more selectively. A lot of our lives can be scanned and scrolled and skipped, but reading remains a more immersive kind of act, dependent on detail. As Woody Allen observed: “I took a course in speed reading and was able to read War and Peace in 20 minutes. It’s about Russia.”