Some reading apps flash one word at a time in the middle of the screen and promise you can read two or three times faster. When researchers tested apps like this at 700 and 1,000 words a minute, people understood less than when they read ordinary text (Acklin and Papesh, 2017).
Your eyes don’t waste much time moving
The pitch behind these apps is that moving your eyes wastes most of your reading time. A major review of reading research disagrees. Each jump of the eyes takes about 25 thousandths of a second and each pause about 250, so your eyes are moving only around 10% of the time. Cutting that out can’t save much (Rayner and others, 2016). That 10% is the authors’ rough estimate, not a measured result.
Looking back helps you understand
Skilled readers look back at earlier words about 10% to 15% of the time. In one experiment, words were hidden once the eyes had passed them, like an app does. When readers tried to look back and the words were gone, their answers dropped to about chance (Schotter, Tran and Rayner, 2014, as described in the 2016 review). Looking back isn’t a bad habit. It is how you catch what you missed.
Even at normal speed, details slipped
In a study of the Spritz app at about 200 words a minute, a normal pace, people did as well on big picture questions as on paper, but worse on questions about details, and their eyes got more tired (described in Rayner and others, 2016). A later study of 172 people found understanding fell once the app went faster than normal reading (Ricciardi and Di Nocera, 2017).
The inner voice is part of reading
Speed reading courses often tell you to stop “saying words in your head”. Older studies found that blocking that inner voice didn’t hurt easy text but did hurt harder text, and repeating words aloud while reading hurt understanding (reviewed in Rayner and others, 2016). It seems to help you hold a sentence together.
So what does work?
When time is short, skimming is honest. Skimming a whole text found more of the main ideas than carefully reading only the first half, but it missed details and the points you have to work out (Duggan and Payne, 2009).
For real speed with understanding, the review found no training that doubles it. In one course, speed rose from about 280 to about 400 words a minute while correct answers fell from 81% to 74%. What helps is slow and ordinary: a bigger vocabulary and more reading.
Why our passages end with questions
The review gives a neat warning. One common speed reading score multiplies speed by the share of answers right. Someone “reading” at 5,000 words a minute and guessing at chance, 25%, would still score 1,250. That is why every nnLabs passage ends with three questions, and a run only counts when all three are right.
Questions people ask
Does speed reading work? Not without losing understanding. Skimming runs at 2 to 4 times normal speed but gets mainly the gist.
Is saying words in my head bad? No. It seems to support understanding, especially for harder text.
See your own speed, with the check
Read one short passage, then answer three questions. No sign up, and nothing about you is saved. Try one passage
Sources
- Acklin, D. and Papesh, M.H. (2017). Modern speed reading apps do not foster reading comprehension. American Journal of Psychology 130(2). doi.org/10.5406/amerjpsyc.130.2.0183
- Rayner, K., Schotter, E.R., Masson, M.E.J., Potter, M.C. and Treiman, R. (2016). So much to read, so little time. Psychological Science in the Public Interest 17(1). doi.org/10.1177/1529100615623267
- Schotter, E.R., Tran, R. and Rayner, K. (2014). Don’t believe what you read (only once). Psychological Science 25(6). doi.org/10.1177/0956797614531148
- Ricciardi, O. and Di Nocera, F. (2017). Computers in Human Behavior 69. doi.org/10.1016/j.chb.2016.12.047
- Duggan, G.B. and Payne, S.J. (2009). Text skimming. Journal of Experimental Psychology: Applied 15(3). doi.org/10.1037/a0016995