May 18th, 2020

Speed reading might not be so amazing after all

When I was a child, I watched a fair amount of television, and while I don't think I watched much more or less television than the average child, what interested me on television might have been a bit different from what interested most children. I enjoyed cartoons and children's shows, but I also watched a lot of infomercials, because it was interesting to see what they were marketing; I assumed that if people were making and selling these products, that must mean that customers were buying them, and so I saw infomercials as a sort of way of better understanding business success. I remember seeing several episodes of Amazing Discoveries, a series of infomercials hosted by the late Mike Levey (who passed away on August 2, 2003). Looking back, it's hard to believe that anyone actually bothered to make or watch such shows; the 1980s and 1990s were a very different time, a time when you could literally make a half-hour-long commercial and people would watch it. Indeed, Levey claimed that none other than David Letterman once called him "the most-viewed person on television", which is a bold claim to make, but it might have been true considering how widely-syndicated Amazing Discoveries was in the early 1990s. That was a time when the middle class still had enough disposable income that they could spend money on things to improve their personal lives.

One episode of Amazing Discoveries which particularly sticks out in my memory is the one on a product called Alpha-Netics, designed by the late Owen Derrell Skousen (who passed away on February 26, 2010). Thanks to the magic of the Internet, you can watch the full episode here on YouTube, but for those of you who don't want to sit through a 28-and-a-half-minute commercial (and, really, I don't blame you), Alpha-Netics boils down to a speed-reading program. There are many such products available on the market, and their effectiveness is disputed, but Amazing Discoveries was popular largely because of its brilliant showmanship: This episode used several stunts to make the product seem "amazing", the most audacious of these being the scene 10 minutes into the episode where they introduce a lawyer by the name of "Fred Hersh" whom they claim reads faster than the person listed in The Guinness Book of World Records as the world's fastest reader. To test Hersh's reading ability, they have him "read" War and Peace for 15 seconds. (The well-known TV personality John Moschitta, known back then as "the world's fastest talker", inexplicably reads the same segment out loud while Hersh reads, which has absolutely nothing to do with the challenge but is apparently done just to make the segment more entertaining.) During the 15 seconds, Hersh literally just flips through the pages of the book, seemingly not even reading what's on the pages, but at the end, they state that Hersh read 18 pages in 15 seconds, and as the book averages about 500 words per page, this comes out to 9,000 words in 15 seconds, or 36,000 words per minute, higher than the Guinness-listed record, leading Levey to claim that a new world record has just been set on the show. Levey then asks Hersh a question about the contents of the segment he read, which Hersh answers with perfect accuracy. If this scene is to be believed, you can remember the contents of a page of text after looking at it for less than one second.

This point is primed by a scene earlier in the episode (at about 4:43 in the video) in which Skousen shows a photograph to Levey for 3 seconds, then has Levey describe what he saw, which Levey does with reasonable accuracy, as the photograph contains a few discrete objects. Then Skousen does this same test again, but this time, the image is a screen full of text; after 3 seconds, Levey can only recite a few of the words he saw. Skousen makes the point that if you can remember the details of a scene in a photograph after just a few seconds, you should be able to do the same with a page full of text. This is an idea which stuck with me through much of my life; it seemed to make sense to me when I was a child. Why can't you just look at a page of text, and in a matter of moments, know what everything on that page says? Surely it must be possible, right?

I'm not aware of a lot of formal study which has been done on this idea, which is probably part of what allows ideas like this to prevail in the public imagination. The general consensus among scientists who've studied this question, however, is that reading speeds of thousands of words per minute is literally superhuman: not possible for a human being to achieve, due to various limitations in the eyes and brain. If you think back to Levey's first "test" in which he views a photograph of some objects, it's apparent that he missed several details in the scene, and so would most people. Yes, he could identify various specific objects, but it's unlikely that most people would be able to recount, with accuracy, all the objects in the scene; you can recognize a fire extinguisher or a banana in less than a second because you've seen these objects before, but you wouldn't notice all the specific visual details of these objects after viewing them for just a few seconds. The same principle applies to reading text: If you look at a page of text, you'll obviously be able to see that there are letters on the page, but reading is about differentiating letters, being able to notice which letters are on the page, and this is something the eye can't do reliably when letters are in its peripheral vision.

"Speed reading", then, is a controversial subject because most of it seems to consist not of actually learning to read faster, but rather developing a set of heuristic tricks to gather the meaning of a piece of text without going through every word. The practice of "scanning" text, which seems to be most of what speed reading is, is mostly just jumping through the text looking for words which give you an idea of what the text is about. If you see the word "amortization", it's a good bet that the text is about finance; if you see the words "plaintiff" or "defendant", it's likely that the text is about a court trial. By jumping through words like this very quickly, you can get an idea of what text is about, but not what it actually says, and indeed, results of various speed-reading courses have shown that people who go through such courses show very poor comprehension of what they've read: Yes, they've learned to flick their eyes more quickly through the text, but at the cost of understanding the details of what they've "read". The Alpha-Netics program claims that it increases not only reading speed, but also comprehension, which appears to be literally impossible.

A point which is only glancingly made during the show is the link between reading and speaking. Levey mentions that because Moschitta has to read every word on the page out loud, this obviously restricts his reading speed. This ties into a popular theory that reading speed is partly limited by how people are taught to read as children: Usually people are taught to read aloud, and so this creates a strong link between vocalization and visual reading. There is a phenomenon called "subvocalization" which is the tendency of a reading person's mouth and throat to mimic how words on a page would be pronounced, even if they are not reading out loud. People who read to themselves tend to sound out how the words would sound if spoken, and because speech is mechanically limited to how fast it can go, speed-reading proponents claim that if you stop subvocalizing--if you break your mental link between written words and spoken words--you can read many times faster. Studies have generally found that while subvocalization is a real thing which people do, speed gains from trying to eliminate subvocalization are modest at best. You may have noticed that people who are not in the habit of reading often tend to read things aloud, because they are more used to hearing spoken words rather than reading written words, and so hearing their own voice read words out loud actually helps them understand the text better. Practicing reading helps to reduce this need for vocalization and allows people to better absorb written words without needing to subvocalize, but even the very fastest readers still maintain some mental link between how words look and how they sound, and they are still constrained by physical human limitations which suggest that claims like those made by the Alpha-Netics program are more fantasy than reality.

Added to this is the growing awareness that trying to accelerate everything is not a great idea and the corresponding rise of the "Slow Movement", which does include slow reading. Ultimately, what really counts in nearly everything is quality, not quantity, and if you read thousands of books in your life but didn't have the time to enjoy them or benefit from them, that's not a better outcome than having read merely hundreds of books but being able to enjoy them and get some meaning and enlightenment from them. Like many people, I get frustrated by how much information there is in the world and my physical inability to absorb it all, but I'm convinced that just trying to accelerate myself and make everything go faster, faster, faster isn't the right solution.