Slashdot Mirror
Why Speed-Reading Apps Don't Work
sciencehabit writes: "Does reading faster mean reading better? That's what speed-reading apps claim, promising to boost not just the number of words you read per minute, but also how well you understand a text. There's just one problem: The same thing that speeds up reading actually gets in the way of comprehension, according to a new study (abstract). Apps like Spritz or the aptly-named Speed Read are built around the idea that these eye movements, called saccades, are a redundant waste of time. It's more efficient, their designers claim, to present words one at a time in a fixed spot on a screen, discouraging saccades and helping you get through a text more quickly. But that's not what researchers have found."
the first time i read it
When I read a page, I can actually see multiple words in a sentence, context from the line(s) above, and generally can access a context of about 10-15 words at a time. While speed reading (as in, actual speed reading a page), you read by going down the center of the page, which preserves a good chunk of the context, and assumes that missing a few words here and there is only going to minorly impair your understanding of the text.
This, on the other hand, provides a minor speed-up at the cost of context, the ability to back-track and no ability to skip words that don't help much with understanding like various particles or flowery prose.
Yep, this approach is idiotic.
Those who can, do. Those who can't, sue.
When I was in high school, they had machines that presented text one line at a time at a set speed. The idea is that we were to gradually speed it up to force us to read faster. There would be a brief comprehension test after which was more of a short term retention test.
Then there was reading texts normally (free reading) and seeing how fast you were reading without the machine.
The result of that was that I could read much faster than natural if I pressed it and my natural speed improved by about 10%. I find that I rarely care to press it as it gives me little time to think about what I'm reading and so poor long term retention so it's good mostly when I need to do more than skim but just need to find a bit of information for immediate use.
Of course, the far more useful 10% increase in natural rate doesn't make as big a bullet point.
Though the article does note that this is the case for a lot of people, but the big advantage of reading over other media (e.g. audio or video) is that reading is self-pacing. When reading information rich texts, it allows me to gloss over details that I already know while focusing upon details that I don't. When I'm in a lousy state of mind (e.g. having difficulty concentrating due to lack of sleep or external concerns) it allows me to slow down. When I'm in a good state of mind (e.g. I'm motivated to read the text or am well rested) it allows me to speed up.
Simply put: I read rather than watch or listen because my mind is in control of the flow of information.
Perhaps someone can pick apart a speed reading idea I've had for awhile and tell me what's wrong with it:
I envision a system where you have a physical book and an audio book. You would read the book while listening to the audio book. Slowly, you would increase the speaking rate of the audio book and work to match your reading speed. Double re-enforcement. Ultimately you would no longer need an accompanying audio book.
This is effectively how we learn to read as toddlers.
Brought to you by Carl's Junior.
The reason speed reading apps don't work is because you either know how to read fast or you don't. The average human should be able to read well over 200 - 1000 words a minute, any less and you have much bigger problems, more then an app can solve. This should be the chart for reading speed:
1. Fast: 1000+ words / minute
2. Normal 200 - 1000 words / minute
3. Slow 100 - 200 words / minutes
4. Unacceptably slow less then 100 words / minute.
People who read at less then 100 words per minute have a completely different problem that can't be solved from a simple app on a phone.
1000 words per minute? Next time you pull numbers out of your ass make sure they're not a joke. No one reads anything of value at that speed. You can scan text quickly but you won't actually be reading it. Anyone claiming otherwise is likely trying to sell you a speed reading course, or stroking their ePeen on Slashdot (see half of the replies I'm going to get from alleged super speed readers).
tl;dr
Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
It was shown quite some time ago that adults "read" by recognizing the shapes of words, not their spelling. If this is true then it would explain the problems described in TFA.
http://en.wikipedia.org/wiki/W...
http://www.dummies.com/how-to/...
16 words a second? That's three times faster than a 300 baud modem.
I should use this sig to advertise my book ISBN-13 : 978-1501515132.
What if you're reading AND juggling?
I'm fully capable of reading at 1000+ words / minute and remembering the information, so next time you want to claim it can't be done, make sure you're not talking to someone who can. I was able to read 300 words / minute in grade 6 and retain information over a 90% level. I was fair and put the average at 200 words / minute, which is still pretty slow, you have to be crawling through a book at that pace.
HAHAHA well I can't juggle so I don't know, but if I could I'd figure that would slow me personally down to less then 100 w/min.
As someone who discovered Spritz when it started making headlines, and tried out a similar RSVP app with the novel I'm currently reading, I can tell you that my comprehension didn't suffer. I tend to adjust the speed while I read, ranging from 500-700 wpm, and I can still clearly recall and describe the plot and detailed events of the book over the sections that I read using the app.
I do agree that it's not an ideal way to read, as the flow of text tends to be robotic and lacks some of the conveyance of emotion. In this regard, it's probably better for reading non-fiction or purely informational texts. I can't say it hurt my comprehension, though.
Unlike porn, which yada yada rimshot hey-ooh!
Reading speed also depends greatly on the complexity of the text being read. I've taught grade five students, given them new selections and timed them, testing comprehension afterwards. The fastest I've seen (by a HUGE margin) was around 1400 words per minute with over 80% comprehension. That exceptional student could read and comprehend college level novels, so grade 5 was a joke, but he couldn't read the college level stuff at more than 250 words per minute.
Can you read the last novel you read at over 1000 words per minute? Not unless you are a hell of a lot faster than I can if you're also understanding it. (I top out around 750 wpm for the grade-9 level stuff you tend to see day to day.) Can you read kindergarten level text that quickly? Almost certainly, yes.
The original poster may have seen those numbers in a seemingly reliable source. Without grade levels involved, the numbers are meaningless. That doesn't mean the poster pulled them "out of his/her ass," but it does mean context is lacking.
- W. Blaine Dowler
http://www.bureau42.com
I was fortunate to be given a speed reading class that took several weeks as a high school graduation present back in the dark ages (1965). I took the course and practiced as I was told, but I never did believe it was doing any good and it was definitely not a fun way to read anything for pleasure. Fast forward to the next year and I am a freshman in engineering, but having to take an "elective" political science course in which I had no interest at all. Since the purpose of the freshman year in engineering is to cull the masses I concentrated on what, to me, were the far more important classes I had. A few days before the final exam I realized that I was not going to pass that polysci class unless I could pull off a near perfect score on the final. The professor was kind enough to confirm my calculations on that point. For two days before the final I read the entire text book through cover to cover TWICE. I scored in the high 90's on the final and passed the course with a C+. When I was taking the exam, I really was just sort of zoned out. Much of the test was essay questions and I would just write whatever came into my head on the subject, not really knowing what I was saying or where it came from.
Now, I would have put this down as a fluke except that I was able to do very similar things for the rest of my career in engineering. Although trained in systems engineering I started out working with computer systems when computers were big iron and I worked on both IBM, Univac, and DEC systems. Then I successfully made the transition to PC's and networks and retired as a Network Engineer and Security Officer. Often I would have to learn enough to get started in a new area about which I knew almost nothing with little time to do so. I would get 5 or 6 books on the subject and absorb them over a weekend and could then get up to speed pretty quickly after that.
I still don't understand how it works, and I am still not sure I really believe it works, but for over 40 years the speed reading class I took in 1965 saved my bacon many times.
I'm fully capable of reading at 1000+ words / minute and remembering the information, so next time you want to claim it can't be done, make sure you're not talking to someone who can.
If you're really a person who can do this, PLEASE volunteer for one of these studies, so we actually have some reliable evidence. Because basically every previous study on this stuff says comprehension goes significantly down as reading speed goes up.
Quite a few studies have shown this (some articles summarizing findings here, here, here, here), and the only ones that seem to ever disagree are those designed by the speed-reading course or software people. Even for professional high-volume readers and people who performed well in generic speed-reading tests showed a maximum of about 75% comprehension at 600 words/minute.
And there are loads of cognitive science studies that demonstrate why this must be so. Lots of research on eye movements during reading and the maximum possible speed they can take things in, the way our retinas work and focus, cognitive constraints on the extent and speed of our "working memory," etc.
By the time you get to your claimed speed of 1000 words/minute, I sincerely doubt you're getting anywhere close to 50% comprehension. Therefore, what you're doing is skimming, not reading.
There's nothing wrong with skimming. it's an incredibly useful skill which I really picked up in graduate school. I have used it all the time when teaching and (when preparing for class) needing to re-read an article I haven't looked at in a long time (and don't really remember) or a new article dealing with a subject I'm already familiar with. I can certainly skim at 1000 words/minute and be prepared to discuss a lot of important points of an article, but if a student queries me on something very specific, I guarantee that we'll have to slow down, I'll go back, and take a look at that specific passage. When you're already fairly familiar with the field or kind of material, you can often zoom on essential elements fairly quickly, and your comprehension rate gets higher -- but you're still not reading. And if you were reading something outside of your discipline, the comprehension would go WAY down at such speeds.
So -- sorry, but your claims to read at that speed and retain information have been debunked by many studies along with many other supporting cognitive science studies that basically show why it can't work.
If indeed you are some person with a freakish skill that you can demonstrate under controlled conditions, please volunteer for a study. Otherwise, I (and any other reasonable person here) should assume that what you're actually doing at 1000 words/minute is skimming, and you're probably only getting a small fraction of the total information.
I know exactly what you're referencing, it's the same studies I've had quoted to me by my teachers, parents and family. I'm not saying I'm better then anyone else but I've never understand the purpose of reading and comprehending in one step. The only purpose to "read" is to move the words from the page to my memory, once I finish reading what ever I want I just process the information. You can easily break 500 words / min reading in this fashion, your brain can process so much faster then your eyes. I've already been in a study for speed reading and proven it's not a joke.
The article is exactly right from my experience, and I'm not going to speed read a book I enjoy. However, there's a lot of times where you need to pick up an idea quickly or in it's general form where that comes in handy. Great example is classroom material. I'm not reading four or more 600+ page textbooks each semester, especially when I don't need 98% of the material.
What I do need most often is a general grasp of what is going on in a particular chapter, then I might go back, work out how to use formulas, go over a specific table or case study, etc.
Speed reading the entire thing leaves out almost all of the trees, but it allows me to get a quick view of the forest.
And let me guess, you probably also learned to read by recognition and not phonics, right?
Your experience seems similar to mine, and that's what I have found the key difference is. I learned to read from being read to every night, and following along. When I was 2 I was reading the books back, and my family just assumed I had memorized them. Then somewhere around turning 3 they realized that I could already read - it started with picking out a sign or text and then finally someone shoved a newspaper in front of me, and while I couldn't read the entire thing I was able to read quite a bit. I was reading "chapter books" by 5. And I've been reading - a lot - ever since then, even if the format has changed (obviously a lot more online/tablet than print these days).
I think the folks saying "you are SKIMMING" may be on to something, but not what they think. Not the quality of reading, but the actual method. Looking at it (pun intended), I would assume that my "normal" reading is what some folks would refer to as "skimming". I don't sit and read every single word, I look at groups of words and see them as one. Now, I still go over all the text - skimming for me is when I'm only looking at a few sentences a paragraph.
I really started to understand this fundamental difference in how folks read when I got out into the professional workplace. Lots of very smart, clever folks (and many not so smart and clever), have very little ability to read anything longer than a few sentences and understand it. Everything has to be bullet points and brief, summary headers - this post would never work there, LOL. And a lot of these folks are a lot more traditionally educated than I am. Of course, this is a separate issue and has many factors, but just watching folks read something that is sent out and it takes them four times as long to read it and I still get more out of it than they do was actually the first time I realized that they simply read differently than I did.
The way I taught myself to speed read is to skim rapidly across a paper and get the gist of what they're saying. I read in major word concepts and understand what they're trying to say while skipping lesser words. It is dead on that you can't comprehend as much, but you can give yourself a personal TL:DR summary. I don't recommend others to learn speed reading because sometimes it engages itself without trying. It is almost a bad habit that it engages when I have impatience with what I'm reading. I mean there are places it is good to have speed reading, but sometimes its hard to control and can be a bad habit. I mean the pros out weigh the cons, but not by much.
Speed reading isn't as useful of a skill as people think it is when they first hear it. It even ruins things designed to capture your imagination at a certain rate of time like Tolkien. If you speed read Tolkien, it is comparable to watching LOTR in fast forward. I mean you can get the gist of what happens by watching a movie in fast forward, but if you want to sit down and enjoy it, you watch it at the rate it was intended for. Where speed reading is really good is focusing a page of all sorts of information like Reddit, and getting to the stuff you want to read more rapidly.
God spoke to me
Great. Again, take part in a study, please. It's all great to "self-diagnose" or test yourself with some online tool that's probably designed to sell you something. But the actual studies done by cognitive scientists and reading experts don't seem to find the speed reading claims with comprehension to be real.
Look -- I read by word groups too: anyone who is fluent in a language and reads A LOT probably does. There probably are some differences in normal reading speed for maximum comprehension (and maybe some of it has to do with the way people are taught or bad habits or whatever) -- the figures I've seen say it might vary between 200 and 400 wpm or so. But 1000 wpm for full comprehension? Nope. Not in any reputable study. And, more relevant to the present discussion, multiple studies have shown that people who train themselves using speedreading methods (like deliberately trying to avoid subvocalization or reading by groupings or whatever) don't perform any better than untrained readers who are asked to skim in comprehension tests (and sometimes trained speedreaders perform worse). There may be some gain in superficial understanding for speedreading techniques, but no demonstrated advantage over normal people skimming in terms of deep comprehension.
In sum, I have no doubt you may read faster than your coworkers on average, and your comprehension may be better. But I have serious doubts if you claim full comprehension at speeds over 500 wpm.
I've already been in a study for speed reading and proven it's not a joke.
Fantastic! I've provided some citations for my claims. Where's your study published? I'm absolutely serious, and I do NOT take this as a joke. I find this area of interest, and if you have something reputable that contradicts what I said, I want to know about it.
Oh, and by the way, the whole thing about the brain going faster than the eyes -- that's probably true for most folks, particularly non-expert readers. But there are limits to cognitive processing speeds, and most experts in visual/language processing see this as the ultimate "speed limit." It may vary from person to person, but I get the sense that most experts are pretty skeptical of anyone claiming full comprehension above 400-500 wpm. Again, any citations that dispute this would be welcome.
I can read the entire slashdot page of comments for this article in 10 seconds. Why? Because I know all I'm skipping is uninformed BS. It seems no-one here has the faintest clue to the reading process. I have first hand experience in psycho-linguistic experiments, and I can tell you that saccades are quite probably an essential part of the reading process: eyes just don't wander around at random, they fixate quite precisely to recognize groups of characters or short words in their context. Very short words, like articles and prepositions can sometimes be skipped altogether (leading to a certain class of reading errors). And in experiments where you show subjects one word at a time, it's obvious that information gets lost when you present the words too quickly.
It might be possible to speed up reading a bit, but the faster, the more it resembles skimming. And when you try to force that onto the reader by presenting every bloody word (something skimming avoids), the reader is going to get tired from the effort, and confused since he's got no way to go back. Very lousy idea.
Many years ago I got pulled into a scam around speed reading. A good friend (at the time) was in the thrall of a conman with an interesting proposal. The elevator pitch went something like this: "Imagine a system that flashes words at you subliminally and when it detects you haven't understood a word (via a biofeedback mechanism) it then flashes the dictionary definition of that word. You could read an entire book in minutes and have complete understanding of the content."
Even though I was young I still could smell bullshit. A small group of similarly-minded people tried to pop the bubble, but when the true believers had invested so much time and emotional energy there was no turning them around. The was more to it than this: crazy mind games, a three-car pile up and other weirdness (including an impromptu cover of "The Rainbow Connection" in an upmarket restaurant), but I won't bore you with the details. The end-point is it soured a friendship which never recovered.
Maybe I'm biased by that experience, but any technology that promises to solve problems by getting people to read faster - instead of, say, with better comprehension - leaves me with the taste of snake-oil in my mouth.
Something about speed reading being great? Right?
systemd is Roko's Basilisk.
I am German, so please excuse me, if my English isn't that good. There once was a study in Germany, what effect almost exclusively lower case in german texts would have. The result was, that texts are better understandable with nouns written with an upper case first letter. They serve as anchors, especially, when reading fast. Furthermore, in English it is common, that the same word is used as a verb and a noun. This actually would be a damn good reason to write nouns with an upper case first letter. When you skim through an English text, and notice, for example, the word "jump", you don't even know, if it is the verb or the noun. This can also be a source of ambiguity. Another big source of ambiguity is the usage of "it" to refer to almost everything, because most words don't have a grammatical gender. You have to be aware of that, when you are creating relative sentences in English. Even people are not always gendered in English. Is the tennis player female or male? You often have to read a fair bit into a text, until he/she (see?) is referenced as "her" or "him". I have seen some discussions between native English speakers on the internet, were someone wrote sth. like "Ah, so the teacher was your mother!" or "I thought, the cat was chasing the car, not the dog." Oh, this last example has another ambiguity. I should write "I thought, the cat, not the dog, was chasing the car". This ambiguity does not occur in German because articles are declined dependent on the gender and relation. http://de.wiktionary.org/wiki/... To put it in a nutshell, it seems logical, that speed reading is especially hard in English. The more possible ambiguities you have, the more likely you will have to stop and think about, what is really meant.
Like he covered everything else, RIP
https://www.youtube.com/watch?v=rbKFgIxv5nU#t=02m31s
I read from memorization, I only have to look over a page and I can break it down later when I need the information. I don't phonically go through and read the words in my head, I let that happen later. The only reason I even look at the page in the first place is to capture it's information. I'm not skimming, I'm actually "loading" the information into my head, I just appear to be skimming. I've read this way for years, starting in about grade 3 when I noticed that there was no reason to sit down and pronounce word by word when I could look at a page a remember it.
I'm going to find the study if I can, I've been in several, the first was in grade 5, the second was in grade 9 and the last one was in university. If I can find the link to the study I'll share it with everyone, I have nothing to hide or worry about, hell I'll even redo the study if someone comp's my time for it.
As I've said in another post, you only have look at a page and remember it, there is no point to read phonically, it's much more efficient and practical to read from images. Most people read by going through a paragraph and reading it to themselves, I read by looking at the page and taking a snapshot of it in my head, it's hundreds of times more efficient as I don't have to process the information phonically at run time, I can "compile" it and store it for later.
If you're talking about reading a page phonically then yes you will run up to the 500 wpm limit, this is a major draw back to not reading using snapshots. I'm at work right now but when I have more time I'll search for the information I'm referencing.
I'm going to find the study if I can, I've been in several, the first was in grade 5, the second was in grade 9 and the last one was in university. If I can find the link to the study I'll share it with everyone
I look forward to reading it.
As I've said in another post, you only have look at a page and remember it, there is no point to read phonically, it's much more efficient and practical to read from images. Most people read by going through a paragraph and reading it to themselves, I read by looking at the page and taking a snapshot of it in my head, it's hundreds of times more efficient as I don't have to process the information phonically at run time, I can "compile" it and store it for later.
Yeah, now you're talking about a completely different phenomenon. You're no longer claiming to be a speed reader. Instead, you're claiming to have an eidetic memory. Those aren't the same thing. A speed reader would process the information as a normal reader would, though at a faster rate, and would be able to answer questions about the reading with the retention capabilities of a normal person.
But you're claiming to have an eidetic memory -- if so, you should not only be able to "read" quickly, but be able to actually quote the page you "read" back word-for-word without errors. If you can't do that, you haven't really "loaded the page image" into your memory for later processing as you claim.
In either case, I'm now even more skeptical than before. Lots of people claim to have eidetic memories too, but basically all studies and attempts to get someone to do this under controlled conditions tend to fail. The people who claim they can do it turn out not to really have a "photographic memory," but rather are very efficient at information processing (as in speed reading, or at least some version of it) -- because when such people are asked to recall a text or an image with nonsensical patterns, they simply can't.
Until you quote a study, I'm done with this discussion. As far as I'm concerned, I appear to have wasted my time having a discussion with some random troll on the internet... claiming to have superpowers. Cheers!
... still write better than you?
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
If you seriously are calling out than vs then, well you're just a duck.
Nobody in the world has ever demonstrated such a capability under scientifically controlled conditions. All the evidence gathered to date suggests that you are lying. And that's a bit sad.
Agreed. There's only been one scientifically controlled study that suggested anything remotely like an eidetic memory, and the subject ("Elizabeth") later married the scientist running the study. Then they refused to participate in any further experiments. So, in effect, nobody in the world has ever successfully demonstrated either an eidetic memory, or (much less) an ability to read significantly faster than average while maintaining decent comprehension. Any anecdotal arguments (for example, Murdoch5's frankly laughable claims) otherwise are almost certainly pure Munchausen-esque nonsense and should be disregarded with prejudice.
Actually you're completely wrong on that, the data suggests that no one can demonstrate long term photographic memory, which is true. If I read a book today, I can probably recall the content for about a week or so, after that it will start to slip and certainly by a mouth or two out I'll retain very little of the information I read over. However that also doesn't matter, usually is something is important enough to remember longer then a week out, you'll have to read it more then once.
Actually you're completely wrong on that, the data suggests that no one can demonstrate long term photographic memory, which is true. If I read a book today, I can probably recall the content for about a week or so, after that it will start to slip and certainly by a mouth or two out I'll retain very little of the information I read over. However that also doesn't matter, usually is something is important enough to remember longer then a week out, you'll have to read it more then once.
Actually you are completely wrong about that. The data suggests that no one can demonstrate even SHORT term photographic memory.
You stated that you can "look over a page" and "load" all the information into your memory before comprehending it. This is an utterly ludicrous claim. You also stated that you can read at over 1000 WPM with very high comprehension. Also a laughable claim that is not supported by any scientific studies.