When you read, do you read one word at a time, or do you actually read two and three words at the same time?

When I was in middle school and high school, I would take classes at the local community college during the summer (mostly because my mom didn’t want me around the house, I think). In about 7th grade, I took a speed reading course. We were tested on our reading speed and comprehension at the beginning of the course and then again at the end. Our grade was a combination of absolute speed (and comprehension) and improvement at the end of the course. I had a B at the beginning of the course. Because I spent so much of my time reading (while hiding from my mom), I found that I had instinctively developed many of the techniques that were taught during the class. While I do not do the looping “scan pattern” that was taught, I do drop (or infer) most of the supporting words in sentences. Novels usually take me 2 to 3 hours to read, depending on the complexity of the plot (lots and lots of characters or subplots require that I slow down and pay more attention), non fiction works take longer since I must focus on new ideas and vocabulary. An unfortunate result of the way that I read is that I do not sound out new words to myself, so I have larger written vocabulary that I do a verbal one – I’ve never figured out how to pronounce many words.

(By the way, the adult level biology course when I was 15 and the film that the instructor showed of natural childbirth was the cementing of my opinion that I would never have children. There were some things that I was just not prepared for at that age.)

Yes. Here’s a paper about it. It’s part of a strategy to read that involves jumping around, identifying key sections and then going back to read more closely.

[edit]
Others think that reading in groups is implausible, although they don’t offer any evidence to support their theory. In an article entitled Encoding multiple words simultaneously in reading is implausible, Erik D. Reichle, Simon P. Liversedge, Alexander Pollatsek and Keith Rayner say:

By processing words in their order of appearance on the page, readers can incrementally access the meaning of each word to construct a sentence representation [30]. (This does not mean that every word has to be fixated in order; only that attention is allocated sequentially [26].) By doing this, serial-attention models acquire any information conveyed by word order (e.g. syntax) without additional assumptions [31]. By contrast, attention-gradient models require additional assumptions and/or predict that words will be identified out of order. For example, in situations involving a rare long word followed by a common short word (e.g. ‘rhinoceros and’), the words should occasionally be identified out of order. (If this prediction is not true, then parallel processing takes on a ghostly quality, having few testable consequences and – among other things – providing no basis for predicting parafoveal-on-foveal effects [16], [17] and [18].) The question then becomes: how would attention-gradient models handle such occurrences? One possibility is that a buffer maintains word meanings, and that some mechanism re-orders out-of-order words. One problem with this solution is how such mistakes are detected without using comprehension difficulty to ‘signal’ such occurrences. This solution also delays post-lexical processing, preventing immediate detection of whether a word is a reasonable continuation of a sentence. It is thus not clear how the attention-gradient hypothesis could possibly explain higher-level language processing and how it influences the immediate eye-movement behaviors observed in reading [32]. We also suspect that recent work using E-Z Reader (Box 1) to examine these issues will prove instructive in revealing challenges that attention-gradient models have to surmount. For example, in E-Z Reader [33], it is assumed that the linguistic information used to constrain the identity of a ‘predictable’ word only becomes available after the preceding words in the sentence have been identified and linguistically processed. In attention-gradient models, linguistic information somehow helps constrain the identity of all of the words in the attention gradient so that the processing of wordn is facilitated by its preceding context even though that context (e.g. wordn-1) has not necessarily been processed.

@syz The funny thing is, probably the part that “traumatized” you the most was the actual delivery…and that’s the EASIEST part about it all! May I ask how old you are, BTW?

@wundayatta Thanks. I’ll go look at it. I wonder if some people just pick it up naturally, like I think I did, while others have to be taught…or is it simply a a natural outcome of reading ALL THE TIME!!

@wundayatta Cool! So I’m NOT imagining things! Can I post this in that other Q about reading out loud to improve comprehension?

@Val123 Ancient. I’m 46.

Based on wundayatta’s link, I am the second type of reader. I don’t do the loops, but I do process groups of words. And I don’t sound out or perceive the words, I just comprehend them. Because of that, I don’t tend to notice things like patterns or rhymes (or what words sound like). If I want to read poetry (which I rarely do), I have to read each word and sound them out in my head. Maybe that’s why I’m not a fan of poetry.

So youse not likely to change your mind, huh! BTW, maybe we get the prize for getting Off The Subject In Record Time!:)

It’s knd of lke th Missing letter effect.

@erichw1504 Thank you!

@syz Among things that drive me nuts (and it happens once in a while.) There will be 3 or 4 sentences, one after the other that are worded in such a way that it just so happens that the first word on each line is the same word. So I’ll see:
When blah blah blah blah blah blah blah blah blah blahblah blah blah blah blah.
when blah blah blah blah blah. Blah blah blah blah blah blahblah blah blah blah blah
when blah blah blah blah blah. Blah blah blah blah blah blah blah blah blah blah blah when blah blah blah blah blah blah. blah blah blah blah blah blah

Stops me dead in my tracks every time! Also, if I notice that the words on a page are arranged in such a way that you have a big diagonal blank space running through the middle of the page….that’ll stop me for a second. But I think that may be from reading too many Hardy Boys and Nancy Drew mysteries where they have to find hidden meanings in some written thing!

You must have a very visually oriented brain. I too will stumble occasionally – it throws off my pattern and I have to go back and try again. But the patterns on the page don’t affect me as much as they do you.

@syz It’s really the voices in my head that see the patterns, not me! :)

@Val123 @syz When patterns like that happen to me while reading I literally have to read the sentences like 15 times to figure wtf just happened and where I am. It drives me nuts.

I read both ways I think. If I’m reading a book, I read every word, but I will have a word from another sentence jump at me sometimes. But sometimes, if I’m reading something I don’t care to read, I’ll just skim and kind of hit the high points.

It’s called chunking, and I do.

I have to pronounce each syllable.

I just look at the pictures.

Depends on the book, sentences, paragraphs, or quick parts of paragraph’s giving me the
the page.
If it’s poetry or something someone wrote to me personally, I read every word, a large portion of technical also.

I usually scan-read, and if it doesn’t seem to be making sense, I go back and re-read more carefully.

@YARNLADY…..Yes that’s it….scan-read. Perfect answer.

Actually, we don’t know anymore how we do this. Like riding a bicycle. Our unconscious mind takes over. Only when children learn to read do they need to make a conscious effort and do it word by word.

Later we just process patterns when we read.

@mattbrowne Do you think that reading is a skill that some people possess innately? For example, I read and read. My girls read and read. My son doesn’t like to read at all,and he is, of course, a much slower reader than the girls and I are. Is that just from lack of practice?

@Val123 – No, biological evolution can’t work this fast, but the capability to understand and speak a language is something humans possess innately. There’s a neurological basis in the so-called Wernicke and Broca areas in our brains. But we need to learn reading and writing and it takes time.

The capability to recognize patterns (in nature) and associate them with something else is also something humans possess innately as well. But the associations is something we learn and cultural evolution can speed up the process. Let’s take Native Americans who lived 5000 years ago as an example. There are two ways a child can learn about animal tracks in nature which might look like this

http://www.istockphoto.com/file_thumbview_approve/3766117/2/istockphoto_3766117-animal-tracks.jpg

1) It spots an animal which runs away, then takes a good look at the tracks left behind.
2) There is no animal in sight and a teacher points to a particular track on the ground and then describes the animal it belongs to

Interpreting signs in nature was key to human survival. Superior pattern recognition capabilities is a result of biological evolution over tens of thousands of years. When our distant ancestors invented written language our brains were more than ready. This article explains the history http://en.wikipedia.org/wiki/History_of_writing

Now what about the speed of reading? Like the experienced ancient hunter, an experienced reader can process patterns and find the correct associations much faster than an inexperienced one. Why? Here’s the neurobiological explanation:

Hebbian theory describes a basic mechanism for synaptic plasticity wherein an increase in synaptic efficacy arises from the presynaptic cell’s repeated and persistent stimulation of the postsynaptic cell. The theory is often summarized as “cells that fire together, wire together”. It is commonly evoked to explain some types of associative learning in which simultaneous activation of cells leads to pronounced increases in synaptic strength. Such learning is known as Hebbian learning, see http://en.wikipedia.org/wiki/Hebbian_theory

So you @Val123 as an experienced reader will take whole chunks of sentences and translate them into meaning in an instant like “biological evolution can’t work this fast”. You don’t scan this letter by letter which would be the other extreme. Your son is probably somewhere in between. He will recognize short common words like “work”, “this” and “fast” right away, but not very long words and/or unknown words. Maybe he’s seen the word “biological” only a few times. This would slow him down significantly.

Yes, reading much slower in most cases is due to lack of practice.