Talk:Subvocalization

[2004] term: subvocalization
Any chance of some references? Kappa 03:08, 25 Dec 2004 (UTC)

I would also like to see some references on the "new and improved" speed reading course theory that was previously posted. I reverted the last version of subvocalizing because I could not find a double standard definition for subvocalizing in any recent or old published prior research on subvocalization.

Regards Doug


 * Open up almost any speed reading book, especially the older ones. They all confuse the two types of subvocalizing.  They don't mention a "double standard" definition because most of the authors have no idea what they are talking about.  They all push "eliminating subvocalizing", and many students in classes who try this think that when they are attempting to "eliminate subvocalization" and still find themsevels "aware of the sounds that the words make" that they are doing something wrong.  That confusion between the two (possible) definitions leads to a lot of frustration on the part of students who try to "eliminate subvocalization".  One type is detrimental, the other is impossible to get rid of and is actually useful to a speed reader.  It behooves an article such as this to make the distinction when the source material is muddled.  Granted, the article was somewhat sloppily written and needs a complete rewrite, but reversion propogates the ambiguity...  perhaps having a second part of this article differentiating subvocalization from "being aware of the sounds that words make" as being very different from "subvocalization" as a serial process of sounding out the words? - Jim Whitaker Metaphorman 05:24, 20 Mar 2005 (UTC)

Open up any comprehensive book on reading and you will find that speed reading literature will say just about anything to get you to buy their product. The research shows that speed reading is extremely limited in its usefulness. Subvocalization is more a subject for reading rather than speed reading. However, I can explain more about subvocalization for reading in general for the sake of clarity. Cheers, DougDoctorDog


 * Quoted from the wikibook by Jim Whitaker:

For the purposes of this article, there are essentially three forms of subvocalization:


 * 1) The first form is visibly mouthing the words read aloud to oneself.
 * 2) The second form is imagining mouthing the words you read aloud to oneself without moving the lips.
 * 3) The third form is being aware that words have sounds associated with them, but not taking the time to wait for each word to be "spoken" before the readers' attention moves to the next word or group of words.

''Subvocalizing in style 1 or 2 are essentially the same [...] These patterns of reading [...] tend to place an absolute limit on reading speed in the neighborhood of 100 to 150 words per minute. These readers wait for the sounds of words to complete before moving to the next word [...] No matter how slow or fast a reader reads, there is no way to eliminate type 3 subvocalization, and according to muscle detection experiments, there is no way to eliminate involuntary muscle movement in the throat that corresponds to subvocalization.''


 * Although I'm not an advocate of speed reading, and don't agree with everything Jim has said, I am however curious about the topic and in this case I find the division into three forms of subvocalization to ring true for me. I find myself subvocalizing by method two most of the time I read. When I hum or count, either in my head or out loud, I can't really use method two, but I am still aware of my brain associating sounds with the words even when I read them very quickly. I believe this is the third method of subvocalization, which is very different from the second, because it both cannot be blocked out (nor is it helpful to try in terms of reading faster), and because it involves fully hearing the sounds just as if one was reading out loud to oneself.


 * I'm not sure if there is actually a clear break between these two, or if it is a continuum, for example when I read I don't think I always wait for the sounds to be fully vocalized before moving on to the next word or sentence.


 * I do think it's important to make a distinction between these two types in this article, otherwise it gives yet another thing to knock speed reading down for, and I think that has been done enough both here and at speed reading already. Richard001 18:16, 28 April 2006 (UTC)


 * I don't even know what waiting for the sound to complete really means above. I hear each word in it's entirety in my head when reading at what's considered a typical speed around 300 words per minute.  So then there must be a fourth form.  i.e. not imagining a mouth moving, but hearing the entire sound.

[2005] Subvocalization & Japanese
I think this should be known:


 * Subvocalization in reading kanji: Can Japanese text be comprehended without it? (Matsunaga, S.)
 * The role of phonology in reading Japanese: Or why I don't hear myself when reading Japanese (Sachiko Kinoshita)

Unfortunately they're not avaible on internet. --GLari 21:28, 30 May 2005 (UTC)


 * This is a very good point. Japanese kanji are varied in pronounciation depending on the word they are used in and whether they are kunyomi or onyomi. Some kanji like the one for 'day' have a large number of different pronounciations. If subvocalization is necessary to read, which is the crux of the argument against speed reading (which is essentially just reading a little faster, without the subvocalization), then wouldn't this render most Japanese unreadable?


 * I can't offer much knowledge on speed reading or kanji myself, as I am only able to read a tiny portion of Japanese kanji and no longer study the language, and I have only recently taken an interest in the possibility of speed reading. I can however share something a friend in my Japanese class said to me - his host mother was reading a newspaper, and he would point out kanji to her and ask what they meant - she couldn't even describe each kanji's meaning, let alone its pronounciation, but she could still fully understand what she was reading.


 * Some information about subvocalization in Japanese (or Chinese, which I imagine is basically the same with regard to kanji having different pronounciation) would be very useful to the discussion of the article. Richard001 17:24, 28 April 2006 (UTC)


 * Edit: Actually, now that I re-read my own statement, I'm not so sure about the subvocalizing part - if she couldn't vocalize it how could she read it out loud? I think it was more a case of not being able to give an independent meaning to the kanji. Richard001 17:33, 28 April 2006 (UTC)

Japanese is no different from any other language in the way it is processed. There are a number of books on this topic (such as Marshall Unger's "Ideogram: Chinese Characters and the Myth of Disembodied Meaning") and studies that show kanji primarily indicate sound, rather than meaning, to a native speaker, and that the subvocalization is there just as in English. The nature of kanji may increase the chance that the reader assigns a wrong reading, but it's just not possible to bypass sound entirely and somehow directly access meaning in the brain. A Japanese learner who knows the meaning of a kanji but not its reading is associating it with the sound of the word in their native language.

[2006] Speed reading
It seems to me like this article is very biased against speed reading. They might back it up with a reference, but I'm pretty sure there are many speedreading references that contradict this. --Brazucs ( TALK | CONTRIBS ) 05:18, 26 April 2006 (UTC)


 * That's the same impression I get. Richard001 00:58, 28 April 2006 (UTC)

The post above by Brazucs is exactly why farcical notions such as photo-reading gain acceptance in the public mind. To say that "I'm pretty sure there are many speedreading references that contradict this" illustrates Brazucs a priori acceptance of the phenomenon, despite the presented evidence to the contrary. Why would one believe in a fantastic claim without the requisite fantastic proof? —The preceding unsigned comment was added by Jeffwishart (talk • contribs).


 * You are correct. The burden of proof should be on the people claiming it's possible to read at super-human speeds.  Giving equal time to all possibilities, regardless of each of their likelihoods for being true relative to each other, is NOT nuetrality.  That in fact is exactly what skews the truth and creates bias.  Perhaps this article sounds biased against speed-reading to some people simply because the claims made by speed-readers and sellers of speed-reading seminars and materials are... well, full of crap!Rglong 21:02, 15 October 2007 (UTC)

I claim no expertise in the area of subvocalization, spead reading, or reading comprehension so I can only share anecdotal experience and my own opinion, which is significant since I am a "textbook case" of a slow to average reader (150-200 wpm) with good comprehension (80-90%) who has struggled to improve his reading. I attribute my consistently high reading comprehension rate to several factors: 1) subvocalization does indeed help my comprehension at the speeds I normally read (100-200wmp), 2) being a perfectionist by nature, I naturally resist skipping any word or fact in an effort to retain all the information, and 3) above average educational background. That said, I do agree that the article entry on subvocalization seems bias and, I believe, should be removed or cleaned up. The point of an encyclopedia is to define terms and concepts and then to place those terms in a broader context, including social, scientific, philosophical, and other areas of knowledge. The author's entry is, I believe, more a "position paper" where, yes, terms are defined and elablorated on to some extent, but mainly for the purpose of stating a particular view of the object's or concept's validity or usefulness. (A point of clarification about subvocalizing, I was a little confused by several definitions I read about subvocalizing. I understand that subvocalization may be moving one's lips without vocalizing, or that it may be not visibly moving one's lips but only moving the muscles inside the mouth/throat. I also take subvocalization to mean mentally repeating of the sound' of a word without any vocal movements at all. This is the type of subvocalizing I've done most of my life.) While an encyclopedia entry may certainly relay views others have about an object's usefuleness, it should be not be the major point of the article and should be done from a "third-person" objective perspective and should be balanced with a presentation of all other significant views. I only see one view presented in the article. Finally, in regard to my anecdotal experience, I have, for brief periods in my life, been able to nearly completely remove subvocalization from my reading practice without losing comprehension. In my experience, comprehension actually improved. Having subvocalized heavily most of my life, it was a very strange experience. To me, it was as though an "extra" step in the process of reading/comprehending had been removed. Most people have heard someone wish they could remove the top of their head and drop in a book, close their head, and retain the knowledge the book contains. My brief periods of reading (successfully, that is, with good comprehension) without subvocalizing felt something like this -- as though subvocalizing were a "processing filter" that had been removed, so that a wide open "information pipe" was left for information to flow freely from the page into my mind. This experience argues against the author's statement suggesting that subvocalization may be impossible to eliminate. During periods where subvocalization was not part of my normal reading practice, my reading rate skyrocketed and my experience of reading was very satisfying since I was not frustrated by the slow flow of information into my head. I long for, and am searching for, a way to retreive and retain the ability to read and comprehend without subvocalizing. My experience may or may not be valid. It remains untested of course. However, my initial argument against the article remains valid, I believe. The author seems to present the concept with obvious bias and does not objectively place their position supporting subvocalizing in a complete context. I would be interested to check back here and see others' views on the relationship between subvocalizing, speed reading, and comprehension. Wikimag 10:29, 15 April 2007 (UTC)wikimag


 * I also do not know a whole lot about the topics of subvocalization and speed reading, and would be more interested in them if there seemed to be more agreement within the topics themselves. From what I read out of this article, is it true to say that even though I don't feel any muscle movement (in terms of physically forming sounds) it is still present because that's how I learned to read? If that's the case, I have to disagree. I remember learning in my intro to psychology class that the ability to read is developed through the recognition of certain letterforms and the associated sounds; however it was said that beyond reading syllables to be pronounced, one learns to simply see a word and associate the meaning with the "shape", if you will, of the letters making up the word.


 * I've known for a long time that I read faster than most of my fellow students, but it wasn't until I went to university and talked about it with people that I realized that what I consider to be a normal reading tempo most other people consider skimming. I have no trouble with comprehension at all, and I scored extremely well on the verbal section of every test (standardized or not); I estimate my speed to be between 600-800 words/minute based on a test I did more than a few years ago. I find that I cannot read very quickly (or at all, sometimes it's distracting enough) if I catch myself saying the words in my head; it's far too slow for me to follow along, the only exception being philosophy (where sometimes it's necessary to re-read several sentences b/c of the length and content of the sentences).


 * I agree with the "extra filter" idea from Wikimag above, as I just stated I find it difficult to read and pay attention to what I'm reading if I "hear" myself "saying" the words as I read them. Not only is there no physical "subvocalization" (nothing anywhere near moving my lips or consciously moving the muscles in my throat/mouth; if there is some micro-muscle movement that I'm not aware of, I won't consider that subvocalization because it's not voluntary, and subvocalization strikes me as a reading/ reading comprehension technique), there is no "mental" subvocalization as well.


 * The only thing I find more annoying when I'm trying to read than my own subvocalization is when other people do it; worse than that is when they actually mumble the words under their breath.

[2007] motor theory of perception
consider:

liberman & mattingly 1985, zatorre et al. 1992 (-> motor theory of perception, perception-action integration, theory of mirror neurons)

[2009 III] subvocalization when doing mental arithmetic or general thought
I find myself subvocalizing *heavily* when I do mental arithmetic, in particular multiplication. It's annoying and slows me down. And I can't shut it up easily, tho I'm confident that it can be unlearned.

When thinking in general (trains of thought, planning,...) I find myself able to not subvocalize and the effect is enormous. My mind doesn't spend time subvocalizing; instead I can continue the train of thought or the planning.

It's like this whole paragraph "reduces" to one thought, not 10 seconds of subvocal speech. The thing is: it has always been just the thought, but with superfluous and reflexive verbiage around it.

It's a shame that there isn't more research into subvocalization outside of speedreading. --134.130.183.101 (talk) 10:28, 22 March 2009 (UTC)

[2009 IX] I don't subvocalise when I read
I started reading early enough, and in a bit of a language poor environment, that I picked up a lot of words by context from *reading* them first. No signing, no speaking. It was obvious because I'd often get to a word when talking, and then pause for a moment, as I figured out how to 'pronounce' it. I index words in my head by their spelling, and will often not realise what someone has said, if it is a homophone that I have mentally assigned to the wrong spelling.

I'm aware I still do this, because in books with exotic sci-fi or russian names, I realised on talking with someone that I didn't have any pronounciation associated with the words, I was recognising them by sight. I found it odd that they thought that was something that tripped them up. I was unaware that people continued subvocalising (ie in their head) past childhood.

Do people look at a picture and mentally describe things as 'green boat, blue sky'? Or to be more accurate, 'gree-n boh-t, blu skii'? Likewise, reading is a different mechanism to vocalisations. Not to mention, if you are 'subvocalising' something that you're "listening" too, then that's not what you'd do while listening anyway, and just means that the brain is doing an inefficient, habitual conversion, rather than something innate.

Anyway, there must be some resources that would demonstrate several of the assertions on this page are bunk, like "It may be impossible to totally eliminate subvocalization because people learn to read by associating the sight of words with their spoken sounds. "

I'm guessing this is just a low-traffic page. —Preceding unsigned comment added by 130.195.86.40 (talk) 04:34, 4 September 2009 (UTC)


 * One problem about reading by recognizing word shapes rather than sounding out the words is that it puts the need for speed in the wrong place. While speed reading is nice, speed listening is more important. If you don't understand some written text, then as you said, you can always read back over something until you get it. But if you don't understand spoken words because you have only experienced them as writing and not sounds, you cannot keep up with the conversation (or lecture, movie, etc.). Also, the path from sound input to word meanings is shorter and more reliable if it doesn't have to detour through the visual cortex to get to those meanings.


 * I'm curious if you can read text which is spelled phonetically (more or less) since you used some in your post:


 * Iff yue sowned owt thiss sentans tew yoreselph, yoo wil komprehenned et.


 * It should be impossible for you to understand phonetically represented words if you store words by their spelling, not their sounds. So each of the above words should look like unintelligible random letters to you.


 * If you learn a word and its meaning without learning its sound, not only could you not speak it when needed, but you would not recognize the word when you heard it. Some might consider that a major disability.


 * Finally, in one paragraph, you talk about subvocalizing when looking at pictures or listening to something. Subvocalization is a function of reading, not seeing or hearing, so I don't know what your point is.
 * 75.15.68.132 (talk) 21:31, 25 November 2010 (UTC)


 * The second anonymous commenter's last paragraph is right on. Subvocalization is linked to language. Reading involves language; looking at and recognizing pictures does not, ;describing  them does, but that's very different. --Thnidu (talk) 16:59, 16 May 2014 (UTC)

[2010 I] Don't get it
A really struggle to understand what subvocalisation is and this article does not seem to shed much light on it. I can understand what it means to siliently mouth words but if you don't do this then what is subvocalistion. Specifically what is the difference between subvocalisation and thinking. —Preceding unsigned comment added by 82.70.147.30 (talk) 16:58, 22 January 2010 (UTC)
 * Subvocalization is the voice inside your head that you hear when you read. It's like you're saying the word without actually saying it.76.218.222.17 (talk) 09:13, 7 September 2010 (UTC)
 * It's also that muscles move when doing so. If you pay attention to your thoughts and body you may realize this at some point. What helped me was noticing my breath was timed to be just as if I were speaking the thought.
 * 50.35.114.147 (talk) 20:16, 2 August 2023 (UTC)

[2010 XII] This article feels a bit like someone's joke
Especially first and third sentence referencing [2] Rayner, Keith and Pollatsek, Alexander (1994) The Psychology of Reading.

Recently... I have read without subvocalization while being actually asleep. I would be insanely happy to permamently get rid of subvoc in real world.

Short reminder regarding section above: one can think without internal voice. That voice can be just as much hard to evade.

81.190.231.67 (talk) 21:01, 6 December 2010 (UTC)

[2012] Detection
Chief Scientist for Neuroengineering at NASA Ames Research Center, Dr. Chuck Jorgensen, has suggested that it could have potential applications for rescue operations people, security and special operations forces, people with vocal cord problems, and might even find a place in gaming.

Sure. Just some time after they'll start using it in interrogation with tortures. 62.221.56.166 (talk) 11:04, 17 January 2012 (UTC)

[2014] Proposed merge with Subvocalization (Learning and Memory)
These articles seems to cover identical concepts, and a merge proposal has apparently been in place, undiscussed, since August 2012. Any reason they should not be merged?--Animalparty-- (talk) 03:16, 27 February 2014 (UTC)


 * Subvocalization (Learning and Memory) (Unlike the one in the section heading, this link is clickable in the mobile version on Android.)
 * Support. The learning and memory article is much more detailed. In editing them it might prove worthwhile to effectively merge them and then break off a different subtopic to its own article, but that's to be seen.
 * Thnidu (talk) 17:04, 16 May 2014 (UTC)

[2015 IX] Streamlined but merge is a good idea
Cleaned the article a bit for it to be easy to read. It is still biased against what simply exists (if we ignore minute muscle movements and concetrate just on the inner voice), but wording is more "maybe" now. There are no words needed to feel something, think or remember placement. While correcting this article I have read some sentences instantly. Some sentences almost instantly with sound of a couple words catched by accident. Reverting to imagining sounds sometimes can feel just awful even though I usually subvocalize. It seems some people do not see such possibility. Then, new books get written. But at the same time not all claims of speedreading marketers are to be trusted, oh no. SafetyControl (talk) 23:44, 8 September 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Subvocalization. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive https://web.archive.org/20070115091615/http://www.tfot.info:80/content/view/80/58/ to http://www.tfot.info/content/view/80/58/

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 01:19, 12 February 2016 (UTC)

Synthetic Telepathy
Does anyone have a reliable source on this? All I can find are conspiracies that the government has this technology or unreliable blogs. Also, the section seems to claim it is possible rather than just a hypothetical. Can't find any sources where subvocalization brain activity was linked to particular words or sentences either. --Malanet (talk) 18:35, 13 April 2017 (UTC)
 * Here's a related patent granted to Facebook in 2020:, and a journal article from a Springer publication in 2016: --Gmarmstrong (talk) 14:17, 4 July 2020 (UTC)