Dual Route reading model

Moderators: Debbie Hepplewhite, maizie, Lesley Drake, Susan Godsland

FEtutor
Posts: 348
Joined: Wed Nov 24, 2004 8:24 pm
Location: London

Re: dual route reading model

Post by FEtutor » Tue Sep 30, 2008 1:42 pm

Hugo wrote:Almost every struggling reader I have ever come across deploys just one word identification strategy and this is to sound out. They have an absolute belief that text is sound written down and only sound written down.
I do not dispute that this is what you have found, Hugo, but I do find it surprising. In my experience in FE supporting beginner and struggling adult readers and spellers, students invariably use the "skip word and guess from context" strategy. They expect themselves to "know" a word (or resign themselves to "not know"). They can be amazed to discover that the sounds they speak are reflected in the complex Code correspondences of the written word, and wonder why they have not been shown this before. If they do have difficulties sounding out it is because they have not practised blending/ segmenting pure sounds and /or only know of one-letter-one-sound. They have not had the opportunity to identify, learn and process more Advanced Code correspondences. So, the trouble I find is not that most struggling students decode too much, but that they do not decode well enough.

The thought occurs to me: maybe a lot of us are affected by our expectations, training, and what we feel we can teach effectively. I write as someone who swallowed Frank Smith, language experience etc, hook line and sinker in my early eighties literacy training (but had little success with it), felt I could teach spelling after dyslexia training in the late eighties, but only had success teaching beginner readers in this decade, after SP training.

mtyler
Posts: 371
Joined: Wed Jul 12, 2006 5:56 pm
Location: Minnesota, USA

Post by mtyler » Tue Sep 30, 2008 4:37 pm

Hugo wrote:
Almost every struggling reader I have ever come across deploys just one word identification strategy and this is to sound out. They have an absolute belief that text is sound written down and only sound written down.

This is not enough information to be an indictment of synthetic phonics, first and only. Several things could be going on:

1. Part word guessing.

The adults with reading problems that I have spoken with and observed reading will only read the first few sounds in a word and then guess. With each correspondence added the number of possibilities narrow, however, in many words there are still quite a number of possible matches 3-4 sounds into the word. The reader tries a few, some don't make sense and they have now got the aural information from the words they have tried mixed up with the information on the page. Confusion ensues.

When I pointed out the words they had missed, they did not read the word--they stated one of the words they had said previously. It was not until I took them correspondence by correspondence through the word that they got it right. It wasn't that they didn't know the pertinent information, it was that they habitually did not use it.

2. They don't know enough of the code.

Many words have rare correspondences. If people have not been schooled in the possibilities they do not have a small enough (or large enough) pool of sounds to try when encountering new words. Some may not understand the possibilities of groupings or have the skill to regroup (I read "awry" as /aw-r-ee/ rather than /u-r-ie/ when I first encountered it). Strings of vowels (such as in "arduous") often confuse people because they have not been taught where it is possible to make breaks in correspondences. Most phonics programs seem to teach only a small subset of possibilities leaving the rest for the student to figure out. Some students may have the skill of blending and know to sound out, but if they don't know what sounds to try and what letter combinations are valid, they may easily be confused and give up.

Melissa
Minnesota, USA

Hugo
Posts: 107
Joined: Mon Apr 07, 2008 7:42 pm

dual route reading theory

Post by Hugo » Tue Sep 30, 2008 5:29 pm

Jenny, you are right to be relaxed about the dual route theory and that it relates to the word identification process in fluent readers. In case everyone is not so happy with the idea that fluent readers identify most words visually and that the visual route is the direct one can I urge Rayner & Pollatsek (1989) upon you. On p. 109 they write that “The common ground for all positions is that direct visual access is important and that sound encoding plays some part.” I absolutely accept that with very early learners (and we start formal schooling much too early, in my view) the situation is different, by definition. In Ellis 1984 & 1993 (second edition), and in Rayner & Pollatsek you can find evidence for the dual route theory as such if you need it. To be honest, the idea that there are two routes available to reading and that the most direct one is the visual one seems to me to have been accepted right across cognitive psychology. There seems to be consensus.

However, I hope it is worth reiterating my view that “phonics” – shown beyond reasonable doubt to be extremely important in early literacy acquisition – covers a multitude of skills and abilities in reality. I think we should use the more precise term “alphabetic principle” to describe the understanding we are trying to instil. It has been shown that phonics without letter awareness is of small utility and letter awareness has been shown to be as powerful a predictor of reading success as phonological awareness is. (Neither has been shown to have a causal relationship, but that’s another story!) I would like to redirect you to a quote from Adams (1990) in a posting of mine a couple back where she says that “…it is possible that the ability to sound words out … is not the primary positive outcome of phonics instruction … the most critical factor beneath fluent word reading is the ability to recognise letters, spelling patterns and whole words effortlessly, automatically and visually.”

What I am closing in upon is the idea, with which I do not perfectly agree, that phonic instruction is, or should be, teaching that reading is “phonic decoding”. There is a subtle, but emphatic, distinction between teaching the alphabetic principle (letters and letter patterns represent sounds) and teaching that text is sound written down and must be primarily, perhaps even exclusively, reconstituted into sound. I have found a reference for the assertion that an attack methodology, once taught, can be very firmly internalised and rigidly adhered to – especially by the “weaker” or less confident pupil. It is Barr, R. (1974) The effect of instruction on pupil reading strategies. RRQ 10: 555-582. So although I absolutely agree that the very early learner is indeed a different prospect to the adult learner especially in terms of orthographic pattern knowledge and understanding, I might not agree, to the same degree, in the absolute desirability of phonologic attack, notwithstanding. My beef is with the exclusive phonics approach. Reading (word identification, I mean) is not so simple. It should be aimed at visual and automatic recognition from as early as possible – visual (and motor) approaches alongside a phonic approach, in other words.

So I think that at least you and I may be quite close on this aspect except that when you say that the best results are achieved where pupils “rely heavily on sounding out” I wonder if that is all they are, in fact, being taught to do. One thing is for sure and that is that we need to get students reading (the Matthew effect is enormous) and we need to get them to like doing it!

I am intrigued by the idea, which seems to me to be extraordinarily unlikely, that our visual recognition literacy lexicon cannot be larger than 2000 items. Where is the evidence?

Melissa, I agree with much you say, especially the remark that "they don't know enough of the code" - except that by "code" I might mean more than you do?!

FE tutor we may have been witnessing similar events? The students I have taught did not use sounding out at all effectively, but they did attacked solely by sound. Extreme strategic impoverishment.
www.hugokerr.info

mtyler
Posts: 371
Joined: Wed Jul 12, 2006 5:56 pm
Location: Minnesota, USA

Post by mtyler » Tue Sep 30, 2008 7:20 pm

Hugo states:

In case everyone is not so happy with the idea that fluent readers identify most words visually and that the visual route is the direct one can I urge Rayner & Pollatsek (1989) upon you. On p. 109 they write that “The common ground for all positions is that direct visual access is important and that sound encoding plays some part.

If by "visual route" you mean understanding a word as a whole-outline shape, etc. without knowledge of its phoneme-grapheme correspondence parts--then recent research disputes the assertion.

http://www.plosone.org/article/info%3Ad ... ne.0000680

This study used various methods to discern what percentage of 3 methods fluent readers use (decoding, sight, and context) accounted for reading rates. The study found that decoding represented 62 %, sight word knowledge 16 % and context 21 %.

Melissa
Minnesota, USA

Hugo
Posts: 107
Joined: Mon Apr 07, 2008 7:42 pm

dual route reading theory

Post by Hugo » Tue Sep 30, 2008 8:16 pm

If by "visual route" you mean understanding a word as a whole-outline shape, etc. without knowledge of its phoneme-grapheme correspondence parts--then recent research disputes the assertion.

No, I don't. The unit we "read" (by which I mean identify) is a variable thing depending on the circumstances. If a very common word (most of the 'key words' perhaps) maybe by the whole word, if a common pattern or morpheme, maybe thus, if a very unfamiliar word maybe by letters or letter patterns. I think the word-shape idea came and went, fleetingly, a while back. I don't think much research disputes that the very initial stages of identification are visual in most instances. Considerable association will instantly result in much enrichment - eg with sounds - but the word will usually have been identified visually.

However, I am about to go and look at the article you kindly appended!
www.hugokerr.info

chew8
Posts: 4161
Joined: Sun Nov 02, 2003 6:26 pm

Post by chew8 » Tue Sep 30, 2008 10:37 pm

Hugo's comments are in black below and mine are in blue.

'What I am closing in upon is the idea, with which I do not perfectly agree, that phonic instruction is, or should be, teaching that reading is “phonic decoding”. There is a subtle, but emphatic, distinction between teaching the alphabetic principle (letters and letter patterns represent sounds) and teaching that text is sound written down and must be primarily, perhaps even exclusively, reconstituted into sound'.

I'm not sure that I understand your problem, Hugo. When I write anything, the spoken forms of the words that I am thinking of putting down are going through my mind, and I often experiment mentally with different words and word-orders before putting anything down - any text that I produce is definitely sound written down. Is this not your experience? However we see it as adults, though, I don't see how we can avoid giving young children the impression that reading involves reconstituting sound from print - isn't this the obvious conclusion for them to draw when we read them stories, for example? And if we are teaching them to read, we will either point to a whole written word and tell them what it is or demonstrate sounding out and blending as the way to work words out - in both cases, we are surely giving the impression that reading involves converting written words into spoken words.

There's yet another point: when children first start learning to read, they already know the spoken forms and meanings of thousands of words. It would seem silly not to captalise on this, where captalising means showing them that if they can translate written words into spoken words, meaning will immediately be obvious.


'So although I absolutely agree that the very early learner is indeed a different prospect to the adult learner especially in terms of orthographic pattern knowledge and understanding, I might not agree, to the same degree, in the absolute desirability of phonologic attack, notwithstanding. My beef is with the exclusive phonics approach. Reading (word identification, I mean) is not so simple. It should be aimed at visual and automatic recognition from as early as possible – visual (and motor) approaches alongside a phonic approach, in other words'.

The problem is that many of us have found that this can be confusing for young children - or have you taught enough of them to know otherwise, Hugo?

'So I think that at least you and I may be quite close on this aspect except that when you say that the best results are achieved where pupils “rely heavily on sounding out” I wonder if that is all they are, in fact, being taught to do.'

Programmes which have good classroom records (e.g. 'Jolly Phonics', 'Read Write Inc.' and 'Fast Phonics First') also start introducing a few of what JP calls 'tricky words' after the first few weeks, but (a) they don't just teach them as visual global wholes (they draw attention to expected and unexpected grapheme-phoneme correspondences), and (b) these words are far outnumbered by the decodable words that are introduced. I happen to think that the introduction of such words is a help rather than a hindrance, but only if they constitute a very small percentage of the total number of words that children are exposed to - the children rely HEAVILY on sounding out even if they don't rely EXCLUSIVELY on it.

'One thing is for sure and that is that we need to get students reading (the Matthew effect is enormous) and we need to get them to like doing it!'

I agree, but this actually tends to happen much more in schools which teach children to rely 'heavily on sounding out' than in schools which teach a bit of sounding out but also teach a lot of 'sight' words.

Jenny C.

noor_warda
Posts: 35
Joined: Tue Aug 26, 2008 7:27 pm

Post by noor_warda » Tue Sep 30, 2008 11:47 pm

I would like to add my personal experiences to this debate. As most of you know already, I am learning to read Arabic, when I first posted here I had zero sight words in Arabic, but could decode any Arabic text that was either written with all the vowel marks, or written without the vowel marks using grammar and vocabulary that I am familiar with. (the short vowel marks are not written in Arabic text intended for fluent readers who are native speakers, they are written in children's books and books that are read by non Arabs)

The situation is pretty much the same, but Jenny asked me to let her know when I start taking in Arabic words as a whole. I don't yet (and I've been learning Arabic for a few years, on and off, and living in an Arabic speaking country for just over a year). However I have been practicing reading a lot lately, and I can report the following sight words: The Arabic word for I (three letters) for in (two letters, written in a special way so it looks like just one symbol so maybe it doesn't count) the word for restaurant (four letters) the word for God (four letters, but the last three are joined as a special symbol not used for any other word, so actually just two symbols) and the phrase God Willing (three words: 2 letters, 3 letters, 2 symbols (God)) The rest of Arabic I have to decode letter by letter (or letter+vowelmark, letter+vowelmark, letter+vowelmark)

I only recently, in the last week or so, started to see the word for restaurant as a sightword. I have been doing lots of practicing reading shop signs as I go out and about, so have decoded the word for restaurant many times as there are a lot of them where I live. For a while, I'd not recognise the word straight away, but after decoding the first two letters, I would recognise the other two as completing the word for restaurant. Recently I have read this word just by looking at it without decoding it.

When it comes to decoding Arabic, I don't have any difficulty understanding the words that I decode if they are in my vocabulary (limited, I don't speak Arabic that well yet). I don't consider myself as a struggling reader in Arabic, even though I occasionally decode letters wrongly because some letters look the same but they are only distinguished by whether the dots are above or below the line, and even though I read very slowly.

Regarding teaching beginners I think it is very important to realise that you can't make a beginner do something by emulating what is going on in the mind of a person who is skilled. When you first learn to drive you have to consciously think your way through several different processes even just to start the car. Every time you have to change gear you have to think your way through the process. However an expert driver can just think "start car" and they start the car, they can change gears without thinking at all. Any skill you learn is like this, you can't expect to teach beginners by shortcutting thinking through and doing all the steps involved in the task, because it is through repeatedly doing whatever is required in the correct way that the skill becomes automatic. You can't jump the process of aquiring a skill and skip straight to automaticity. Therefore trying to make beginners emulate the way experts do something is not a logical way to teach. You have to break the skill down into its components and have the beginner practice these until they develop automaticity.

By teaching sightwords you are not skipping this process, or emulating the way skilled readers read (because the skilled reader can tackle unfamiliar words using their very well honed sounding out and blending skills), you are just dramatically increasing the number of individual components that need to be learned. Thus in the early stages it may appear to be quicker because a child learning sightwords can read ten words after learning ten units of information, whilst the child learning phonics can only read ten sounds and not blend (YET) - however once the child can blend words the number of words they can read grows exponentially as they learn more correspondences, overtaking and racing ahead of the sightword reading child who can only learn one new word at a time.

When it comes to the way reading is often taught, it is actually expecting a child to run before they can walk. I have been thinking carefully about what really stopped me from making progress in reading English to the point that I was diagnosed dyslexic at age 19 after struggling through secondary school labelled by teachers as "obviously bright but lazy" and employing all manner of "coping" aka reading avoidance strategies. Not only that, but I realised that a problem I'd had when I first started learning to read Arabic as an adult was actually rooted in the way I was taught reading in primary school. When I started learning Arabic, I found reading without the vowel marks easier than reading with vowel marks, even when I didn't understand the words and had to guess the short vowels using my incomplete knowledge of grammar. Everyone else thought this bizarre and illogical, but I had my ed-psych report saying I was dyslexic so the weirdness in this I put down to that.

What I now realise was actually going on, was that having twice as many symbols to decode was putting me into some kind of "panic mode" and instead of systematically decoding, my eyes were all over the place not taking in much information and certainly not taking it in in the right order. This problem was remedied before I even knew the cause of it, by a patient teacher who made me go through the text very slowly and systematically decode all the symbols in the correct order. She gave me back the TIME it takes to decode properly and therefore read properly. Having read on this forum of how a lot of struggling readers are guessing, thinking it's better to say anything than nothing, I'd imagine something similar to what I was experiencing is going on in their heads.

Looking back at how I read as a child, I can see the same thing going on... I used to "read" Asterix books, I enjoyed them, and figured out the story from the pictures and reading some of the words and guessing the rest. The names of the characters in Asterix are long and complicated, I used to completely guess them from the first letter or two and the last two letters. When I was diagnosed with dyslexia age 19, I still had problems with the middle of words! Why? Well I remember in primary school there was a LOT of emphasis on reading quickly and "with expression" - the children who read quickly were praised a lot and considered the best readers, anyone reading out loud was encouraged to make their voice go up and down as they read. If you didn't know a word the teacher would tell you what it was.

However, expecting a child to read quickly with expression before they've mastered decoding is like taking someone on the motorway before they've mastered changing gear. It doesn't make them read like expert readers, it makes them employ coping strategies. It is really vitally important to give learner readers time to decode properly. Even something as simple as jumping in to say the the word that they are decoding before they've read it should not be done, ever! This annoys me greatly when someone does this to me when I'm reading Arabic, and it comes from them not understanding the difference between struggling and decoding. (Arabs don't do this, its English people that do! Probably because all Arabs learn to read Arabic the synthetic phonics way) Decoding slowly is not struggling!!! If you do this to a child you are giving them the message that taking your time is not acceptable, that having to go through the word sound by sound is not acceptable and you should magically know what word it is already. I don't even think correcting mistakes is necessary, just ask the child to read the word again if they got one sound in it wrong (e.g. mixing up d and b which is a similar mistake to the kind I sometimes make when decoding Arabic). If they don't actually know a particular sound, that is when you need to tell them what the sound is. And forget this "reading with expression"! - thankfully I was never taught to guess words, but all I can see this doing is re-enforcing the message to the child that they should already know what the word is without decoding it.

And it is important to note here that what I knew about phonics 99% of it I learned from my parents, not the school. We did a little phonics work at school; someone on this forum said that at one time primary schools were teaching "look and say" initially with some analytic phonics later on... well thats what my primary school did. There was little to no encouragement to use phonics for reading at school, although my dad used to stop me guessing and make me decode words properly, even drawing pencil lines between each syllable and making me decode syllable by syllable and teaching me some complex code too. Given the long amount of time it's taken me to get any sight words at all in Arabic, I honestly believe that if my parents hadn't taught me phonics I'd have been completely illiterate. The only thing they did wrong is they didn't teach me enough phonics and they trusted the school too much, because I wasn't a "free reader" from what they taught me. So the try to read as quickly as possible mentality I got into at school prevailed and my reading didn't progress as it should have.

What helped me most other than my parents teaching me some phonics, was the dyslexic tutor at university. Having been diagnosed with dyslexia I decided to give education another go (I left school believing that I was completely incompatible with the education system, I had quite a few other problems besides haivng to continuously employ reading avoidance strategies, and when I was diagnosed with dyslexia the ed psych said my level of intelligence was such that if I'd had appropriate help with the dyslexia reated problems I could have got straight A's and gone to Oxford or Cambridge, which encouraged me to think about going to university) - this tutor did some phonics work with me, and a lot of work on study skills, and also got me going back to systematically decoding text as slowly as I needed to, encouraging me to re-read sentences slowly and carefully until I understood them. After about a year at university, I was reading well enough to read the scientific journals that were necessary for doing the coursework, and I also realised just how easy and regular scientific word spellings are, and I didn't need the dyslexia tutoring after my first year. It took me quite a while longer to read well enough to enjoy adult reading age fiction - and what helped here is having something to read that was interesting enough to want to slowly and systematically decode my way through because I needed an awful lot of practice at this! I recently read "The Hobbit" and thoroughly enjoyed it; this is a book I once thought I'd never be able to read. I have become a fluent reader, and I became one by systematically decoding words until either they became sightwords or I decode them so quickly and automatically I don't notice I'm doing it.

I do believe that we dyslexic people have something different about our brains; when at university studying neurobiology I learned about this and read a fair bit of research, what I found was that dyslexic people had more symmetrical brains with less lateralisation (i.e. fewer skills being focussed in a specific half of the brain) than most people, but the research quite clearly showed that lots of other people had the same symmetrical brain/less lateralisation and could read perfectly well, but still had various dyslexic traits such as a tendancy towards ambidextrousness, confusing left and right, reversing symbols and a lot of skills that are associated with dyslexia too. I now believe that the "missing link" in the data - why some symmetrically brained people can read well but other's can't - that is the way they were taught to read. I take longer to remember symbols and have a lot of difficulty learning sequences (even my own phone number!) and when I take in new information I take in the "global whole" first and let the details fall into place later. Disciplining my eyes to systematically decode in a sequence is difficult for me (but clearly not impossible seeing as I can do it!) - Its clear that with a brain like mine I can't learn to read using the "whole word" method. I have to learn the rules (the global whole) and apply them to specific cases (the details). People with brains like mine I strongly suspect will take longer than average to learn to read using synthetic phonics, in particular taking longer to learn to recognise the graphemes and to go from slowly decoding sound by sound to taking in words as whole entities (whether because decoding is automatic and subconscious or because they are now processing sightwords) however that does not mean that these children should be taught with non-SP methods, just that they need a lot more practice to do these two particular skills automatically. Learning words as whole entities is a LOT LOT more difficult for people like me, there is no "global whole" that can be applied to specific cases with learning whole words, just more and more things that are very hard to learn.

I would like to challenge the illogicality of the "whole language" method whereby they claim that individual phonemes/graphemes are meaningless. They are not!!! b stands for a sound, /b/; they all stand for sounds! I can't learn meaningless strings of information easily at all, by considering the whole word as the smallest unit of meaning, the letters in it become a meaningless string of just the kind that I find it extremely difficult to learn. Heiroglypics would be easier to learn, because the symbol in the glyph bears some similarity to the word it codes for. And this guessing nonsense sounds like an admittance of failure to teach a skill that eliminates the necessity for guessing. As for reading ahead for the context - I find it difficult to keep my eyes in the right place on the text. Even if I look away for a moment it takes me a while to find where I was before looking away (I found copying from a blackboard or textbook extremely extremely difficult at school because of this) - so if I were to keep on trying to look ahead for context I think I'd just lose my place and spend more time trying to find the words I was trying to read than actually reading anything.

Hugo
Posts: 107
Joined: Mon Apr 07, 2008 7:42 pm

dual route reading model

Post by Hugo » Wed Oct 01, 2008 2:38 pm

Jenny, I don’t feel I have a problem! Our difficulty is a linguistic one. We are talking about different things and repeatedly passing in the night. When I assert that word identification is primarily visually achieved I am talking about pure cognitive psychology of the first milliseconds of a reading’s life. The most efficient way to recognise a visual signal will always be by direct visual means, direct to meaning. That is how all commonly visualised things are recognised and words are, in this respect, just another visual stimulus. Spreading activation will ensure loads of other associations also rapidly become current, the sound of the language retrieved from a page a thunderously important aspect of it. This is not the same as saying that text is sound written down. Text is meaning written down in visual symbols. It’s a subtle point, but it is not a small one as so much praxis flows from it. However, I think we may have exhausted the subject and our readers?

I am not at all suggesting that we do not capitalise on the verbal knowledge and phonological correspondences in language, nor the value of sounding out as an approach to decoding. I am merely arguing for proportion. To be personal a moment, I absolutely have to say that you write in a much more proportionate and reasonable way than many do on the subject of phonics. This is very unlike many in the field who insist upon pure and exclusive phonics to the exclusion of all other method. You and I know that many people resist any dilution of it. As you will know, in the USA there are schools using scripted phonic methods which are rigidly enforced to the letter by admin staff creeping soundlessly along the corridors in stockinged feet, script in hand, and applying their authoritarian ears to the doors of classrooms to check that no teacher is deviating from it in any way. The profession is deeply and widely disempowered. Those who are even slightly inclined to critique their imposed system get labelled as boneheaded troublemakers at best, job security is also an issue for many.

The fact is that, as I think you and I agree in principle, reading is a complex skill in which many different cognitive abilities mesh, in variable proportions, and that a fluent reader needs more than one means of attack upon text. We disagree only in degree, in reality, I think.

I have never taught children in a classroom and I absolutely defer to your exertise in this regard of course, but would like to reiterate that you seem to me to be much wiser and more flexible than many who may be obliged to swallow dogma and reproduce narrow, or shallow, pedagogies. You and I both know that there are many people, apparently with access to some high powered ears, who are proclaiming single silver bullet solutions and ferocious inflexibility in their pursuit. Research is exaggerated and taken out of context, nuanced thinking evaporates. Trained teachers are not treasured. Experience counts for little. American teachers with years of experience, and some Brits too, frequently write that they have to circumvent and even subvert the ‘system’ approaches they are required to at least declare they are following. Younger teachers seem “dazed and confused”. It seems to me that grizzled politicos in the US dictate classroom detail, and that it is time that the teaching profession was better trusted.

The Arabic question is very interesting. I, too, learned Arabic, not so much literacy in it as speaking it. It’s a wonderful language. My subjective experience, for what this is worth (it was years ago and prior to my becoming interested in literacy as such) was that I sight read quite a few words precisely because their system (vowel-free and calligraphically individual) seemed to be more ‘symbolic’ than ours. No vowels, muffled sound signal is how I recall it now, but what that observation is worth I am not certain.

The “panic mode” reference to “eyes all over the place” really resonated with me. I think that affect is a far more important effect than we presently allow. Most ABE students use very strong emotional language about their experiences in respect of literacy, and addressing affective issues and control helps enormously. And I remember Tony Martin powerfully describing how the notes swim round the page for him when he is with his piano teacher.

But I feel written out on the subject of dual route access to meaning!
www.hugokerr.info

FEtutor
Posts: 348
Joined: Wed Nov 24, 2004 8:24 pm
Location: London

Re: dual route reading theory

Post by FEtutor » Wed Oct 01, 2008 3:24 pm

Hugo wrote: FE tutor we may have been witnessing similar events? The students I have taught did not use sounding out at all effectively, but they did attacked solely by sound. Extreme strategic impoverishment.
Hugo, I'm confused. The students I meet have been using guessing/ context because they cannot decode effectively, but you say the students you meet only use ineffective decoding ("they did attacked solely by sound"): not the same strategies to my mind. But surely both groups would need effective phonics teaching (such as Noor Warda describes in her inspirational post) if their reading is to improve? Once they can access the words then they can use context etc to monitor for meaning.

Noor- would it be OK to show parts of your post to an FE student I teach?

noor_warda
Posts: 35
Joined: Tue Aug 26, 2008 7:27 pm

Post by noor_warda » Wed Oct 01, 2008 6:38 pm

Yes it would be okay to use anything from my last post - I'm glad that my experiences can help someone else.

JIM CURRAN
Posts: 3127
Joined: Fri Oct 31, 2003 7:18 am

Subject

Post by JIM CURRAN » Thu Oct 02, 2008 1:48 pm

Research by Van Orden ( 1987, Van Orden, Johnston & Hale 1988 ) tends to support the theory that phonological information is immediately activated during the course of word recognition and that its activation is obligatory.

Subjects were presented with words which appeared rapidly on screen and asked if the word belonged to a particular category. One of the categories used was “flower”. When subjects were presented with the word “rows” they placed it in the flower category. Another category was “parts of the body” when subjected were presented with the word “ hare” they answered in the affirmative.

Hugo
Posts: 107
Joined: Mon Apr 07, 2008 7:42 pm

dual route reading theory

Post by Hugo » Thu Oct 02, 2008 5:08 pm

Jim, phonological information will be activated and probably is obligatory (associative connections always are, though their intensity may vary according to circumstance). The immediacy is debatable - it cannot be quite as immediate as the visual information in the case of familiar words. Nothing is absolutely immediate, of course.

With regard to the research you quote this is the pseudohomophone effect and the interpretation of it is somewhat contentious. Read Taft, Marcus (1991) Reading & the mental lexicon. LEA, or Rayner K & Pollatsek A (1989) The psychology of reading. Prentice-Hall.
www.hugokerr.info

JIM CURRAN
Posts: 3127
Joined: Fri Oct 31, 2003 7:18 am

Subject

Post by JIM CURRAN » Fri Oct 03, 2008 7:16 am

‘Seidenberg ( 1999) discussed how connectionist models lead to a different view of English orthography than do dual- route models. The latter tend to imply that the orthography is a mess - there is a large class of words ( the ‘exception’ words ) that need a different route because they are so irregular. Connectionists models emphasize the fact that, as Seidenberg ( 1999 ) put it , ‘have’ is not pronounced ‘ banana’. It’s pronunciation has similarities to other words that share its orthographic overlap ( hive, had, etc.). Even the notorious ‘pint’ ( a word residing on everyone’s list of ‘exceptions’ ) has three of four phonemes pronounced regularly. Connectionist models capture these facts of the predictability and nonarbitrariness of words deemed ‘exceptions’ in ways that dual-route models do not. Connectionist models also give us a different slant on how much of the orthography should be considered ‘decodable’.’

( Stanovich- Progress In Understanding Reading page 217 )

Hugo
Posts: 107
Joined: Mon Apr 07, 2008 7:42 pm

dual route reading model

Post by Hugo » Fri Oct 03, 2008 9:59 am

I have not read anything yet which makes me think that a connectionist model and a dual-route model are necessarily mutually exclusive. A connectionist model nicely explains how, although we have no rules on board, we can behave as if we were following them (and underpins my refusal to teach rules) but does not counter my view that a visual signal will, if it is at all familiar, be visually managed in the first instance. I have great difficulty seeing what is the problem with this view! I am only talking about the very initial stage of stimulus recognition, not management of recognised symbol thereafter.

To go back to common sense. Text is squiggles on paper - pure visual signal. I cannot believe that these symbols are recognised as, then neurologically (and expensively) treated as, completely different to almost all other visual stimuli. Like my bullock example, I do not see cattle, recognise them as noise-emitting creatures, discard my visual acumen and search a phonological store of animal noises and then decide that I have seen cattle. It must be a similar story with text. Having decided it is cattle I am seeing, then many other aspects of cattle and circumstance will flood around and cause me to have a richer experience than pure identification, but initially I did that visually. This is NOT to deny the effect and use of the enriching non-visual information (nor the use of noise when teaching a child about cattle!). The world is a holistic affair (and see Gibson on "affordances" - that we perhaps only really "perceive" that which has salience, or utility, for us).

Incidentally, I have corresponded with Stanovich, who is considerably sceptical about reading disability & dyslexia. The page you refer to seems to me confused because of his underlying unease about this. But I find connectionist theory just a mechanism explanation - it would explain the nuts and bolts of cognitive management of stimuli but not affect the dual route theory in that both routes, so to speak, will probably be connectively managed. (Difficult to see how it could be otherwise.)
www.hugokerr.info

kenm
Posts: 1495
Joined: Sat Dec 17, 2005 5:19 pm
Location: Berkshire

Post by kenm » Fri Oct 03, 2008 11:17 am

Hugo wrote:The immediacy is debatable - it cannot be quite as immediate as the visual information in the case of familiar words.
Why do you qualify this? The retina and the visual cortex are organised into successive stages to extract the features of a scene. The first stage of processing is the extraction of boundaries between regions that differ in a characteristic such as brightness or hue. This takes place within each small area of the visual field. The next stage is the association of all the boundaries of the same orientation within the area to a neuron in the next stage. The supporting evidence for these stages is strong, being partly from anatomy, partly from perceptual experiments,* and partly from experiments with kittens raised in unusual visual environments, the last showing how some of the mechanisms for vision develop only with exposure to appropriate stimuli.

The pattern recognition specific to reading presumably makes use of the features produced by the stages I have described, but I have not been able to find any information about this in detail corresponding to what is known about the earlier phases. I can hypothesize that a set of features in a particular spatial arrangement could combine to stimulate one (or a few normally functioning together) that correspond to a particular letter. The alternative, that a larger set of features could combine directly to stimulate a "meaning" neuron will be considered below. The next phase is the interesting one. Given neurons corresponding to a set of letters in a sequence, is it plausible that they could directly stimulate a "meaning" neuron? This presents an enormous problem: accommodating all the interconnections that would be required. Deriving meaning from the earlier features without using letters as an intermediate stage would require even more interconnections. Since recognition of the meanings of words without use of their sounds would require the construction of some such network as I have described, in parallel to the similar one that exists for spoken language (and which has been refined over some tens of thousands of years), I am not surprised that 2000 words seems to be a typical maximum for individuals who are so unfortunate as to have been led along this path.

At this stage of the argument, my question for Hugo is, "What brain mechanism do you think makes a word familiar?" For me, it is the connection (or connections, one per phoneme?) to the speech perception areas, and what is familiar is its sound.

* A particular one of these involves experimental subjects observing the flash of a xenon light which depletes the visual pigments and leaves an afterimage on the retina. This afterimage then changes in appearance as the neurons associated with its component features fatigue and cease striking. The subject matches the incomplete pattern against the various patterns derived from subsets of the features. I was a subject in such an experiment in about 1976. The pattern was a plus sign (+) within a circle. The features that disappeared were the vertical bar, the horizontal bar and, IIRC, the circle.
"... the innovator has as enemies all those who have done well under the old regime, and only lukewarm allies among those who may do well under the new." Niccolo Macchiavelli, "The Prince", Chapter 6

Post Reply

Who is online

Users browsing this forum: No registered users and 8 guests