Following on from my previous posting, I want to consider what the implications are for what our eyes are doing when we are learning to read?
Certainly, because the span of fixations are more limited, the beginning reader needs more fixations and saccades to hold text in foveal view. This and the fact that publishers increase font size may, the authors speculate, lead beginning readers to look at the initial letters in a word and to guess. Of course, as we are well aware, many teachers promoting multi-cueing techniques reinforce this tendency by asking young children to look at the first letter or the accompanying illustrations and to guess what word might come next.
Such a strategy may seem to offer a quick solution, especially if a word is guessed correctly. However, this is rarely the case! Multi-cueing must always collapse back into a whole language approach, which, to paraphrase Diane McGuinness, promises everything and delivers nothing’*.
On the other hand, in their research article ‘Literacy Development: Insights from Research on Skilled Reading’, Jane Ashby and Keith Rayner insist that by attending carefully to the detail of words and linking print to sound, a child is embedding and anticipating advances in later reading development. After all, it is the internal details, the complexities of the spellings of many of the vowel sounds, that are fundamental to successful decoding. The corollary of this is that it is vitally important to teach beginning readers using high quality phonics programmes because children who can recode spellings into speech sounds are able to match them to their oral/aural repertoire. This skill is also an indispensable device in ‘generalizing the meaning of spoken words to written words [and] is a valuable self-teaching tool’.
In addition, being able to identify (read) words without having to resort to context has a number of crucial ramifications:
First, it helps to build high quality representations of word-specific sound-spelling correspondences.
Second, the ability to process text automatically enables a reader to apply themselves entirely and without distraction to such things as ambiguity of language, lexical choice, the ‘vagaries’ of plot construction, as well as the complexities of syntax and grammar in more challenging texts. As I have pointed out before in postings, if the cognitive load of decoding text is low, resources can be allocated to other, higher order skills.
Third, automatic word recognition (or decoding) also reduces the difference between reading and listening comprehension. In the beginning, readers’ listening comprehension skills vastly exceed their reading comprehension skills; yet, as decoding ability improves to the point of automaticity, the disparity between the two reduces to the point where written text is easily comprehensible.
Moreover, given how lexically impoverished everyday speech is in comparison with written language, reading will offer vastly more opportunities for learning new words than oral language alone can offer [cf Keith Stanovich’s ‘Measuring Print exposure’ in Progress in Understanding Reading].
Interestingly, the authors also point out that ‘[b]ooks with short words allow children to register all the letters in a word during a fixation’ (p.58), a contention which would lend strong support to the use, in the beginning stages of learning to read, of decodable readers containing short words.
However, the central message conveyed by Ashby and Rayner I will leave in their words:
“Instruction that develops a child’s ability to read unfamiliar words accurately (and familiar words quickly) will, by definition, build the efficient word-recognition processes that are necessary for text comprehension.”
Ashby, J. & Rayner, K., ‘Literacy Development: Insights from Research on Skilled Reading’, in Dickinson, D.K and Neuman, S., Eds, (2006), Handbook of Early Literacy Research, Vol 2, London, Guilford Press, pp 52-63
McGuinness, D., (2004), Early Reading Instruction, London, The MIT Press.
Related
I disagree that decoding using SP builds high quality representations of word specific spellings. In order for a word specific spelling to be known the writer must know which spellings of which sounds are used in the particular word. In order for this to happen they need to internalise the whole sequence of letters for the word, not simply know which letter *might* spell one of its constituent phonemes or, in reading, which sound. *might* be represented by the graphemes. it's not hugely helpful to know that a_e, ea, ay etc spell the sound /ae/ if you don't know which word used which variation. So most people seem to use a sense of 'what looks right' thereby matching the whole word written with a whole word in the lexicon. Some words are very counter-intuitive: 'yacht', 'viscount'….
In order for cognitive resources to be fully allocated to higher order reading, I would agree, word recognition has to be automatic. The misconception you are working with here is that decoding through SP is the only route to automacity. Ths is clearly untrue as SP as taught in England at present is not the same as previous successful approaches to reading instruction. This point also applies to your last 2 claims about reading comprehension and vocabulary extension. These are achieved through automatic reading with *understanding* not through knowledge of the alphabetic 'code', useful though awareness of phonic correspondences might be.
Self-teaching is an integral part of reading progress, as shown by David Share. This seems to happen when children have built up a lexicon of words and can generalise from them to new words encountered. It would be wrong to think, therefore, that SP is a powerful tool to aid this. The powerful tool is word knowledge however acquired. The original build up of words may occur through SP, but not necessarily, particularly if SP concentrates on nonword exercises and decoding isolated graphemes. This could well slow down the process of recognising and reading real words with *understanding* – something essential to reading skill and indeed, motivation to read.
Thank you for your comment, Ruth.
The answer to the reason why a writer would know which spelling to use is from exposure. Exposure comes from plenty of reading practice but reading practice alone doesn't guarantee that readers notice which particular spelling of a sound needs to be used in any specific word. First, the (beginning) reader needs to be taught the variations in spelling sounds – always in the context of real words! Otherwise, how would they ever get to know what the possibilities are! Multiple opportunities to practise working with the various representations of any particular sound does build high quality representations of the spellings of that sound because there are common patterns in the language and the brain is quick to reconise these patterns – an argument made very well by Diane McGuinness in her Early Reading Instruction.
Of course, a beginning reader wouldn't know which spelling of the sound /ae/ to use until they had had plenty of practice. This practice is gained by being taught the possibilities for spelling /ae/, as well as from the exposure they get from reading.
As for your persistent claim that readers skip words, I think Bonnie Macmillan's summary of the evidence in Why Schoolchildren Can't Read from a number of studies (Just and Carpenter, 1987; Patterson and Colheart, 1987; Perfetti, 1995) makes clear that 'the skilful reader attends to every letter in each word from left to right order'.
You obviously, also accept the Dual-Route approach which contends that all words eventually become sight words, going straight to meaning without the phonological step. I think I've made clear that I do not agree with this view. As I keep on insisting, adult, skilful readers are not aware of what they are doing when they are reading. This is because they are so well practised that it happens under the level of their conscious attention. What can be plainly seen in beginning readers is the process slowed down and we can see how it gets faster and faster with the proper kind of practice (i.e. SP or L.P.) until it becomes automatic.
Thank you for your comments. This kind of argument can go on interminably for which reason I shan't be publishing any more of your comments.