[senco-forum] Reading

Thanks, Mary, that gives me a different angle to look at!  

I don't do all this 'questioning' to be awkard!  It's just that a) the whole subject  is fascinating and b) other people's responses can be very enlightening/thought provoking!

Philip's point about muscle movements still brings me back to thinking of automaticity and long term, rather than short term, memory.  If the child can talk then the muscle movements must already be mastered, but I suppose that producing them in response to a written stimulus must be more difficult than 'just talking' . And,  the phonemes in speech are coarticulated, rather than separately enunciated.  There must be a degree of complexity involved in the deliberate coarticulation of discrete phonemes required for 'reading' a word.  

I liked the Resnick & Beck approach to blending because I, too, find that some children blend better if they do it incrementally, in the way Resnick & Beck describe.  It does appear to put less strain on working memory (and it stops this dreadful habit of looking away from the word as they try to blend it!).  

I'm not sure if I have found an answer to my original 'question', but it's been interesting.  Thanks.

Maggie 

Mary Kelly <mary.kelly4 at ntlworld.com> wrote:        v\:* {behavior:url(#default#VML);} o\:* {behavior:url(#default#VML);} w\:* {behavior:url(#default#VML);} .shape {behavior:url(#default#VML);}      st1\:*{behavior:url(#default#ieooui) }           Even if the string of letters is on the page in front of the child the letters must still be translated into an internal representation of the “sound” (I think Philip tells us it’s not actually the sound but the pattern of muscle movements needed to articulate the sound). That is, you can’t blend the letters on the page, you can only blend something you can manipulate in your mind. I think competent readers need only glance back at the page to refresh their memory of what’s there because they can perform this translation with automaticity. Severely dyslexic children almost have to go back to the beginning and start again, I think. I think the achievement of automaticity may be one of the reasons RML is apparently so successful –
 all that looking at the grapheme chart and chanting the phonemes? Please understand that all these are working hypotheses that seem to help me in my attempts to unpick what goes on for each child – I don’t speak from any academic/research background in this, though I read a lot around the subject!
  I certainly agree with the quotation you’ve included – I find it often helps to have the child blend the rime into a “chunk” and then add the onset as a “chunk” – two things to blend rather than up to five or six.
  What d’you think of that?  

  Mary

---------------------------------

  From: Maggie Downie [mailto:maizie2004 at yahoo.co.uk] 
 Sent: 15 July 2007 21:58
 To: Mary Kelly; senco-forum at lists.becta.org.uk
 Subject: RE: [senco-forum] Reading - Working Memory

 Mary Kelly <mary.kelly4 at ntlworld.com> wrote:
  My understanding is that (for a reader who is encountering a word that is
 relatively new to him/her) the phoneme (or possibly the motor memory of the
 movement needed to say the phoneme) for the first letter (or letter group)
 within the syllable has to be held in working memory while the next is
 decoded, then that has to be held there and so on, until the syllable is
 assembled and blended. Then the syllable has to be held there while the next
 syllable is decoded, and the next, until all the syllables have been decoded
 and can be assembled into the word.
 In other words, you can only process what you have in working memory. Items
 from long-term memory have to be fetched to working memory for processing.
  What I am speculating about is whether this is actually a necessary part of the decoding/blending process?  I am coming from my observations of my poorest 'blenders'.  These are the children who sound out each grapheme, then, LOOK AWAY and try to blend the sound sequence from memory alone.  This, to me, is analogous to a skilled reader decoding the syllable sequence of an unknown, complex multisyllable word (one of the constituents of shampoo, for example!) and then attempting to reproduce the syllable sequence from memory alone.  Unless one has a very good memory, it can't be done without several 're-readings'  of the word.  

 Is this a 'working memory' task, or am I completely adrift in my understanding?  If it is 'working memory', then my theory is that the 'memorising' of the sound sequence is unnecessary as it can be rehearsed and consolidated by re-reading the sequence.

 A good hunt around the web found this snippet:

 "Resnick and Beck (1976) note that an important
 feature of blending instruction is merging different
 sounds successively—that is, /m/, /me/, /met/. Teachers
 should avoid using sequences in which the merging
 does not occur until each sound has been produced,
 such as /m/, /e/, /t/, /met/. Among the reasons that
 successive blending is preferable is that it avoids the
 need to keep a string of isolated sounds in memory."

 "The Role of Decoding in Learning to Read" Beck & Juel (2002?)

 (http://www.scholastic.com/dodea/Module_2/resources/dodea_m2_pa_roledecod.pdf)

 Maggie

---------------------------------

  Yahoo! Answers - Get better answers from someone who knows. Try it now.

---------------------------------
 Yahoo! Answers - Get better answers from someone who knows. Tryit now.