SALT - Colourful Semantics

[StarkAndWitchesWillFindYou]

Any ideas? Thoughts?

Is there any chance this could be appropriate for ds?

Stark, if I remember correctly, your ds uses quite long chunks of echoed language which he combines to make greater meanings but although others can decode it, it is quite disordered e.g.

why do you think mummy can I go the kitchen yes?

(paraphrasing)

If so, possibly. It could be used as a teaching tool to help him analyse the chunks a bit better BUT I would be careful to make it very functional and relevant to him. It's a visual strategy and a prompt, so you would need to think about how that would fade.

Text fading is a possibility here e.g. teaching the entire sentence structure with colours and then covering aspects (backward chaining) until how to put the structure together is learned off by heart. It definitely has potential if adequately tweaked/individualised.

It's been a while since I did this sort of thing so forgive me if I sound a bit waffly, I would need some examples of what you would like it to achieve to give it a go. And I forget some of the colours as they are slightly different for Colourful Semantics and Narrative colour coding systems.

(arses in disguise - and not because moondog told me to change my name! Namechanged for sahm thread as I was afraid it might go haywire but it was all very civil in the end)

Sorry, I should have made it clear in my OP.

This is what his school SLT tells me she is about to start doing with him.

It doesn't sound great from what you say. I mean it 'could' work, but only done in a kind of ABA context which I doubt the SLT (in particular this one) would have a clue about.

Oh , not another thing I have to criticise. I'm fed up of it.

Is there anything like it that would be better?

I mean, I can't complain really if it is the best there is for this kind of issue.

I'm so out of my depth here .

The SALT spoke to me on the phone today at last with enthusiasm about this 'solution' to what I was telling her ther problems were that I wanted to be worked on. Don't want her to give up for not being able to do anything right.

She has told me that she will either a)copy her notes each session to send to me or b)write other notes if I tell her what I want and that I have an open invitation to any session without notice and she'd particularly welcome attendance at a quarter to a third for observe and take ideas for home.

She also said she apologises (bet this won't be on any records) that the original targets were so cautious and she believes we can up the pace as ds learns so fast.

So, you know, - we have the makings of a good working relationship. But utterly pointless if ds doesn't progress much.

Thanks for responding though, and for what it is worth I like your name and don't think you should change it.

and for anyone else that wants some pretty good free pictures

Well, what I would be inclined to do if I were you would be to get the info the SLT comes up with and adapt it to be ultra-SMART and we'll try to work out some sort of tracking..

I know it's not ideal but I think there is a grain of usefulness there somewhere which is a starting point. I understand the rationale for it and I don't think it's a particularly poor one... but that won't help with teaching methodology.

If you want to give me an idea of some sentences he uses/where the issues are, I will have a stab at making it more useful and then maybe someone with better ABA skills (aka moondog) could clean it up? Then you are prepared..

Sorry, cross posted with you there. I have to go have a driving lesson now (groan) but will be back later.

Okay. So, Shall I try to get the pictures from her and ABA ds into learning them?

I mean one fundamental issue I have with these types of visual things is that they mean nothing to ds. May as well just teach him to sight read.

He'll never make a sentence because he just doesn't look at things that don't interest him unless you know how to get him to.

I see there is no evidence for the approach , but did come across a shapes system that seemed quite good, although it was done with real words etc.

I really do appreciate your input here. It's facinating stuff, but really scary because I really don't know what I'm doing/arguing for.

Can you believe that only last June ds was being introduced to PECSs, and initiating communication. THAT I could do!

Oh, one last thing, if you or any other poster reading can help with.

Any recommendations for a book on basic grammar construction.

I had NO education on this and lost all ability to learn a foreign language because of it. .

Examples of ds' language:

ds 'why do you want the can I have the mummy go in the kitchen please?'

me 'I don't know ds, why?'

ds 'because the do you want the dinner'

ds 'mummy, shall we want to go upstairs and put the covers?'

me 'okay, I'll go upstairs and get the covers'

___

ds 'shall we need to sleeping'

me 'yes, it is time to go to sleep'

___

ds 'we're not bumping lights'

me 'eh?'

Dh needs the computer so no time to write a proper response tonight but I think text fading is the way to go here.

You will need to use ABA approaches as you say: however, I believe there is potentially some merit to teaching the colour coding here.

There is a resource here that shows really what needed to happen between initiating for single objects and the multiword utterances you have now: www.tulareselpa.org/Autism/Downloads.shtm

e.g. three meaning relationships e.g. Jamie eats fruit loops where one word/phrase corresponds to one bit of meaning.

What your ds appears to be doing is remembering chunks of language associated with a specific context and assembling them in this way - it's just that the context/assembling is somewhat idiosyncratic to him. The best example of this is the "why"//"because the do you want the dinner?" example. So "because" is associated with "why" but getting what he wants is associated with "do you want the" and then he is saying "the dinner" as he could probably (I'm guessing here, maybe not) also change this element to be
"because the do you want the breakfast".

So it is two to three meanings he is using here with seven words, either:

[because the] [do you want the] [dinner]
or
[because the] [do you want the dinner]

The longer elements don't quite fit together in the way they should for this reason.

Nothing wrong with using a lot of words corresponding to each meaning, it's how we build up complex meaning... except of course he is breaking the rules for how we do this so although you understand his meaning the grammar is all over the place and, where it is truly idiosyncratic, even you can't work it out.

However, can I say that the first example, is a really positive sign that he is working hard at the rules.. I would expect he has had a few "goes" at this. The other great positive is that he is really aware that he needs to initiate with you so is using markers for this..

He may need some help with getting it straight though (as I guess you've figured).

Can he sight read at all? I know you said it to highlight the issue with the intervention but might be a very valuable tool here. If he could sight read, I would be inclined to
deal with these in the moment using pen and paper e.g.

I think in the first example, he is asking can he go in the kitchen - not why at all..

so

if he says:

'why do you want the can I have the mummy go in the kitchen please?"

and he could read,

I would write

"Can I go in the kitchen please?" and get him to read it and then say "Why?" and then
when he responds with
"because the do you want the dinner"
write
"because I want dinner" Then, having done this, I would start to tackle these questions using communication temptations, changing the place/what he wants e.g.
"Can I go in the bathroom?" "Why?" "because I want the [insert reinforcer]"
Then you fade the text so..
Can I go in the kitchen please
Can I go in the kitchen
Can I go in
Can I go
Can I
Can...
eventually fading out all text prompts

I can blend this with the colourful semantics too if you want.. I would be thinking of using the sentence stems to teach different places with the colour.. e.g. kitchen/bathroom/bedroom etc.

I just need to think some more about it..

Does any of this make sense?

So next question......

Why is it I can get more help and sense specific to ds' needs from a random person on an internet forum than I can from the 18.5 additional support hours on top of ds' 15 hours of nursery provision?

(sorry, not saying you're random, but you know what I mean)

That is an amazingly helpful post, thank you.

DS meant that he wanted me to go into the kitchen and start cooking his dinner I think, but still, your example was very clear about how we would work on it.

He can't sight read. I had this wierd thing about allowing his decoding to shoot far ahead of his comprehension, but I bet he could learn fairly quickly.

Also, teaching him to read as a part of this strategy could help his comprehension.

I'm just a bit scared of it all as in the past I have been certain that whatever mistakes I make, doing it my way will mean progress for ds at a faster rate than any other, but this is different. I don't want to make the problem worse. It's much more complicated than motivating him to speak in the first place, using names to address or simply learning a million nouns.

You are right to be concerned about the decoding/comprehension divide, but if you do it carefully and make sure that what he is learning is linked to meaning for him, it should be okay.

As he can't read, I think I might at least have a go at blending the text fading and colourful semantics, I think I would be inclined to take whatever we were targeting and then add a colour coded symbol to it with a picture, say, of the kitchen (as in the resource pack) - just for the key words, as in Pecs.

So if his meaning was more like
Mummy go in the kitchen
Mummy make dinner

I would do it almost PECS style with different coloured symbols for
[Mummy] [go] and [in the kitchen] (breaking down prepositions later, add a please now if you really want!) depending on meaning
so this would be
Agent action location
(or in simpler terms, who what doing where).

You can still backward chain, so start by giving him a sentence strip with these things on and remove elements. If this proves too hard you can alternatively work at getting him to assemble the sentence from different choices or use two as a forced choice e.g. mummy make the dinner or daddy go to the shop etc. It should teach him that words go in specific places in relationship to other words to make meaning (theoretically).

The reason I would consider this vs just text if he's not reading now is it allows you to expose him to text in meaningful context, thereby working on decoding and comp together plus it has applications for other language development activities e.g. sorting and categorise people/places/objects in the early stages, responding appropriately to questions and/or put adjectives/adverbs in the right place.

It won't just happen but it's building in supports that make sense across a lot of different targets yet still planning to fade the prompt. It also means that if he is having, say, colourful semantic work at school that the work you do may enable him to gain more from that.

And as you know about PECS you do also know what to do.

Don't worry too much about messing it up. I have met quite a few verbal kids with ASD on ABA programmes who have gone through this phase and have developed age appropriate concrete language: it may actually be as much a quirk of his learning style, but I guess it's not worth messing about with!

(sorry last line makes no sense - I mean it's not worth just leaving!)

Thank you. I really do appreciate your time and thoughts on this.

I have some things to think about and figure out the practicalities.

I might try hard and get a copy of the resources that the school SLT is using and observe a couple of sessions so I can speed up the process at home. DS will have to learn the meaning of the visuals. He has had very little practice.

PECS went a little bit wrong for us, in that ds already knew a billion nouns when we started but he would say them to the ceiling. So we got him to hand us the picture. He learned quickly that handing us the picture was the activity that got the reward so handed us any old picture and said the desired object accurately. We saw no reason to teach him to discriminate visually since he was doing it verbally iyswim.

Still, I have those lovely home-made PECS boards in a drawer that I can get out, with the sentence strips etc. that we never used.

Almost giddy at the thought of getting them out again.

So am I reading it correctly that the PECS taught him the initiation side of things quite well? It does sound that he understands and recognises "initiation" types of sentences and is

I am in two minds about teaching the visuals. On the one hand, it is teaching a prompt and one that is not exactly functional with reference to the rest of the population. On the other, teaching him to pay attention to visuals might just be a conduit to teaching him a bit more about how individual chunks of sentences work in communication. It's hard to know without knowing your son whether working at teaching this prompt will work or not. Even then.. Did you say you did PECS yourself? Do you have an ABA team.. sorry, can't quite piece things together here.

It's interesting what he did with the PECS, what that's about? Do you think the visuals simply didn't mean much to him or was his way of doing it his way, making sense of it/controlling it?

The thing about visuals, as I've probably said before, is that they are often introduced as prompts because they are easier to manipulate and longer lasting than verbal prompts (which are the hardest to fade). So all those old classics - the first/then board, the visual timetable etc - are often used on ABA programmes but the difference is that there is a clear vision for why they are added and how they will be removed. They shouldn't be a 'forever' thing but.. and I suppose, this is the but.. some crutches you need a bit longer than others. I suppose my instinctive reaction to the colour coding is that it helps to be clear about what type of sentences you are working on and do it somewhat systematically vs haphazardly.

The other thing that strikes me today is that he is trying to use language to control your behaviour (a good thing, of course, that he realises he needs to talk to you to do this) so he might get a kick out of a manding game: what we used to call the Robot Game. So you could make up your sentences like "mummy go to the kitchen", "daddy go to the bathroom" and put them in a pot and when he pulls out one, act on it. You could throw in some wild cards, like "touch your nose mummy" or "pretend to be an aeroplane" (obviously, here, choosing something reinforcing to him).

Hi thanks.

DS DID discriminate for a little bit, but what happened was his little sister would remove some of the pictures and, rather than be without his new found voice he would simply use the card that was usually next to the one he wanted but say the one he wanted. Once that was reinforced he realised he could use any picture but as long as he said what he wanted he would get it.

(This might sound slap-dash but we had done plenty of work on prefered reinforcers and did know 100% that what he was verbally requesting WAS what he wanted rather than the picture or anything else)

DS' trouble with visuals symbols (I think) is that the more specific the picture the more difficulty he has with generalising it. So if you show him a picture of a toy train, it doesn't represent HIS toy train, or any other toy trains.

The more abstract the picture/symbol, the more able he is to use it to represent many items in the same category. I seem to remember that we used a star to represent reward, and a question mark to represent his choice of activity on something else we were working on.

I get the impression that if he has to make up sentences with a picture of a man, and then is presented with a photograph of a man, he won't realise that the symbol of the man represents the man in the picture, and if he does eventually learn it, he won't realise that that same symbol also represents a different man in a different picture that he is expected to make a sentence about. sorry does that make sense?

I think it the picture WASN'T a man, he'd have a better chance as this visual wouldn't interfere or distract or confuse. So symbolic representation in the form of the written word might be better, although there will need to be prep work while we teach him to read the words first.

BUT, if I said this they'd say I was nuts or pushy or deluded etc.

But I really am worried about the way the sentence is represented.

Absolutely NO issues about the overall strategy though. I think colours for the component parts would be very helpful.

Does that make any sense at all?

I don't think anyone should say you were nuts for that! It makes perfect sense. It's one of the key things I remember from my very, very earliest training day on autism, long before I was an SLT: I think the training mentioned something about how Temple Grandin said that each cat she met forced her to relearn and relabel the word "cat" to include the new features of that particular "cat". I have vague recollections of doing a PT programme about this back in the day - matching lots of 2d pics of an item to a 3d object, then 2d-2d then 2d-label then randomly rotating the stimuli. Moondog would be the lady to seek advice from on this.

It is a tricky one, though..

I think the immediate way around it is to approach it as you would with PECS, so to use a visual that is crystal clear (white background, no shadows) with functional meaning. So "mummy" or "daddy" or his sister to begin with, with very clear visuals, and just keep it so static (as with text) that he learns the association over time. That is much easier said than done of course. Text makes more sense but as he can't read you need to avoid playing into the disjuncture between the symbol (the text, picture, language, anything representation) and what it represents (the person, activity, location) which is his natural default.

Sorry, it would have been
3d-3d
2d-3d
2d-2d
2d-text

Sorry. Only just realised you have posted additional stuff.

Thank you. I have some useful sound arguments for my meeting with the SALT tomorrow, who has surprisingly turned out more supportive than I was led to believe she was.

I observed a session today. At first I was a bit disappointed because I thought ds was being allowed to get away with pretend helplessness and not performing as well as he can. I also thought it was little wonder, as the pace was SO slow and the instructions a bit varied.

Then I got a verbal feedback report and was even more disappointed because the SLT had told me that she was amazed as that had been the best he had performed and the 1:1 agreed. (perhaps it was because I was there and he knows I know etc.)

BUT, she showed me the colourful semantics and she just usess the coloured SHAPES for ds to point to as he says each meaning. No wierd open-to-interpretation visuals. She also explained some of ds' difficulties which I felt was accurate if not as profound as she was suggesting. The most important thing though was that I could see how what she was doing 'could' help him order his language better.

She also agreed to give me the resources so I can teach him the concept myself as I suggested this might be good for me to do as I feel I can do this easily and quickly. She didn't argue but agreed.

Overall, I think that she has analysed his problems quite well. Found a solution that could work. Has absolutely no idea how to get ds to learn it but has allowed me to have a go at addressing this first.

All and all I think it went well.

Ha I am following you around the board .

That's good at least! As there were obviously other hiccups. SALTs don't get any training in instructional methodology really so it's not surprising she doesn't know how to teach it. We are better assessors/evaluators than teachers on the whole.

We need discrete trial training. Or P/T. I could really do with revisiting it myself, working out how to do it with unruly teens

'not performing as well as he can. I also thought it was little wonder, as the pace was SO slow and the instructions a bit varied.'

Exactly!!!!

'SALTs don't get any training in instructional methodology really so it's not surprising she doesn't know how to teach it. We are better assessors/evaluators than teachers on the whole.

We need discrete trial training. Or P/T.'

Yes!!!

Hence this being more or less my entire mission at work.

Watching your thread with interest! The link is excellent too...thankyou. :)