Eye gaze while listening and speaking

[PeterPan]

https://pubs.asha.org/doi/10.1044/1092-4388%282005/099%29  Came across this study that gives % for typically developing. It's saying in a 3 minute conversation and TD would have eye gaze 62% of the time while listening and 43% of the time while speaking. Then it contrasts with numbers that are slightly higher for TBI. There are probably numbers out there for ASD, but it makes me think that if we force the issue and try too hard, we'd actually push the % up abnormally high. 

I'm finding all kinds of other neat things too doing this survey of narrative langauge and autism on ASHAWire. Fell into some rabbit holes on gesture, etc. and haven't quite sorted them out.

[PeterPan]

https://pubs.asha.org/doi/10.1044/persp1.SIG2.66  So much for the thought of ASD as some kind of right-brain strength. The figurative language deficits are a right hemisphere issue.

[PeterPan]

https://pubs.asha.org/doi/10.1044/2020_LSHSS-20-00110  Rich vocabulary instruction makes no difference, the vocab wasn't the problem, move on and teach your kids where they really are I guess.

[Lecka]

I have definitely heard that it’s important to watch out for kids being told to make eye contact and then coming across as *staring* and making people feel uncomfortable.

 

[PeterPan]

https://pubs.asha.org/doi/pdf/10.1044/cicsd_34_S_44  Planning to read this next. It's a flowchart for analyzing syntactical structures in a speech sample. Targeted at younger kids but still applicable.

 

[PeterPan]

https://pubs.asha.org/doi/10.1044/1092-4388%282005/031%29  Long term outcomes for late talkers.

[PeterPan]

https://pubs.asha.org/doi/pdf/10.1044/2020_AJSLP-19-00090  Gibberish in which they admit what they tried to do didn't work.

I swear, any time I need to feel confident that my crazy ideas could have as much validity as anyone's, I just need to read the literature.

[PeterPan]

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7225018/  Don't get too excited, but Balthazar published an article in 2020 comparing Shade Coding, MetaTaal, and her Complex Sentence Intervention. 

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7225018/pdf/LSHSS-51-226.pdf  prettier link as pdf

She also explores how procedural deficits affect DLD and why you need explicit, metalinguistic instruction, not only implicit.

Edited July 7, 2021 by PeterPan

[PeterPan]

https://pubs.asha.org/doi/10.1044/2020_JSLHR-20-00135  Trying to make heads or tails of this. It looks like it's saying working memory compensates for procedural and declarative memory weaknesses, yes?

[Cake and Pi]

On 7/6/2021 at 9:11 PM, PeterPan said:

https://pubs.asha.org/doi/10.1044/2020_JSLHR-20-00135  Trying to make heads or tails of this. It looks like it's saying working memory compensates for procedural and declarative memory weaknesses, yes?

I read it as procedural and declarative verbal memory may *appear* impaired in kids with DLD, but they're actually consistent with nonverbal IQ and working memory, which were presumably lower in this group than the TD control group.

[Cake and Pi]

Brain imaging supports the notion that eye contact may be distressing for autistic individuals.
https://www.nature.com/articles/s41598-017-03378-5

[PeterPan]

7 hours ago, Cake and Pi said:

Brain imaging supports the notion that eye contact may be distressing for autistic individuals.
https://www.nature.com/articles/s41598-017-03378-5

And yet we have all kinds of social thinking and conversation instructional people forcing eye contact, explicitly demanding eye contact. So I thought it was really ironic that the data was showing even TD people don't have some kind of mythical high level of eye contact. And doing it too much becomes conspicuous too.

7 hours ago, Cake and Pi said:

I read it as procedural and declarative verbal memory may *appear* impaired in kids with DLD, but they're actually consistent with nonverbal IQ and working memory, which were presumably lower in this group than the TD control group.

Yeah I found multiple studies where they were trying to draw correlation between the nonverbal IQ and various language features, and I couldn't figure out why. (I think I get why, they're saying kids are just flat low, but it still doesn't make sense to me. I also don't have serious nonverbal IQ scores vs. WISC type IQ scores to make any discussion of nonverbal IQ in my ds make sense.) In the conclusion of this article they go back to focusing on the working memory.

There's a whole strand of thought right now that interventions (for syntax, whatever) should be *visual* because you're supporting for procedural learning weaknesses and tapping into a strength. This is why they're studying it as SLPs, because they are using their conclusions to extrapolate and drive intervention. But to me, the whole visual instruction for syntax thing is iffy. I've looked at the programs and they don't make sense to me. I've done diagramming (which isn't really SO different from shape coding) with ds and zero click.

Now I'll do anything it takes with ds. I'm just saying I'm questioning how they're thinking through this. If working memory compensates for procedural memory or correlates to procedural memory, then upping working memory would allow the student to go back to a more naturalistic, native way of working on syntax (ie auditory). I'm also not sure I agree with the idea that paper bugs, legos, and colorful markers around words are the ultimate visualization. I'm happy to use them, just not yet convinced. Numbers move for me for math, back and forth in equations, so that makes sense. And written math is the crutch for working memory that can't hold everything. So are we saying that written language is the crutch for working memory that can't hold all the oral language? 

My theory is they aren't getting enough dosing (enough sessions, enough therapy time) with the material they've already taught to drive the new skill into narrative to where they can use the new skill to build working memory. So then they'd be constantly going forward without enough working memory to do the next thing via auditory and would have to use visual. Also here on the boards we talk about WM as being very important while in the therapy world it's surprising how much they dance around it. They definitely do that with RAN/RAS, with a few brave souls asking why they test for it if you can't intervene and make a difference and others saying of course you can and should work on it. My own small experience is it makes a radical difference. So how much are they poo-pooing WM, not building it up, and therefore falling to these abnormal, highly visual methods? 

Speech is not visual and speech is not multiple choice. The visual of speech is *visualization* and to get visualization we need enough working memory that we can hold the thoughts in the head till we get the picture. 

Ultimately the weakness of the system is the lack of time they can do the intervention. I was thinking about this. 

The procedural deficit hypothesis claims that children with DLD are impaired at grammatical rule learning because of a deficit in their procedural memory system. To compensate, they rely on their declarative memory system.

 

Ok, I think I figured something out. This is what I'm taking issue with. Not completely disagreeing, because I agree implicit learning didn't work and that they fall back to scripting. But WHAT were they supposed to learn implicitly? The bits of language, parts to whole (pitch, phonemes, syllables, words, morphemes, stress, phrases, syntax, etc.). And so far, for going back and learning the bits (which we've done with the APD materials) has a visual ever been key? We used a GESTURE to work on pitch and the rest was auditory. GESTURE. 

I spent some time going through ASHAWire on gestures, but I made my search terms pretty narrow. (gesture syntax clause) There's enough research that they have gesture types (deictic=pointing, represntational like a shape, conventional=accepted by culture like nodding, and beat=emphasizing aspects of speech). Then they had functions, extending (adding extra info not giving with the speech, like making the shape of the cat's tail when you say "the cat had a tail like that") vs. redundant (refinforcing the spoken, so "the cat had a curly tail" while making the curly tail). https://www.hanen.org/SiteAssets/Articles---Printer-Friendly/Research-in-your-Daily-Work/Printer-Friendly---Closer-Look-at-Gestures.aspx  And the research showed just as much or even more gesturing going on but more extending instead of redundant, compensating for the vagueness of the DLD speech. 

So then I have questions in my own mind about the role of *gesture* as the visual to create meaning and understanding in syntax. How did you learn clauses, relative pronouns, referents, etc. in a foreign language? Me, I used a lot of gestures. I didn't sit there with legos or slips of paper. I used gestures, and I thought about the constructions a long time till I had a mental picture. 

Now I need to remember some more or find my notebooks. That was a LONG time ago, lol. I *may* have done some little figure drawings representationally to improve my visualization of the structures. I either had Barbies handy or I mentally used them. That would have been the summer of 1993, a long time ago, lol. I don't think that written analysis was pivotal. It was always visualization, because visualization is the strength.

I don't know. I seem to spend a lot of time questioning phds, hahaha. But I figure, if they are struggling to get carryover to narrative, there's room for questioning. And in some of those studies, they were getting only 1 of 3 constructions to stick at all after 9 weeks of intervention, weak carryover to narrative. There's room for questioning.

Right now it's very fashionable in the SLP community to decry APD. And I agree, APD as a construct is silly and should not be happening, because it's clearly a language processing problem, not a hearing problem. It should be the SLPs who are treating it. So what if their syntax intervention is the final stage of intervention for an APD problem that they spent their entire careers denying? What if they never fully grappled with the consequences of the actual implicit learning deficit, that the dc wasn't learning ANY of the bits of language, parts to whole? Or they think they did but they didn't really sit down and think what they learned from the prior intervention that worked? Or it didn't work?

I think for most kids, there's not really this assumption of much hope. More like bits of hope, bits of breakthrough. Me, I'm not working with those assumptions. I can devote (and often do) 2-3 hours a day to language. I don't have their time limitations and I don't have the system limitations that tell them to push kids through and work on the goals (even if they're way high). So what if working on PITCH is filling in a developmental step that would help the dc discriminate bits of speech to then, several steps down the road, be able to understand the syntax in a more natural way? Still needing explicit learning, but explicit instruction starting with the bits. 

Me, I worry I'll dither around in these weeds so long, doing what I think is filling in holes, that I'll run out of time to do the higher level stuff. But I'm just saying what I see and am thinking about. It's like SKILL, where they put it out, the bigwig SLPs are like oh yeah, then you read the data and realize it didn't stick for autism, it was spotty. And I think some of these SLP researchers KNOW it's really hard to get it to stick (with autism) and don't know WHY. That's what they really want to know. 

https://asha.figshare.com/articles/dataset/Expecting_questions_modulates_effort_Pupillometry_Chapman_Hallowell_2020_/13480368/1  I thought this study was really interesting. She found with pupillometry that expecting questions increases effort. Think about all the implications of that. Right now I'm using Harkins' book on conjunctions, and she gives you a narrative to read to the dc during which he needs to *listen* for the target structure (a conjunction) and be prepared to answer a question. And the questions are not necessarily about the target either! Like it might be about the first part of the sentence or the overall meaning. 

I don't think that a physical visual (legos, shape coding, whatever) would somehow make that go better. Working memory makes it go better, sure. And we had already done the explicit instruction by the time we get to this stage of listening for the target to answer questions. And because it's in a narrative, he's visualizing. I use her book out of order btw, because I don't find her instructional order within the chapters builds logically.

So that's why I'm saying I think WM is the compensator and visualization, not physical visuals, are the goal. At least that's what I'm working on in my mind. I'm happy to have those tools of bugs, shape coding, legos, and all this other stuff they're doing or even completely change my mind. I'm not yet sure it's what is essential.

Edited July 8, 2021 by PeterPan

[PeterPan]

So maybe my paper someday will be Barbie Loves Syntax.  

In all seriousness, I think I have some cute figure dolls back from my dd's day. I could dig them out. They'd be a way better visualization tool than paper bugs, mercy. 

Ukrainetz 2018 Sketch and Speak would be a visual way to approach syntax. Still, Barbies would be good, haha. Or maybe playmobil. 

Total aside, but I keep thinking about this article. http://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC6802893&blobtype=pdf  It's not really on point, since it's looking at simpler syntax. However if you haven't seen different kinds of diagramming it's an example. We did this in an advanced grammar class I took in college, and it's sort of a linguistics/TESOL piece that informs how I think of grammar. To me it's very fluid and flexible. So I get why they want to use the visual, but still I'm not sure visual vs. visualization makes sense.

I need to read more about what Balthazar was using and the legos. In Barton, lIPS, etc. you use place holders for language. That would take you back to that working memory thing. I wouldn't have an issue on the placeholder being a tile vs. barbie. You're just changing the level of abstraction. But it seems to me if explicit instruction with visualization is the goal, barbie is ideal. That's how I learned physics haha. Barbie and Ken, attraction. 

Edited July 8, 2021 by PeterPan

[Cake and Pi]

On 7/8/2021 at 5:37 AM, PeterPan said:

And yet we have all kinds of social thinking and conversation instructional people forcing eye contact, explicitly demanding eye contact. So I thought it was really ironic that the data was showing even TD people don't have some kind of mythical high level of eye contact. And doing it too much becomes conspicuous too.

Yup. It might be good to remove eye contact from the goals list altogether. Always seemed like a waste of time to me anyway.

On 7/8/2021 at 5:51 AM, PeterPan said:

So maybe my paper someday will be Barbie Loves Syntax.

Lol! I'd like to see that.

 

On 7/8/2021 at 5:37 AM, PeterPan said:

I agree implicit learning didn't work and that they fall back to scripting.

The longer I'm on this journey, the more I think scripting is perhaps an acceptable end-point. I mean, it works. 

Right now I'm teaching syntax to DS#4 basically through rote memorization -- and it's WORKING! It's not a very efficient process, but it's better than anything else I've tried so far. I'm using the language strand from Reading Mastery, which I *think* may be equivalent to the separate SRA Language for Learning program, but I haven't quite figured out how the different DI programs overlap or don't. It's really just drilling short scripts with sentence frames and errorless learning.

For example...
Teacher: This is a clock. What's this?
Kid: A clock.
Teacher: Yes, a clock. Say the whole thing.
Kid: This is a clock.
Teacher: This is a pencil. What's this?
Kid: A pencil.
Teacher: Yes, a pencil. Say the whole thing.
Kid: This is a pencil.
Teacher: This is a book. What's this?
Kid: A book.
Teacher: Yes, a book. Say the whole thing.
Kid: This is a book....

Rinse, repeat like 5 bajillion times per lesson with various nous, then do the same thing again in tomorrow's lesson and the day after and the day after and the day after and after for a couple of months.... Tada! kid can now say "This is a _____" about things he has a label for. 🙂

[Lecka]

That is not scripting as far as “autism scripting.”

That is quality speech therapy.  

[Lecka]

Separately — “functional scripting” is totally a thing 🙂. And can totally be functional 🙂

[Lecka]

What you are saying — I think I have heard called a “carrier phrase.”

https://www.speechbuddy.com/blog/speech-therapy-techniques/when-and-how-to-use-carrier-phrases-in-speech-therapy/

[Lecka]

Autism seriously has its own lexicon where words do not mean what they normally mean, it is often very odd!  
 

Edit:  I mean scripting wrt autism often means something is totally memorized but even like reciting something from tv.  

Edited July 15, 2021 by Lecka

[Lecka]

This isn’t even an example of scripting, but I used to hold up a bag of bread and say “do you want a peanut butter and jelly sandwich?”

For a while my son was calling bread, you guessed it, “do you want a peanut butter and jelly sandwich.”

I was literally holding it up and labeling it that way (even though I meant for the bread to represent a sandwich).

Scripting can (among other things) be like that where there is some language that is remembered and it’s known it “goes with” something, but it’s not being used as individual words meaning things.  

 

[Cake and Pi]

3 hours ago, Lecka said:

Autism seriously has its own lexicon where words do not mean what they normally mean, it is often very odd!  
 

Edit:  I mean scripting wrt autism often means something is totally memorized but even like reciting something from tv.  

I was told that scripting and delayed echolalia were the same thing but that scripting is the more commonly used term in behavioral sciences (such as by an ABA) and delayed echolalia is more commonly used by SLPs and self-advocates? My understanding was that scripting/delayed echolalia could be either functional and to communicate or non-functional, like a stim, and could be a recitation of just about anything: lines from TV shows, things family or friends have said, things electronic toys say, etc. Is that different than what you're saying?

4 hours ago, Lecka said:

This isn’t even an example of scripting, but I used to hold up a bag of bread and say “do you want a peanut butter and jelly sandwich?”

For a while my son was calling bread, you guessed it, “do you want a peanut butter and jelly sandwich.”

And I was told that this kind of thing *is* scripting/ delayed echolalia. I wonder if these terms have regional differences in use and application. 

5 hours ago, Lecka said:

That is not scripting as far as “autism scripting.”

That is quality speech therapy.  

I'm glad it sounds quality. Sometimes I second guess what I'm doing with him. 😄

I might not have expressed my thoughts clearly. The first bit of my post was my immediate response to the discussion on syntax. If my kid is repeating verbatim a sentence or phrase that works to get his point across, I find that preferable to the word salad or non-word vocalizations (that I often can't understand) that we might get otherwise. The second part of my post, which I didn't do a very good job of transitioning to, was about my attempt at home-cooked language therapy and trying to harnesses the tendency to echo/scrip to end up with more language with correct syntax. 

[Lecka]

There is a word I am blanking on for a label (like what an item is called) when the child has learned the wrong label, but is consistently using the label to refer to a specific item.  
 

I could see it also being scripting!

 

But there is something like — I can’t remember — when a specific word is being used for a specific thing, it’s just not the right word.

 

My son also had a phrase he used to ask for any DVD, but it was something like “do you want to watch Dora or Diego?”  And that meant “I want to watch a DVD.”  
 

A lot of scripting is more random than that or — it would make sense if you happened to know where it came from.

If I am literally holding up a DVD, and instead of saying “DVD” I say “do you want to watch Dora or Diego,” then it’s like I am teaching him the wrong label.

 

But I can see how it would still be called scripting.

 

But it’s not — like other scripting.  There was a while my son used to say “I just knocked down a building” at totally random times and nobody knew what he meant, where it came from, if he meant something, if he just liked how it sounded (maybe it was said with a lot of excitement).  That I thought was 100% scripting.  
 

The thing is that some people do talk about “teaching scripts” that are going to be familiar, common things to say — but that is different from “scripting.”  Sigh. 

 

Teaching scripts can be really appropriate.  
 

Edit:  could it have been “mislabeling?”

Edited July 16, 2021 by Lecka

[Lecka]

Anyway — I guess I would say, for my son saying “I just knocked down a building” would definitely be delayed echolalia or scripting.

But if he thinks an item is called the wrong name but always uses the same thing and is clearly being used as a request or a label, totally consistently, I don’t see how that is delayed echolalia *period.*. I could see someone calling it functional scripting.

 

Is there such a thing as functional delayed echolalia?  Or echolalia that is considered meaningful communication?  I have not heard of it.  I have only heard of that with functional scripting.

 

I think of delayed echolalia as pretty definitely being — not intended to communicate anything specific.  Which — I could just not be aware of more nuances there, that are there.  
 

My son never scripted very much though, in the scheme of things.  It was just once in a while.  And I learned to do better at labeling things clearly and that really cut down on the mislabeling!!!!!!  

[PeterPan]

11 hours ago, Cake and Pi said:

And I was told that this kind of thing *is* scripting/ delayed echolalia. I wonder if these terms have regional differences in use and application. 

I think it's more that these camps (intervention, ABA, SLPs) have so many turf wars that they don't talk together very well. 

11 hours ago, Cake and Pi said:

 If my kid is repeating verbatim a sentence or phrase that works to get his point across, I find that preferable to the word salad or non-word vocalizations

I'm probably just pretty touchy about it, and like you say it depends on where you're coming from. My ds had all of his speech motor planned for 10 years, and they were essentially scripting, programming, memorizing the phonological processing and motor planning for every word, phrase, sentence. They didn't work on expressive language, saying it wasn't their thing, but reality was that it needed another SLP to do that, which we couldn't make happen. He was already getting 2 hours a week of SLP and had no autism diagnosis yet and we were burnt out. (driving long ways to therapy, financially, etc.)

So then he hit 10 and was failing the SPELT. At that point, scripting is no longer so cool because you're wanting original language, kwim? 

https://www.northernspeech.com/echolalia-autism/natural-language-acquisition-in-autism-echolalia-to-self-generated-language-level-1/  Two part video course and there's a book. I think she's identifying stages and how you would go through that transition from echolalia to original speech. 

Edited July 16, 2021 by PeterPan

[Lecka]

So to me — the good side of any functional language, even if it’s memorized sentences, is it’s functional, and that is wonderful.

The downsize can be — it’s not necessarily building towards generating new, original sentences.

But a lot of times, memorizing a pattern can start to help with generating new, original sentences that follow that pattern, and then that can be a step towards even more original language.  It really can be a step that way.  

[Rosie_0801]

12 hours ago, Lecka said:

Is there such a thing as functional delayed echolalia?  Or echolalia that is considered meaningful communication?  I have not heard of it.  I have only heard of that with functional scripting.

My kid refused to learn to speak from me, so I had to be pretty careful what she watched on telly.

She'd been watching Beatrix Potter cartoons about Mrs Tiggywinkle and unwanted guests who were eating her out of house and home. Anyway, one day her father was looking for his perpetually misplaced hearing aid and she came bursting into the kitchen from the next room exclaiming "There IS honey in my cupboard!" (whereas the original line was "there is no honey in my cupboard") and sure enough, his hearing aid was in the kitchen cupboard.

I don't know if that's an example of what you mean? It was definitely echolalia in play because she had all the words to say "it's in the cupboard" but she didn't know how to say that yet.

[Lecka]

It is hard to know.  
 

Because there are these stereotypes “autistic kids don’t ever know what they are saying,” and it’s like a radical idea to say they are intentionally communicating.  I HATE that attitude.

The best I have heard is functional communication.

But honestly I think — how cool your daughter sounds with that story!  That is the kind of thing where I feel like the stereotype is going to be “let’s be super-negative and act like this does not actually represent communication.”  Grrrrr.