Mistakes to Avoid when Increasing Language Skills in Children with Autism

Years ago, when I first began teaching Lucas to speak, I didn’t know about the building blocks of language. When Lucas wanted a snack, I taught him to say “open cabinet”. It was only when he began using the phrase “open cabinet” for when he was frustrated and couldn’t open a water bottle that I realized that the phrase had been overgeneralized for him. Today, I’m going to be talking about some of the mistakes to avoid when increasing language skills for toddlers.

In my book The Verbal Behavior Approach, which I based on B.F. Skinner’s book Verbal Behavior (1957), I teach parents and professionals that it is important to start with one-word utterances, and then build onto that foundation with two-word utterances. Recently I attended a lecture given by Dr. Carbone that combined the same basis of B.F. Skinner’s work with Roger Brown’s work. While it is similar to my own work, it does have some differences that I want to go over today.

Dr. Carbone talked a lot about Autoclitics in his lecture. Before teaching language to a child with autism, the primary verbal operants must be in place before autoclitics can be introduced. Autoclitics are secondary verbal operants that modify other words, for example adding the plural endings or the “-ing” and “-ed” endings in verbs. Dr. Carbone explains that a typical child needs 300-400 two word utterances before they can expand into adding past tense and plural endings. However, even getting to two-word utterances in the first place can be tricky.

When I taught Lucas to ask for a snack when he was younger, I taught him five syllables or two words. By jumping ahead in language development, I wasn’t strengthening that foundation that he needed to build on to speak in more complex sentences. Also, when we jump ahead and prompt too many words, we may not be able to understand a child with poor articulation.  I have seen some amazing language growth in children with autism when they followed the approach I laid out in my first book. When we go slow and get language learning right the first time, we can help children with autism reach their fullest potential to communicate as best as they can with the world around them.


  • Why focusing on articulation matters so much when adding in new phrases.
  • The importance of teaching one phrase at a time for early language learners.
  • Common mistakes that parents and professionals make because they’re too focused on adding new words too quickly.

Start making a difference for your child or client with autism or signs of autism through free training!

Attend a FREE Workshop!


MaryBarbera.com/workshop (Sign up for a free workshop online for parents and professionals)
#093: Autism Detection: A Summary of Dr. Ami Klin’s Autism Work
#026: Should You Focus on Talking in Sentences for Children with Autism? | Delayed Speech Development
#078: Autism Success Story with Michelle C.
#056: Building Echoic Control in Children with Autism
Why Teaching Carrier Phrases is a Bad Idea
Teaching Colors to Children with Autism & What Autism Colors to Teach First
Skinner’s Autoclictic presentation by Dr. Carbone at National Autism Conference

Transcript for Podcast Episode: 094
Mistakes to Avoid when Increasing Language Skills in Children with Autism
Hosted by: Dr. Mary Barbera

You're listening to The Turn Autism Around podcast episode number 94. I'm your host, Dr. Mary Barbera, and I have another solo show for you today. We are talking about avoiding common mistakes to teach toddlers to talk or to talk more. I think there's a lot of us who are making critical mistakes. And I'm going to go over the best ways to go from one to two-word utterances and beyond without making these mistakes. So let's get to that.

Hi. Welcome back to another episode of The Turn Autism Around podcast. Today, I'm doing another solo show where we're going to talk about talking and how to get very young children to talk or to talk more.

I think we're making a lot of mistakes and I want to go over those. One of the reasons why I wanted to do this episode is because I attended several lectures through the National Autism Conference at Penn State. Last week I presented on the work of Dr. Ami Klin, who I saw present at the National Autism Conference in person at Penn State in 2019. I saw a couple lectures he provided this year 2020. And I wanted to summarize his research because it's so critical for you all who are listening, parents, professionals, anybody who's listening, especially if you're if you've been listening for a while.

These are really critical pieces of information that I learned some things. So if I learned some things, I want to bring it right to you. So last week, episode number 93 was the work of Dr. Klin. It wasn't an interview and it was a summary of his work and some of the new things that I've been learning. And today it is based on a lecture I heard National Autism Conference with Dr. Vincent Carbone in 2020. We're going to link that in the show notes. So anytime I say episode 93 or episode 94, this is 94. So that is MaryBarbera.dot com/94 will get you to the show notes, which will be the audio, the video and all the links we mentioned throughout. So super valuable resource. There's even transcripts that are free of charge.

So we put a lot of time and effort in to our podcasts and our podcast pages to serve you better. So check out all of those resources. So way back in episode number twenty six, I did a podcast on teaching, talking in sentences and seven mistakes I learned over the years.

So this is going to probably be a little similar, but I'm not going to review those mistakes because they're going to be weaved in here. But I do want to talk about some of the things I learned from this two-and-a-half-hour lecture with Dr. Benson Carbone. His lecture was called Skinner's Autoclitic. Which I'm pretty sure I have never done a video blog or mentioned the autoclitic in a podcast. So it is kind of. So seeing that title Skinner's Autoclitic, for many of you, especially for the parents, you probably have never even heard that term. For many other professionals out there, you may have heard the term. You may not really understand it. Skinner B.F. Skinner wrote a book called Verbal Behavior in 1957, and the first eleven chapters are all about primary verbal operates; the mand, tactic, echoic, and intraverbal.

And that's what I based my whole book on The Verbal Behavior approach. It's all the primary autoclitics. It's getting kids to talk first in one-word utterances and two-word utterances. And eventually we'll get to, you know, through. Not in my book, but through my intermediate learner course we get to prepositions and pronouns and all kinds of things that do involve some autoclitic language. But so basically, we need we need primary language. We need the primary verbal operates. We need mands and tacts and intraverbals and echoics before we can start worrying about autoclitic.

Because basically autoclitics are secondary verbal operands. So we need the primary verbal operands going to some degree and then we add autoclitics to change the meaning of things like I wanted something instead of I want. So that's past tense. Also, even tone of your voice like it's going to rain today basically means like get your umbrella. It's going to rain today. But if I say is, is it going to rain today or rain today with a question, even the intonation of my voice changes the meaning to be like, OK, let me check my weather app, not get an umbrella. So autoclitic is added to primary verbal operates to change the meaning for the listener. So that's why I never really get into it, because it is quite technical and it's important for analyzing typical language, for furthering our understanding of kids who are talking in sentences and conversational.

But my mission is really to get to kids that are either not talking or minimally talking and not conversational. So I want to I think there's plenty of room to grow with primary verbal operands. So I have never really gotten into autoclitics. But this lecture that Dr. Carbone did basically combined, the Skinner's autoclitic and the work of Roger Brown, who had never heard of. He wrote a book called A First Language, and it's 1973 seminal work. Dr. Brown per the bio and the back cover was a professor of social psychology at Harvard University.

And him and his colleagues have spent years, probably decades, researching how typical language develops in infants, toddlers and preschoolers. And so after carbon's lecture, I search for that and got a used copy of it. It's quite intense and I'm probably not going to read the whole thing ever. And, you know, I just was really fascinated by this lecture, which is free of charge and it's gonna be linked in the show notes by Dr. Carbone.

But I wanted to for parents out there and professionals who really don't want to. Don't have the time or it's not their focus. Their child's not even talking or minimally vocal, you know, diving into Dr. Carbines lectures, very meaty and very advanced. So I wanted to cover the need to nose some of the things I either learned or a kind of validated or I got some new numbers that I wanted to share. So, OK. So we get the Roger Brown and B.F. Skinner's autoclitic work. That's what Dr. Carbone was trying to tie together to show us how we should be programing for young children with autism, with signs of autism who are not talking.

As you probably know, I wrote a book called The Verbal Behavior Approach: How to Teach Children with Autism and Related Disorders. It's in 13 languages. It's still selling better than ever, even though it was published in 2007. As you may or may not know, I am publishing a second book through Hay House in the spring. It's gonna be called Turn Autism Around and it's going to be for parents and early intervention professionals who work with kids one to five years of age. So this this preschool age was one to five year old. Typical language development is an area of high, high interest for me. So that's another reason that I wanted to attend the lecture and I wanted to do this podcast to kind of catch you up.

So according to Carbone, Roger Brown has five main stages of language. And in Brown's book, he talks a lot about morphemes, which is the smallest unit of language that conveys meaning. Again, the audio clitic also changes everything to convey meaning, different meaning. So they are similar. So morphemes are what speech therapist talk about.

And so he gives the example. I wanted to eat the cookies. So that's six words and it's eight morphemes because I wanted adds the ED to the end of want. And to eat the cookies and that S on the end of cookies is plural. And so you get extra credit basically for the two for the EDI's added to want and the S added to cookies. And so even though it's six words, it's eight morphemes.

I talk a lot in my work and in my new book for sure about syllable length. And that's through the brilliant work of Dr. Barbara Esh, who's a S.O.P and a BCBA-D And she talks a lot about syllable length and how, you know, refrigerator is one word and five syllables.

And I want the cookies is four words and five syllables. So morphemes are kind of extra credit. And I talk in my new book about kids that naturally add start to add plurals and contractions and how, as behavior analysts, I've seen a lot of well-meaning professionals and parents who really want to push that language and get children talking in sentences, end up doing a disservice to the children, end up messing up their language even more. So one of the really important things that Carbone said in his lecture was that, according to Brown, the first stage of his language only starts with two word utterances.

And we can't get to two words before we get to one word. And I see a lot of people messing up once you get words expanding that. So a typical child, according to Brown, needs 300 to 400. Two-word utterances before they can expand, before they naturally expand to adding ED, which is past tense, adding ING adding S's for plurals. Adding contractions. I did a podcast interview with Michelle C. And that is podcast episode number seventy-eight.

Mary Barbara dot com forward slash seventy-eight. And her daughter. I'm actually in the process of writing up her daughter as a case study to be hopefully published in a peer reviewed journal because she joined my toddler online course during COVID. And when she joined, we have language samples done and her daughter had two words in the one-hour baseline language sample and only a month, maybe a month in a day or two later, she did another language sample and she had hundreds of words and two word phrases with some contractions and some mommy's car kind of thing. So she naturally added words and her daughter, naturally, who was just two years of age, started picking up the plurals and the contractions, which is amazing.

And I tell my online participants all the time, like the more natural language comes in, the better the less teaching and pushing we get.

We do especially in the early years. We really have to push the right things. Otherwise we end up making a mess. So what I believe we should do. And actually this is what Carbone suggests as well, is we need to teach single word mands, which are requests. We want children to one things. I also in my early learner programs, I teach combined mands, tacts, and echoics. Everything is under multiple control. Which is using a shoe box. Cut a slit in the shoe box, say, shoe, shoe, shoe.

As we have a picture of a shoe putting it, saying the word three times, giving it to the child on the third time or pairing up or using stimulus, stimulus pairing to pair up the picture and the word for the picture. We want to use reinforcers in the beginning, like mommy, or we could call it ma or mama. Shoe.

Well, shoes aren't really reinforcer, but to get your shoes on, to go outside. It may be a reinforcer apple, candy water or anything the child likes. And then some typical things like dog and things that the child sees a lot. Cup. These are all one words, mostly one and two syllable words is where we want to start. And so we want to combine that. So if a child says if we're doing the shoe box and we're saying Apple, Apple, Apple, the child says Apple or any word approximation, that is because they want to put the card in the shoe box that cause and effect action. It is part mand. It is part tact because they can see the Apple picture. And that's part of it.

And this is why kids like Michelle C's daughter make so much progress because we combine all of these and we empower the parents to do these easy lessons. So we want to really strengthen the mands and the tacts and develop a quick control along the way, if at all possible. So as I'm saying words up to three times as I'm going up the steps to take a bath. I'm not saying Johnny. Let's go up the steps. I'm saying up, up, up. And once we develop a echoic control, then we can start teaching even more words. I did do a podcast on developing echoic control so we can link that in the show notes.

The next step after the child knows a couple hundred nouns, reinforcers and common things and mostly one and two syllable words. Then we're going to want to teach things like adjectives and verbs. Actions are verbs. Open. Come up. Go. Kick. Throw in the adjectives such as colors. Big, little. But those are complex.

And you can't just teach them and you can't. When you are teaching kick and throw and open, you need to not teach it in two-word phrases. Because I made this mistake years ago. I said to Lucas, you know, I was trying to push the length of utterance before I knew, I wasn't a behavior analyst back then. I didn't, you know, know how to teach language. And so he really liked the snacks in the cabinet. So I'd say, say open cabinet. And he'd say open cabinet. And we practiced that way too much, but we didn't practice open in any other situation. So then when he was having trouble opening a water bottle for like months or years, he'd say open cabinet because it was over generalized. Right. So when we are teaching push or open or whatever, first of all, we need to teach them in isolation and then we need to carefully start teaching them combined. But if you don't have to teach all of this.

Each step of the way, if we if we teach the basics and have language come in naturally, great.

Sometimes it does, sometimes it doesn't, but we need to be careful if we are teaching two-word phrases. Push truck. We have the auto. Push other things, we have to not only teach opened door and prompt that adjust as many times we need to prompt and transfer closed door, otherwise we're going to make a mess of things and then we need to get, you know, hundreds. Actually, what I said, 300 ish. Two-word utterances that are with spontaneous manding, with tacting, hundreds of things. Listener responding where you touch banana. That all has to be strong before we worry about increasing anywhere beyond two-word pivotal phrases. OK.

So Carbone kind of outlined in his lecture 50 spontaneous mands before we can expand to two things like I want the book or can I play outside like kids that are saying that can spontaneously request 50 or more items. They can label tact 200 to 300 words. They can touch and listen or discriminate between, you know, hundreds of items. And that's really been my work over the years, is getting kids to mand. To tact. He's also saying they need to answer some what and where questions before we can go to try to prompt or model. Can I play outside? Or I want versus I see. That is very difficult discrimination that most times gets messed up. So we don't want to do that. We want to avoid some of these mistakes like carrier phrases, which I did a video blog on a while back. We don't want to prompt as soon as they have 50 words prompt.

I want banana because you know, I want banana is five syllables. It's three words. Those carrier phrases can get very rote. And, you know, I want to teach. Eat banana. I don't want to teach. I want to teach peeled banana. I want to teach that a banana is yellow.

I don't want to just put a pat carrier phrase in front. That's not going to be helpful in my experience. And also in Dr. Carbos experience and also as it relates to Brown�s language development. We also want to avoid the mistake of just plowing through. If articulation is not good, you know, pretzel if it sounds like. And then you go adding i1 or give me or eat pretzel, then it might sound like Iowa, Abida or Ibota. And then people say, well, you can't understand them. Let's put them on a device. Which even on a device, if we are trying to shape up huge length of utterances before they get very spontaneous with language, even on a device is going to be detrimental.

So you want to watch for articulation if if they have articulation problems. Working with a speech language pathologist, we've done probably five or six interviews with speech pathologist. Many of them are BCBA and SLPs. You can check the shows for those, but we have to absolutely work with speech pathologists and consider articulation and consider improving that before we go expanding to other words and adding words. And we also, like I said, we want to be super careful about combining those two-word utterances. When you teach colors, you don't teach red truck and red balloon. You teach red. You teach it with construction paper.

I think I did a video blog on how to teach colors. I see it in my especially in my intermediate learner course, which is part of my verbal behavior bundle. Years of mistakes of well-meaning professionals, programing for kids, want to expand the language, but just make a mess of language. I know I made similar mistakes years ago with my own son and with my clients, but I really do think some of my procedures, like the procedure that I developed to teach colors and intervertebral categories and prepositions and pronouns, you know, really work. And if the kids are at the right level, they really work.

And then Carbone says that Brown suggests that after five or six years of age, that typical language is so complex that trying to figure out mean length of utterances is not very helpful. But until then, from ages one to five, we really have to look at how typically developing kids are learning to speak and just bringing it back. Real quick to last week's podcast with the work of Dr. KLIN, Dr. Ami Klin. He suggests that early signs of autism are kids getting off track with social language, with caretaker child interactions. Babbling, imitation, cooing, eye contact and that we need to bring it back even before we talk about language. We need to bring back that social interaction piece that is often missing.

Really important, in my opinion. And based on the work of Roger Brown as well as the work of Carbone this lecture and the work of Dr. Sunberg and all the all the mentors I've had in the world, the verbal behavior is super important that we not stress length of utterance talking in sentences before a child is ready for it. Otherwise, we will not get the child to their fullest potential language wise. And that's my goal, to have each child reach your fullest potential and be as safe, as independent and as happy as possible. And I want all of you to reach your fullest potential, too. So we're all in this together. Hope you enjoyed that short lecture.

If you did, it wasn't really a lecture podcast, solo show. If you enjoy the podcast, if you enjoyed other episodes, I would love it if you would subscribe to the podcast wherever you're listening. I would love it if you would leave me a five-star rating and review that certainly does help with trying to spread the word and trying to get some traction with the podcast. We are almost at three hundred thousand downloads now and a little over a year and a half. And we're going to continue because I think the parents and professionals and other interested people who listen are getting something out of it. I get, you know, comments and reviews and emails.

And I'm not going to stop until we can really help as many people and as many kids and families as possible. So that's my mission. I am super excited about getting my book in on time, actually, with a couple of days early. And that new book is going to be called Turn Autism Around and it is going to be out in April of 2021. So stay tuned for info on that and have a great one. And I'll talk to you next week.