Home » Library » Teach » Special Situations

Chaos to Calm: Using TAGteach to Overcome the Challenges of Autism

Filed in - Special Situations

Editor's note: Being a parent has its challenges, but for parents of children with autism the challenges can be particularly daunting—and isolating. How do you manage difficult behaviors? What do you do if your child can't understand your words? As the mother of an out-of-control child with severe autism and no words, Martha Gabler was desperate for solutions. After reading Karen Pryor's bestselling book Don't Shoot the Dog and attending a TAGteach seminar, Martha found the solution she was looking for…in positive reinforcement. Martha's new book, Chaos to Calm: Discovering Solutions to the Everyday Problems of Autism, is her personal account of how she transformed her family from chaos to calm—and how you can, too.

Excerpted from Chaos to Calm: Discovering Solutions to the Everyday Problems of Autism

Foreword by Karen Pryor

I remember vividly the moment when I first heard of autism. It was in the early '60s. I was standing in our training area at Sea Life Park in Hawaii, next to one of the dolphin tanks, chatting with Gregory Bateson, the famous psychologist and philosopher who worked at the research institute next door. Gregory used the word "autism." I must have looked puzzled, because he clarified by describing some aspects of autism, including absence of language.

Children without words. I was stunned by the news, and by a simultaneous realization: we dolphin trainers could help. I carefully filed that conversation with Gregory away in my mind: autistic children. Would I ever have a chance to meet some?

Years after my career as a dolphin trainer, I wrote a book about how to use conditioned reinforcers and shaping and stimulus control and the rest of B.F. Skinner's arcane vocabulary in daily life, not just with dolphins but with one's family, in the office, in sports, and even with yourself. I wrote it with parents in mind. Alas my publishers, Simon & Schuster, gave the book a sort of jokey name, referring to the absence of punishment in this technology: Don't Shoot the Dog. Naturally not one parent in a thousand would pick it up. Dog owners, however, loved it: after all this was also a totally new way to train dogs.

Children without words. I was stunned by the news, and by a simultaneous realization: we dolphin trainers could help.

Psychology professors also found a use for the book. They used it as a teaching text, and sometimes asked me to speak at their gatherings. In the early '90s I was attending an annual meeting of the Association for Behavior Analysis, when I got on an elevator with a group of women from a school that specialized in helping children with autism.

"Oh," I said, excited, "I've always wanted to try working with your children the way we work with dolphins." Oops. They were deeply offended by the idea that we might use a method with children that we use with animals—except for one. Myrna Libby. By the time we got off the elevator Myrna, a program director at the school, and, it turned out, a breeder of golden retrievers, had extracted a promise from me to visit the school if I ever got to Boston.

In 1998, divorced and with my family grown, I moved to Boston to be near my daughter and her growing family. I called Myrna Libby. She invited me out to the school to have a look around. "Bring clickers," she said.

The clickers Myrna was referring to were toy clickers we now used to train dogs (dog trainers didn't want to go around carrying a dolphin trainer's whistle in their teeth). The training system I'd used in my career as a dolphin trainer had proved to be a very successful way of communicating without words, with ANY learner. It was not a question of intelligence; this is the way any species with a nervous system is designed to learn good news. By finding the right marker for the species—a splash, a blink of light, a touch—you could train anything or anyone.

By finding the right marker for the species—a splash, a blink of light, a touch—you could train anything or anyone.

Today, clickers are in use with human learners of all sorts, all over the world, but in 1998 it was a new idea, and some people objected. "Clickers are for dogs, everyone knows that; how dare you use them with children?" I like the answer one behavior analyst gives to parents: "Every medicine you give to your child or take yourself was worked out first with animals. Aren't you glad we developed these techniques with animals before using them for people?"

For 18 months I spent one day a week as a consultant at the school. At the time they had 300 employees and about 200 children, mostly in residence (there are many more today). I could see from the start that positive reinforcement was the primary tool at this school. Most of the children seemed to be content, even happy, and even those that came in with miserably bad behaviors soon improved. Under Myrna's guidance I taught teachers to be more like dolphin trainers: to ration their words, to observe emotional signals closely, and to use the powerful marker signal, usually in the form of a clicker, as well as the food treats they were already using, to reach the bodies and minds of these children.

The people whom I found heartbreaking were not the children, who seemed to be doing pretty well, but the parents.

The people whom I found heartbreaking were not the children, who seemed to be doing pretty well, but the parents. In the halls, in the lobby, in the cafeteria, I often saw parents, side by side, visiting a child already in the school, or there for consultation about bringing a child. They had done all they could, on their own. They were worn out, tired, anxious, and deep in grief. To give a child over to others is a wrenching decision. And yet living with a child with autism—well, in this book, Martha Gabler makes it very clear just what parents can go through, trying to cope, not just now and then, but day and night, year after year.

There is outside help, if you can find it—and if you can afford it. Applied behavior analysis, teaching one-on-one with the child, the method used at the major autism schools, has been scientifically demonstrated to be useful and beneficial. But no one helps the parent much: until Martha Gabler. The Gablers are themselves parents of a very challenging non-verbal autistic son. Martha's story takes us through her own experiences, learning how to use the tools of reinforcement, step by step, to build behavior one can live with.

The wonderful thing about this technology is that once you understand the principles you can use them on your own. Martha is the poster Mom for this phenomenon. Yes, she had a book—mine; and she had some human-applications instruction in seminars led by Theresa McKeon and Joan Orr, the founders of TAGteach. But it was Martha, working on her own, who picked up the tools—the marker, the reinforcers, shaping, and cues—and made them fit her own circumstances and her own son's potential. As she shows us, the underlying deficits might not change; but the behavior, and the level of communication—those can change. And a parent can do it.

One of the gifts of reinforcement-based interaction is that it brings awareness of how things are for the learner, not just for yourself: and the outcome of that awareness is respect. How brave they can be, these children. And how funny; sometimes, concealed under the inability to communicate, there is a great sense of humor. I hope Martha Gabler's wonderful book reaches the hands of every parent with a tantruming child, every parent spending sleepless nights, every parent with a son or daughter who can't be taken to the supermarket or on a bus without stress and anxiety. May this book bring you hope. And help. And some peace. And maybe even joy.

Abridged from Chapter 7: Going to Sleep

Of all the difficult, vexing behaviors of children with autism, the most grievous problem is not sleeping.

Of all the difficult, vexing behaviors of children with autism, the most grievous problem is not sleeping. Our children with autism often cannot fall asleep or stay asleep, leaving the parents worn out, exhausted, and short-tempered. Continuous lack of sleep causes huge problems for parents at their jobs, and for siblings who are continually being awakened by shrieking or commotion. It extends the agony of autism throughout the night. It is exhausting and depressing beyond description.

Many parents find themselves lying down with their child for hours to get him to sleep, bringing the child into their own bed, taking the child to another bed or room in the house, and then having to deal with the problem all over again during the inevitable night-time wake-ups. Struggling with sleep problems for four to eight hours a night is not unusual. We dealt with those problems for years, and I found TAGteach to offer useful solutions. TAGteach was not around when Doug was small, and it would have made our lives so much easier at bedtime and during middle-of-the-night wake-ups. When I learned about TAGteach, I was able to teach Doug to lie still, be quiet, and go to sleep.

In tackling sleeping problems, we found the following steps helpful: (1) set a consistent wake-up time early in the morning and a consistent bedtime in the evening, (2) avoid large meals or snacks late in the evening, and (3) reinforce the tag points Quiet Mouth, Hands Still, Feet Still, Head on Pillow, Yawns, and Exhales.

To start, we implemented Step 1: set consistent wake-up and bedtimes. We got Doug up, dressed and moving every day by 7:00 a.m. at the latest, or even 6:30 or 6:00. We did not let him sleep late, take naps, or doze off in the car. We did whatever it took to keep him awake and moving for the entire day, and enacted all kinds of antics to keep him from falling asleep in the car or on the sofa during the day.

A consistent bedtime is also very important. Ironically, we learned that an earlier bedtime worked better for Doug than a later one. We initially thought that if he went to bed later, he would be more tired. Wrong. We finally figured out that a later bedtime meant he was more tired, but also more wound up, agitated, and cranky. For us, an earlier bedtime works better.

We then implemented Step 2: make sure Doug had no big meals or heavy snacks in the evening, because this might give him an energy boost.

The big job was implementing Step 3, reinforcing the tag points Quiet Mouth, Hands Still, Feet Still, Head On Pillow, Yawns, and Exhales. In the evening, we went through the usual family bedtime routine. Then I gently put Doug to bed, explained that it was time to be quiet, stay in bed, and go to sleep. (Doug was about nine or ten at this point, and whether he was able to understand everything, I don't know. As a parent, I felt I should explain my expectations.) I turned off all the lights in the room and hallway, and went out of the room. From that point on, I reinforced the following tag points—Quiet Mouth, Hands Still, Feet Still, Head on Pillow, Yawns, and Exhales. There are two ways to reinforce these desired behaviors: from outside the room or from inside the room. In the beginning, I always reinforced from outside the room, but I recently found that I could also reinforce from inside the room. For either approach, I found the following materials helpful:

There are two ways to reinforce these desired behaviors: from outside the room or from inside the room.
  • comfortable, loose clothing
  • a flashlight (it is better to use a flashlight at night for marking the behavior rather than a clicker, because the flashlight is silent and will not disrupt other family members)
  • a comfortable, padded high stool to sit on (I found it works better to sit on a high stool than a chair, because it's easier and less tiring to jump up constantly from a stool than a chair)

From Outside the Room: I sat outside the bedroom (on the stool) with the bedroom door slightly ajar so I could observe Doug, then waited until I could mark and reinforce any of the following tag points: Quiet Mouth, Hands Still, Feet Still, or Anything Still. Behavior is variable, so a child who is shrieking and bouncing around will, at some point for a split second, stop shrieking or stop moving some part of his body. When Doug had a split second of Quiet Mouth or Still Body Part, I flashed the light, went in, and gave him a gentle pat on the shoulder or arm. I did not talk to him or explain, and left immediately. Then, I sat outside and waited for the next instance of Quiet Mouth, Still Body Part, Head On Pillow, or Yawns. It took a long time, but eventually he fell asleep.

To review, here is the procedure I used:

  • I flashed the light and went in the room and patted his arm every single time he was quiet or still.
  • I did not go in if he was screaming or jumping, no matter how loud and anguished he sounded.
  • If he bolted from the room, I took him by the hand and gently lead him back to the bed. Then, I returned to my post outside the room and waited for him to do any one of the tag points listed above; I kept flashing the light and rewarding him with my presence and attention. I kept this up until he calmed down. Eventually he did calm down, get into bed, and lie still.
  • After time went by and he appeared to be drifting off to sleep, I decided it might be better to continue to go into the room, but not pat him—I didn't want to wake him—but instead I leaned quietly over the bed or patted the pillow.

Naturally at the beginning, there was a powerful extinction burst of loud, disruptive behavior, and I simply had to outlast it.

This process took a long time for us: several hours per night the first few weeks, to get the new, desired behaviors going; then several months to teach them so my child could keep them up. But they worked. For a long time, Doug went to sleep nicely.

But time passes; I got lazy and failed to do the "maintenance" work to keep this behavior in good shape. Recently I had to patch it up again, since it had descended into a shambles. Doug was staying in bed, but had developed a terrible repertoire of noises, including an incredibly annoying burbling slurp that drove me wild; plus, he bolted out of the room constantly. The noise was awful: shriek, yowl, slurp/slurp/slurp, pounding footsteps, and slamming doors. When I restarted the process, I decided to take notes: I kept track of what time Doug went to bed and what time he fell asleep. I also counted how many times I flashed the light and went into the room to pat him on the shoulder (to mark and reinforce the desired behaviors of Quiet Mouth and Still Body Part). Here is the raw data from my notes from the first few days:


Time Started

Time Finished

Total Time

No. of Flashes


10:25 p.m.

2:00 a.m.

3 hrs. 35 min.



9:47 p.m.

11:30 p.m.

1 hr. 43 min.



10:17 p.m.

11:25 p.m.

1 hr. 8 min.



10:27 p.m.

12:45 a.m.

2 hrs. 18 min.



10:10 p.m.

12:22 a.m.

2 hrs. 12 min.



10:15 p.m.

11:34 p.m.

1 hr. 19 min.


The first night, Thursday, it took a long, long time. Friday and Saturday nights were shorter, and then, extinction burst! Sunday and Monday took a long time again. On Tuesday, the time decreased. My notes showed some progress, but also that it was still taking a long time to get him to sleep. I kept up the routine of sitting outside the room, but fatigue and impatience drove me to try another approach, and I experimented with working from inside the room. I had never tried this because I thought it would not work; if I was already in the room, how could my presence be reinforcing for my child? When I tried working from inside the room, I found that things moved faster. I was better able to decrease the amount of time he was awake and increase the amount of time he was asleep. He also stopped making noises and stopped bolting out of the room.

From Inside the Room: I tried this one evening when I simply could not face sitting outside on the stool. I was tired. My legs ached. So I plopped into a chair next to the bed. I sat in the chair and flashed the light and patted my son's arm every time he was quiet or still. I faced away from him if he was making noise or moving, but faced him to pat his arm when he was quiet or still. This approach, of sitting next to the bed, unexpectedly turned out to be more efficient for us. By chance, the chair I plopped into was a desk chair that swiveled. This made it easy to swivel toward him when he was quiet or still, and to swivel away when he started to shriek, bounce, or, worst of all, slurp. Since I was sitting right next to him, I could reinforce Quiet Mouth and Arms/Legs Still, Head on Pillow, or Yawns continuously; I fell into a steady routine of flash/pat, flash/pat, flash/pat. This intensive stream of reinforcement calmed him down more quickly and he fell asleep in less time than it took when I stayed outside the door. It worked; I was astonished, but pleased with the outcome.

It worked; I was astonished, but pleased with the outcome.

Throughout this sleep-training project I made mistakes. In the beginning, when I was outside the room, I wasn't able to reinforce heavily enough. When building a behavior, it is important to do continuous reinforcement, and my reinforcement was not continuous enough. His vocal stimming was non-stop, so progress was slow until I sat next to him and figured out how to mark and reinforce each microsecond of Quiet Mouth between the slurps. I could only reinforce at such a heavy rate while sitting right next to him. Also, I learned that when he was quiet, I had to keep marking and reinforcing through the entire quiet moment. In fact, doing intensive flash/patting was the only way to keep the Quiet Mouth going. With time, he had longer and longer periods of quiet.

The next mistake was keeping this high-intensity reinforcement going on too long. After a behavior has been built, it is time to change from a continuous reinforcement schedule to a variable reinforcement schedule (reinforce at different intervals). I realized at one point that there were longer periods of quiet between the slurps and bounces, so I tried to see if it would work to wait a bit longer before I flashed the light and patted his arm. To do this I would flash/pat, then wait one, two, or three seconds, to see if he could stay quiet for a longer time. He was able to stay quiet, so I continued to experiment and lengthen the time between the flash/pats and thus reduce and vary the rate of reinforcement.

Another mistake was marking a behavior too soon or too late. Sometimes, when I anticipated a micro-second of quiet between the slurps and flashed the light, he would suddenly do an extended slurp, so I inadvertently marked and reinforced an undesired behavior. Sometimes he would be quiet, and I failed to catch it promptly. Fortunately, TAGteach is very forgiving of such errors. Most of the time I marked and reinforced appropriately, and the desired behaviors grew stronger.

Was he trying to reinforce me? I don't know. But it was very sweet.

Despite the mistakes and frustration, after five months Doug was going to sleep nicely in a reasonable amount of time, and staying asleep through the night. Even if he was not asleep, he was now able to be quiet in his room and stay in bed, so we can sleep. As always with autism, there were touching and humorous moments. Doug really appreciated the help with falling asleep. He would carefully pick up the flashlight and hand it to me with an appealing look that said, "Here it is. Be sure to do this tonight." He even pushed the chair into position next to his bed and set the flashlight on the table right next to the chair, so everything would be ready to go. Also, while I was flashing and patting his arm, he would reach over to me and, very gently, pat my arm! Was he trying to reinforce me? I don't know. But it was very sweet. On the other hand, there were times when he would yank on my arm and yell, "Out!" It was an adventure with laughter and moments of frustration, but we ended up with a good outcome: for twenty to thirty minutes of work, we got six to seven hours of uninterrupted sleep—a real luxury for autism parents. 

About the author

Julie Gordon is the Content & Communications Manager at Karen Pryor Clicker Training. She oversees editorial development and content management for the company’s websites, and regularly contributes articles and blogs.

Post new comment

  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <embed> <object> <div>
  • Lines and paragraphs break automatically.
  • Glossary terms will be automatically marked with links to their descriptions. If there are certain phrases or sections of text that should be excluded from glossary marking and linking, use the special markup, [no-glossary] ... [/no-glossary]. Additionally, these HTML elements will not be scanned: a, abbr, acronym, code, pre.
  • Each email address will be obfuscated in a human readable fashion or (if JavaScript is enabled) replaced with a spamproof clickable link.

More information about formatting options

To prevent automated spam submissions leave this field empty.