Posts Tagged 'sound alert'

Reason #4,679 to Avoid Punishment

[Please note: May is Lyme Disease Awareness Month and MCS Awareness Month. I will be trying to do posts relating to both illnesses throughout the month. This piece isn’t actually about either disease, but a lot of it relates to my being on intravenous medication — that is what the infusion pump is for. I have been on IV antibiotics for two years, due to chronic Lyme. So, even when I don’t write about Lyme, it is always a major factor in every day.]

Last night something happened that reminded me how glad I am that I have almost completely eliminated any form of intentional punishment (causing any aversive situation) in raising or training my dogs. So, I decided to write this post about it.

I would say I have had a harder time eliminating every form of punishment from my interactions with a dog than some because of my worst trait. In my continuing effort at transparency, I am going to admit something that most people work very hard to keep private: I have a bad temper. I have worked on my anger all my adult life, and I probably always will. It is a source of deep pain and shame for me. The moments I regret most deeply in my life all relate to my anger.

I have found that there are two types of people (obviously an oversimplification is coming up) who used to do traditional/correction training and have switched to clicker.

The first group is composed of people who really didn’t like to do corrections — many say they always felt repulsed by them — but who followed the rules, because they were the rules, and they were told this was the only way to do it. Karen Pryor writes about feeling this way.

I have always envied people in that group, because I think they have an intrinsic kindness, mellowness, and commonsense that I will always lack. I was in the second group.

Like those in the first group, people in the second group were trying to accomplish something, and this was the way we were supposed to do it. However, unlike the first group, we were not naturally repulsed by the methods. In fact, usually we experienced it this way: if a dog was doing something to earn a correction, it meant they were being “bad,” so we were anywhere from mildly frustrated to incredibly pissed off about that (depending on the frequency and severity of the infraction) by the time we did a “correction,” and therefore administering punishment felt good.

Well, “good” is not really the right word. When I felt angry at my dog, and I did something that I’d learned was the proper response, like a leash pop or an alpha roll, I didn’t actually feel good during or after that maneuver. I was feeling angry and frustrated before the “correction”; during or afterwards, I felt even angrier, but it was accompanied by a righteous sort of “so there!”

I wanted the dog to do the behavior I wanted (or stop doing the behavior I didn’t want), but I also was just acting out of emotion: the dog does something appalling in front of a neighbor, and I feel humiliated, and I react out of that emotion, etc. But it could all be justified in the name of training.

I wasn’t actually training the dog, of course. It was just me letting my emotions get the better of me, which I could do with impunity because I was doing what “the experts” told me to do. I was actually damaging my relationship to my dogs, damaging my dogs’ sense of trust of me, and their sense of safety in the world.

There are very, very few good reasons to ever use punishment, and I have seen some checklists for the very rare instances when it may be justified. However, I haven’t yet seen a checklist that includes what I’d consider to be two of the most important criteria:

  • You are in a totally mellow, non-angry (dare I say peaceful and accepting?) frame of mind.
  • You do not have a tendency to flare up.

I include the second one because, even if you start out mellow, administering punishment, in itself, might trip off emotions of anger or righteous indignation or whatnot, especially if you are the kind of person who has tendencies in that area.

All that said, it’s nice when the theory pays off for you — not just the enjoyment of doing things positively, but the realization that not doing something negative has saved your ass! Usually the fallout from punishment becomes obvious after the fact (often, a good deal after the fact). However, the dividends aid by avoiding punishment also may not be obvious till later.

A while back, I wrote a post about training Barnum to alert me to my infusion pump alarm.

The idea was to teach him to jump on my bed and wake me when the alarm went off, and then keep me awake while I disconnected and flushed the line. I’ve been training this by clicking and treating (with high-value treats) for alerting me to the alarm initially, and then giving him dinner as a very high-value treat when he alerts again — either to the alarm going off again, or to my timer going off. (If I’ve set one of them to go off again.)

Last night, I didn’t have his usual high-value first-round treats, so I planned to just give him his dinner when the pump went off.

As I was doing my infusion, I just knew there was no way I was going to stay awake for it. I thought, “Fine, I’ll fall asleep, and either the alarm or Barnum or both will wake me, and then I’ll deal with the line and feed Barnum.”

That’s the last thought I had until . . . “Blehp! Blehp! Blehp!” And Barnum is on the bed, and I’m really confused.

Then I realize he’s woken me up, and that the pump alarm is going off, and I look around for his treats — and can’t find them. “Good dog! Good boy!” I tell him.

But I can’t find his treats. He keeps jumping on the bed. I tell him to get down, and he does, then he jumps back up! “I’m sorry!” I keep telling him, “I don’t have your treats. You did a good job. Now off!”

Because he won’t stay off, I eventually use some lower-value treats (I always have some form of dog treat around) to get him off the bed.

But he just eats the treats and then jumps back on the damn bed! This is really irritating the heck out of me!

The last thing I need when I am trying to flush and disconnect the line and I am desperately trying to stay awake is to have Barnum be such a pest! In truth, I am attending to my line because I can’t fall back asleep because Barnum is being so dang annoying!

I finish dealing with my line and immediately fall back asleep.

Some time during the middle of the night, I wake up, needing to pee. Barnum is curled up, asleep in his crate. As I go past the crate, I see his food bowl on top — with his dinner in it. Oops. I’m not sure how this happened, but I feel guilty that I didn’t feed him.

I put his bowl in his crate, and he seems more confused than me: “What? It’s the middle of the night. Why are you feeding me when we’re both asleep? I’ll think about it. . . .”

He must have decided it was worth waking up for because in the morning, his bowl was empty.

So, what happened last night?

Barnum knew his dinner was waiting for him on top of his crate. All past experience told him that if he jumped on my bed often enough after the alarm went off, he would get fed. So, he kept doing it.

I have, indeed, been training him to escalate his behavior until he gets his “grand prize” treat. I have trained intelligent disobedience in him, although I thought we still had a long way to go till it was proofed. But, I wasn’t thinking clearly about any of this stuff last night at two o’clock in the morning. I realized it today, after I woke up and recalled the events of the previous night, and pieced together the fractured bits of memory.

When he woke me up last night, I forgot about his food dish and most of our other training. But it was because Barnum kept jumping on the bed that I disconnected and flushed my line — I’m certain of that.

So, the training in that area is working, despite my discombobulation! What I’m really glad about is that I did not lose it at Barnum for being a pest, but instead, reinforced the desired behavior (even though in the moment, I didn’t realize it was the desired behavior) of keeping me awake until I flushed my line. I kept giving him treats, keeping the behavior chain going!

I’m really glad I went with the low-value treats to manage his “misbehavior” instead of yelling at him or whatnot, because I’d have punished him for exactly what I wanted him to do. Punishment stops behavior, and sometimes, even when the dog is annoying the crap out of you, he should keep doing it!

-Sharon, the muse of Gadget, and Barnum, SDiT?

Advertisements

An Alert Pup!

Even when Barnum was a tiny puppy, I knew he had great potential for sound alert. He’s very sensitive to, and inquisitive about, noises. He also seems to have terrific hearing, which is a nice change of pace. (Gadget, I believe, was hard of hearing, and I think Jersey’s hearing was fine, but she just didn’t care much about sounds that were not food related.)

Why, as a hearing person, do I want Barnum to respond to certain sounds?

  • To alert me to sounds, such as timers and alarms, that I don’t hear because I’ve fallen asleep;
  • To alert me to sounds that I’ve forgotten I heard (timers and alarms), because of my cognitive impairment and memory problems; and
  • To make him faster and more effective at retrieving the telephone when it is ringing, and I can’t get to it.

My plan was to train Barnum to perform the sound alerts that Gadget did, primarily alerting me to the stove timer going off. This was a very difficult skill for Gadget to master; it took several months of intensive training. Nonetheless, I’ve been confident that Barnum would learn this alert more solidly and quickly because he is much more attuned to sounds than Gadget was, and because I am a better trainer now.

I trained Gadget to alert to the stove timer as part of my effort to stop burning my meals to cinders.

However, a couple of years later, this skill became useful for an unexpected reason: I got Lyme disease, and for many months, I had to take antibiotics at 12-hour intervals. It is very rare for me to be awake and coherent for both ends of a 12-hour shift, every day.

Thus, Gadget became my medicine reminder. After taking my dose, I’d reset the timer for 12 hours. If I happened to be asleep when it went off, Gadget would hear it, open my bedroom door, run to get his orange squeaky “alert balls” and bring one back and wake me up with it. (You can see a [poor] demonstration of this skill in the second video in this previous post.)

Barnum Takes the (Sound) Stage

In a very laid-back way, I have been encouraging Barnum to pay attention to sounds ever since he was a little puppy. The sounds I’ve focused on are the ring of the cordless phone and the beeping of two timers: the stove timer and my digital “personal timer” (which lives on my overbed table, and which I use to remind myself of phone appointments or other happenings I’m likely to forget).

By “laid back,” I mean that I hadn’t set up training sessions to work on this. I just took advantage of opportunities that presented themselves (also known as “capturing,” in clicker parlance). In other words, if Barnum and I happened to be near the cordless phone when it rang, I’d  act excited, and say, “It’s the phone!” Then I’d toss him a treat. If he oriented to the phone (i.e., if he looked at or moved toward the phone), that definitely earned a click/treat.

I used the same approach with the kitchen timer.

I didn’t really consider this training; more like “pre-training.” My primary goal was to prevent Barnum from becoming so acclimated to these sounds that he stopped noticing them when the “real” training began.

A few weeks ago, because it was convenient, I decided to get slightly more proactive with building a foundation for sound alerts. Whenever my personal timer went off, I’d toss Barnum a high-value treat. He soon was leaping up and running over whenever he heard my timer, and I thought, “Huh, I should probably do something useful with this behavior.”

This urge was strengthened when we were in the living room a couple of weeks ago, and the phone rang. Barnum turned toward the phone and barked at it! I never taught him to do that. I was pleased and impressed.

Barnum Responds to Infusion Pump Alarm!

That’s when I realized I could — and should — train Barnum to be my infusion-pump alarm back-up.

Those of you with chronic Lyme disease or who have had chemotherapy or other long-term intravenous medication are probably familiar with the ins and outs of infusion pumps. For those who aren’t, here’s a crash course in home-infusion therapy, and how a service dog can be of use.

First, there’s the PICC line. PICC stands for “peripherally inserted central catheter.” It’s a flexible little tube that goes inside a vein from your upper arm to your heart, delivering medications into your bloodstream much more efficiently than by any other method. (Here’s a site on PICCs if you want more information.)

Closeup of Sharons arm with PICC line coming out of biceps

I know it looks weird, but it doesn't hurt, and I'm quite used to it after 18 months.

[Photo description: Sharon’s inner upper arm and elbow with PICC line and dressing. The PICC line is a very thin white plastic tube coming out of a round “biopatch” — which looks like a nickel-sized styrofoam disk — in Sharon’s biceps. The biopatch covers the entry site of the line. Several steri-strips hold the biopatch and line in place. A hypoallergenic clear sterile dressing that looks like a piece of plastic covers the PICC area, with two pieces of hypoallergenic medical tape holding down the dressing. The line emerges from under the dressing to a red clasp, which is opened when flushing or infusing. A white plastic cap connects the line to a clear extension tube, which is a port into which syringes of medication or saline can be inserted for infusing or flushing.]

The above photo description tells you the rest of what you need to know (or maybe more than you wanted to know!). The only unusual thing others with PICCs might notice is that this is not a typical PICC dressing. Most of the stuff normally used for cleaning and dressing PICC sites makes my skin erupt and blister in a nasty way. (Thanks, MCS!)

At any rate, I do two infusions a day, 12 hours apart.

To do an infusion, I have to clean the port with a disinfectant, then flush my line with saline (the smaller white syringe in the picture below). Then I hook up an extension tube between the antibiotic and my PICC line. Infusing the antibiotic (the really big syringe attached to the pump in the picture below) takes about 45 minutes. The pump ensures that all the medicine goes in at a slow, steady rate.

When the tube of antibiotic is empty, the pump alarm sounds, which is a loud, grating, “Beep, beep, beep,” and a red light flashes. The I unhook my line, clean it, and flush it with saline again, and finally with heparin (the yellow syringe below).

Infusion Pump with extension tubing, saline flush, heparin flush

The red light on the pump indicates the alarm is going off, and it's time to unhook the extension tubing and flush my PICC line.

[Photo description: two 12 millilitre syringes, without needles, one yellow, one white, lie on a red background next to a 60 millilitre syringe which is attached to a rectangular, brown Bard pump, about nine inches long and three inches wide. Extension tubing comes out of the big syringe. A red light is on at the bottom of the pump, next to the word, “Alarm.”]

Rather, that’s what’s supposed to happen, and most of the time, it does. However, on occasion, I fall asleep while I’m infusing. When this occurs, I wake up many hours later, my line still connected to the pump, the flushes laying nearby, unused. Tellingly, the pump switch has magically been moved from “Alarm” to “Off”!

Actually, I’m pretty sure no magic is involved. The only reasonable explanation is that I wake up just enough to switch the alarm off and then fall immediately back to asleep. I have no idea how long the pump is alarming before I turn it off in my mostly unconscious state. In the morning, I almost never remember waking up and turning off the alarm.

This is less than ideal! The line should really be flushed immediately after infusing. Also, if I were to ever startle awake and move quickly, I could potentially pull the line out of place (particularly because my dressing does not allow for as secure an attachment as is standard practice for PICCs).

Sharon Falls Asleep while Infusing

Exciting reenactment: I fall asleep while infusing.

[Photo description: Sharon lies asleep in bed, propped up with pillows. She wears gray sweatpants and a purple sweatshirt over a pink nightshirt. On her left side the sweatshirt has been pulled up to expose her upper arm, where the PICC dressing and tubing are visible, attached by a long, thin, clear line to the infusion pump, which rests against Sharon’s thigh. On her legs, the saline and heparin flushes lie, waiting to be used after the infusion is over.]

I realized that a really useful skill, and one that would probably not be hard to teach, would be to train Barnum to wake me when my infusion pump goes off.

I started out with classical conditioning, just tossing Barnum one of his favorite treats each time the alarm went off. He made the connection very quickly — within a few days.

Then I moved to operant conditioning, shaping Barnum’s behavior. Once he showed any reaction to the pump alarm, I clicked for the response. As always when shaping, I started with whatever small behavior he offered, such as cocking his ear toward the sound, or turning his head toward it, or just looking up from a nap. Eventually I withheld the click until he was trotting over.

My plan had been to keep upping the ante to get him to jump on the bed, nudge me, etc., until I woke and reward him.

Sunday morning was the first, and only, time I cued a behavior — jumping on the bed (“Paws up!”) — before clicking and tossing the treat.

That night, as usual, I flushed with saline, connected the pump and began infusing. I had my clicker and Barnum’s special treat ready for when the alarm went off. About half an hour into the infusion, I got the suddenly sleepy, struggling-to-stay-awake feeling. I thought that knowing I had to be awake to capture and click Barnum’s response to the alarm would keep me awake.

I was dreaming — literally!

The next thing I knew, I was awakened. All was confusion. The bed was moving, and I didn’t know why. There was a strange noise.

I struggled to orient myself.

The first thing that registered was that Barnum was on my bed, panting and smiling at me. Then, the strange sound resolved, as well — it was the infusion pump alarm.

Amazing dog! Barnum had jumped on the bed of his own accord and woken me when the alarm went off!

I praised him as I scrambled for the clicker, and tossed him his prized treat. I continued to praise effusively, tossing a large handful of other treats I happened to have in my bed.

Barnum "Paws Up" on Bed
I’m here! Wake up! I want my treat!
[Photo description: Barnum, a shaggy black brindle bouvier, with only one eye peeking out from under his hair, has jumped partway onto the bed, with his front legs and chest resting on a dark raspberry-colored comforter. He looks relaxed and slightly expectant. Sharon’s powerchair is visible in the background.]

I was so excited! With Gadget, I had had to spend weeks just on feigning sleep during timer-alert sessions. I will plan to train this way with Barnum — pretending to be asleep when the alarm goes off. He’ll have to really work for that treat by jumping on the bed, nudging me, licking me, etc.

However, I had thought that if I actually did fall asleep while Barnum was at this stage of training, he would have popped up from the floor, looked for his treat, and when he didn’t get it — when I just lay there, conked out — he would give up and go back to bed.

Instead, Barnum made a mental leap that required not just thinking (i.e., problem solving), but self-confidence, too.

I believe our recent focus on lots of free-shaping sessions (as I described in a previous post) led to this breakthrough.

Now, saying that Barnum is a service-dog-in-training feels more legitimate. The pay-off for our training is no longer just fun or incremental steps in increased obedience and communication; the real-world application of practical service skills is becoming apparent.

Barnum’s increased exuberance and eagerness to offer behavior in the absence of a cue or other encouragement from me is thrilling!

An Alert Pup . . . and a Groggy Human

Any assistance dog partner will tell you that there are two members of the team, and both have to do their jobs. Sunday night, Barnum rose to the occasion, while I fell asleep on the job!

After I shut off the alarm and finished fussing over Barnum, I turned my attention to unwrapping my saline and heparin flushes and looking for my sterile wipes and line caps — the supplies necessary to finish the infusion process. . . . The next thing I knew, it was morning, my left arm was still out of my shirt sleeve, and my PICC line was stretched down to the pump!

These are the times when both the pros and cons of being a disabled person training my own service dog come into sharp focus.

On the down side, I had underestimated my functionality. I had thought that being awakened by Barnum, and being forced to interact with him, would keep me lively enough to finish the infusion process. I had not factored in how my cognitive impairment (in this case, memory and concentration issues) would intersect with my sleep disorders. I had not given enough weight to how hard it is for me to “stay on task” when my body and brain switch from “nothing you do will allow you to fall asleep” to “nothing you do will allow you to stay awake”!

On the upside, I now know that I have to build in back-up systems after the initial wake-up, to keep me awake, or to wake me repeatedly, until I finish flushing my line. This is perhaps the biggest bonus of partner-training: I can trouble-shoot and adjust skills as we go along, because nobody knows my disabilities better than I do.

Another bonus to doing all this practice with him is that I am also training myself.

This was the unexpected perk to training Gadget to alert me to the oven timer. People who don’t have brain injury often don’t understand that their suggestions of seemingly straightforward solutions to memory problems, such as “write a reminder note” or “set a timer” require the cognitively impaired person to remember to write the note or set the timer, and also to remember where the note is, to remember what the timer going off means, or to remember to jump up the instant the timer goes off and go do the thing the timer is reminding one of. I write notes all the time to remind myself of what to do, and then I misplace and forget about the existence of the note!

Indeed, before training the memory alert with Gadget, the biggest impediments to using a timer to prevent me from burning my food had been:

  1. Forgetting to set the timer in the first place;
  2. If I did set the timer, not responding the moment the timer went off (e.g., if I was writing an email, I’d think, “I’ll just finish this sentence,” and then forget that the timer ever went off) and keep on keyboarding; and
  3. If I did get up to go to the stove as soon as I heard the timer, I usually became distracted by something on the way (even though the stove was only a few feet from my desk), and forget the original reason for getting up.

However, when I worked for months on training a complex timer alert with Gadget, I spent a lot of time focused on setting the timer, remembering to keep clicker and treats handy to respond when the timer went off, etc.  It became a form of cognitive-behavioral therapy for me. The end result was that I was much more likely to remember to set the timer when I put something on to cook, and I was also more likely to respond to it when it went off, sometimes because Gadget alerted me, but sometimes simply because my mind was better trained to hone in on the timer.

Thus, I’ve decided that the best way to address this added dimension of the infusion pump alarm with Barnum is to train us both to following this protocol:

  1. When the pump goes off, Barnum bugs me (jumps on the bed, nudges me, licks me, etc.), until I wake up.
  2. The instant I wake up, I hit my personal timer (set for six or seven minutes).
  3. I turn off the infusion alarm, and reward Barnum for having alerted me.
  4. I attempt to stay awake to unhook, infuse, and put things away.
  5. However, if I do fall asleep before I’m done, the other timer will go off, and Barnum will alert to that, leading us to repeat steps one through four.
  6. If I do get my line taken care of, and the timer goes off before I’m done, I reward Barnum for alerting to it anyway.

In other words, I’ve had to add a step. Or a step that repeats itself. So, it’s really just more of the same. Which is sort of the essence of clicker training: once you build a foundation, it’s all variations on a theme, requiring flexibility, creativity, and knowing the rules by which you’re playing.

This is also the essence of a service dog partnership.

And of life.

I’m so proud of my dog! And I’m not feeling too bad about myself, either. Nice change of pace.

– Sharon, the muse of Gadget (I’d have done better with the timer alert if you’d trained it better!), and Barnum (What’s that sound?!) Medical-alert-dog-in-the-making!


Receive new blog posts right in your email!

Join 572 other followers

Follow AfterGadget on Twitter

Want to Support this Blog?

About this Blog

Assistance Dog Blog Carnival

Read Previous After Gadget Posts