Documenting the Alert

Senator and I have been working on his scent training for almost a year now.  He's made some real progress.  He catches most of my highs and lows.  He does it quickly and he is learning to distinguish between them by giving separate alerts.  His public access has been suffering somewhat lately.  I've had some health problems related to my immunosuppression.  I haven't been taking him on many walks or outings lately.  I seem to be over the worst now and am hoping to get back where we were with all of his training.

My concern at the moment, is that he seems to have lost his excitement for some of this.  I've tried switching treats and that seems to help.  I'm planning on playing some scent searching games to up his enthusiasm.  He really enjoyed those when he was first learning to recognize the scent, and I think we will do those same lessons again.  But this time we will also practice the correct alert with them.

I've also been looking for some scientific approaches to teaching DADs and there really aren't any.  The one that I mentioned in my last post (ICAN and Lilly) seems to be the only one.  I've decided to start documenting some of my training.  It will be a good way to mark his progress and to see what we need to be working on more.  I'm measuring how long it takes him to alert and whether he offers the correct behavior for the high or low blood sugar.  He should sit for a high and down for a low after the initial alert which is to paw me.

Date                      Time taken to alert                         Alert give for high or low BG
11/25                           10 sec.                                                 Low for High BG
11/25                           25 sec.                                                 Low for High BG
11/26                           20 sec.                                                 High for High BG
11/26                           15 sec.                                                 Low for Low BG
11/26                           25 sec.                                                 High for High BG
11/27                           15 sec.                                                 High for High BG
11/27                           20 sec.                                                  Low for Low BG
11/27                           15 sec.                                                  High for High BG
11/28                           10 sec.                                                   Low for Low BG
11/28                           15 sec.                                                          ?
11/29                           10 sec.                                                   High for High BG
11/29                           10 sec.                                                   High for High BG
11/30                           10 sec.                                                   High for High BG
11/30                           15 sec.                                                    Low for High BG
12/1                             15 sec.                                                   High for High BG
12/2                             2 min.                                                    Low for Log BG  (old sample)
12/2                             15 sec.                                                    High for High BG (live sample)
12/3                             45 sec.                                                    Low for Low BG
12/3                              5 sec.                                                     Low for Low BG (live sample)
12/4                             15 sec.                                                    High for High BG
12/4                             30 sec.                                                    Low for High BG
12/4                             15 sec.                                                    High for High BG
12/5                             15 sec.                                                    Low for High BG
12/5                              5 sec.                                                    High for High BG

An exciting DAD study and more scent training

Eli Lilly and the Indiana Canine Assistance Network (ICAN) are conducting a study to see exactly what it is that our diabetes alert dogs are sensing when they alert for low blood sugars.  I'm so glad to hear that someone is finally doing this.  If it can be determined what the chemical compound is, then it might be able to be recreated for testing.  The product would be reliable and the trainer could always be sure that they are training with something that the dog can sense.  Sometimes now, I have to wonder about the stability and accuracy of what I am training with.  If the sample isn't exactly "right", they the dog is not getting trained, only confused.  Here is the link to the study.  

I volunteered to send a sample in for this study.  Interestingly, they are using sweat, not saliva samples.  I collected  my sample by rubbing four absorbent cloths over my back and forehead and then blowing on these samples and mailing them in.  There is also a control sample where I did the same thing during a normal blood sugar.

I've been trying some sweat low blood sugar samples.  My low blood sugars usually occur during/after my walks.  I have been using the clothes that I was wearing as the low scent to test Senator, and they work really well.  I keep them in a big plastic bag and they seem good for 3-4 days.  I just put them under my clothes or sometimes just carry them into the room and Senator senses them almost immediately.

I'm continuing to teach him to tell me if I am high or low after the initial alert.  He seems to be doing very well with it.  I've decided to back off of the training with both alerts for awhile.  I'm concerned that I might be confusing him.  For now, I am continuing to do about three alerts per day.  I make sure that I expose him to any natural alert that I have during the day and also do some stored samples to supplement these.  After he paws me for the initial alert, I ask him "what is it".  He does either a down or a sit and I give him a treat.  I repeat this about 5 times and then hug and thank him.  I would estimate that he gets the right alert about 80% of the time.  I tend to have more lows, so he is exposed more to them and tends to default to them (I think).  All in all, I think we are doing well.  I'm trying to get some more public access training experience for him now.

Working with high and low alert together

I decided that it was time to move on and begin training the alerts together.  I haven't read how other people do this part, so I am just winging it.  I'm trying to keep it as simple as I can so as not to confuse Senator any more than necessary.

Our house is arranged so that I can walk in a large circle between the kitchen, living room, and dining room.  So, I put the scents into the small plastic containers with holed punched in them that I had used before.  I put the high scented one in the kitchen and the low scent one in the dining room.  I had the appropriate treat within reach of both.  I put Senator on a leash and walked him in this circle.  When he came to the low scent container, I would say good low and give him the treat for lows.  Then we moved on and did the same with the high scent.  We made about 5 times around and then quit.  He seemed very excited and to enjoy this.

The next day, I followed this same procedure.

On the third day, I quit saying good high or good low.  Instead, I said "what is is" when he got to the container.  If he looked confused or didn't do the appropriate response (down for low, sit for high), I would say "down for low" or "sit for high" and he would do that and get his treat.  After a few rounds, I didn't have to give him the verbal prompt.  Again, we did about 5 or 6 rounds.

On the fourth day, we continued with no immediate prompts from me.  I would say "what is it" when he reached the container and he would do the response.  If he got it wrong, I would correct him and then treat him.  I would say that he got it right more than 50% of the time.  He seems to default to the down when not sure.

The next day, I did the same thing to begin.  His percentage of correct seemed higher.  After about 5 times around, I attached his leash to a door nob in the kitchen and retrieved both containers.  I brought them to him one at the time, had him sniff the container and again asked him "what is it?".  He did well with this.

I plan to continue with this protocol for awhile until his percentage reaches near 100%.

I've had a few natural lows in the meantime.  After the initial alert, I ask him "what is it"  He's been getting it right each time.  It will be interesting to see if he gets it right with a high BG now.  That will say a lot.

Training separate alerts - High alert

After training the low alert for a few weeks, I decided it was time to move on to the high alert.  The alert for a high BG will be first the pawing alert and then a sit.

I tried to do this exactly the same as when I trained the low alert (paw, then down).  I used a scent sample that I had in the freezer.  He alerted to it quickly and then I said sit.  After about two repetitions, he had it down well.  I decided to also use different treats at this stage.  The low treat was cheese and the high alert is now imitation bacon  strips.  I thought it would be an extra cue for him to learn which behavior to do.  I'll phase the separate treats out with time.

The second session, he did the sit right away with no further cues.  I think he prefers sitting to a down which may be part of it.  For each session, I have him repeat the paw and the sit for about 5 or 6 times.  Each time I say "good high" as he sits.

The third trial was about the same.  He seems to have caught on quickly to the fact that a second behavior is required of him after the initial alert.  I doubt that he has actually connected the second behavior to the scent at this point, but hopefully he will at some point.

I had a low after a walk last night and he caught it right away.  He alerted me in the living room.  I said "yes, good low" and we ran into the kitchen which is where I keep all the treats.  I got out a piece of cheese and he did a paw and a down.  He's a smart one all right.

Training separate alerts- Low alert

After much thought and some false starts, I'm finally starting to train Senator to distinguish between the high and low blood sugars with his alert.

I originally wanted to use the left hand low and right hand high for the alerts.  I even began training him to mark left/low and right/high.  Then, I decided that they were too similar and that it would be too easy for me to lead him towards the correct hand.  So, I changed my mind and am going to train him to do a down for low and a sit for high.  Both would follow the initial alert which is for him to paw me.  He has this down so well and even has learned to escalate it at times.  So, I don't want to change that part.  My goal is for him to first paw me and then do either the down or sit to tell me which it is.

I began by holding a container with the low scent in my hand and having him smell it.  I would say "touch" for him to paw me and then "down".  I'm using a separate treat (cheese) for this training.

I've done three sessions now.  The second session was similar to the first.  This time, I had him first find the low on me and then went through the same routine with me commanding the touch and down.  I say "yes" when he does the down and this is when he gets the cheese reward.  I repeat this about 6 or 7 times.

The third session, I again had him find the low on me.  Then, he pawed me and I said down for the first time.  For the remaining 6 times, I didn't say "down".  He just did it.  I did lead him a little with body language, but he seems to be making the connection.

On the fourth session, he did the down with no verbal or physical cues.  He has definitely made the connection between the alert and the down.  I'll continue to do this for awhile before I start him on the alert for high blood sugar.

Trip to Daytona Beach as a service dog

Our annual family vacation was at Daytona Beach this year.  My Dad gets us all places to stay using his time share points.  Its wonderful.  We usually number in the 20s and everyone has a great time.

I decided to bring Senator with us.  A week is just too long to interrupt his training.  So, we drove down with him in the back seat of our little car.  He did just great.  He was excited for awhile and then just slept for the rest of the trip.  It was quite hot.  I had to relent and turn on the AC so he wouldn't suffer the heat.

I had been worried about the check in.  I called twice to let them know that I would be bringing my service dog, and both times they said that it would be fine as long as I had the proper paperwork.  I didn't know what they meant by that so posted the question on the diabetes alert dog forum.  The answer was that there is no such paperwork and that shows how little they know.  I brought a copy of the Florida law on service dogs, but I never needed it.  They didn't ask me any questions at all and had arranged to have me on a lower floor as I had requested.  That worked out well because the elevators were usually very busy and crowded.  We could take the stairs if we felt like it.  The rest of the family was on the 18th floor which would have made that very difficult.

He behaved very well.  It was different having to have his vest on so much of the time, but he was good about that.  People that we met were very interested to hear what he could do.  I met a few diabetics who were excited about the prospect of a DAD.  Mostly, we had people approach us who were missing their own dogs.  A few even had me look at pictures on their cell phones.  And he received countless hugs from kids.  It was fun.

Really, his only weak point were the birds.  Seagulls mostly.  He wanted to chase them.  And they were everywhere and used to people.  I think that if I had had more time, I could have got him used to them, but it was a challenge.  Fortunately, he could walk right by people and toys with no problem or attempt at interaction.  We enjoyed our walks and I let him experiment with the waves somewhat.  At first he was afraid and backed away.  Then he got used to them and waded out farther.  A few broke on him which sent him back toward shore.  But, he liked the water and I think it cooled him off too.

We continued to practice our alerts.  Everyone was anxious to see how he did them and he put on a good show.  He had a few natural ones on me too.  All in all, it was a good experience. I was curious to see what it would be like to have to have him behave as a service dog so much of the day.  I did have to miss out on a few things to be with him, but there were more times that I was glad he was there.

Retrieving the meter

Now that Senator is progressing so nicely with his scent training, I'm trying to teach him some other commands that would be beneficial for a Diabetes Alert Dog.  We decided that it would be useful for him to be able to retrieve my meter so I have started working on that skill.  He might be genetically a Golden Retriever, but he is only golden.  He has no desire for retrieving or even picking things up.  We have our work cut out for us.

I started this by attempting to find something that he might be interested in picking up.  Toys, no, neither soft or hard plastic, or cloth covered. Shoes, no.  Kitchen utensils, no.  Then, when we were outside one time, I tried some sticks.  He likes to chew sticks.  Finally, something he could get his teeth into!  It was a slow start.  I clicked and treated him for just putting his teeth on it at first.  Then we moved on to having him pick it off of the ground.

I discussed this with Margaret, and she suggested trying this with both Senator and Cruise.  This seems to work.  Cruise is more inclined to pick things up and his food drive makes him learn things quickly.  The competition between them really speeded things up too.  I moved from the sticks to the meter in a day.  Cruise was much faster than Senator.  I sat between them and had them take turns, but Cruise would jump over my legs and get the meter before Senator knew what happened.  I decided to have Cruise practice some down stays and that helped.  I was able to have Senator lift the meter and even start to bring it my way before I would treat him.  Its a slow process, but I know he'll get it.

Yesterday, we tried this at Margaret's house and with her dog as the leader.  He didn't do quite as well.  She gave me some good ideas to try.  I was repeating myself too much and probably clicking as he was dropping the meter. So, I'm working on both of those.  Of course, when I got home and tried it again with Senator and Cruise, he did great.  He was even bringing the meter all the way to me.  I was able to throw the meter about 5 ft away and have him retrieve it all the way to my hand.  Very encouraging.  I know he'll eventually get this one.

Scent detection video

The first video shows Gary greeting Senator before he has the scent with him.
The second video shows Gary greeting Senator with the scent hidden in his pocket.

Learning the high scent

I decided that it was time to start training Senator to alert to the high blood sugar scent.  I made a sample when I was at about 220 and began on a Tuesday evening.  I did the same procedure that we did when I started the low scent.  I had him smell the low scent in a small plastic cup and then gave him a treat.  I did this several times.  Next, I had the scent in the cup in my left hand and another scent in a small plastic container with holes in my right hand.  I told him to " find the high".  We did this several times and called it a day.

I repeated this same procedure the next morning.

Then on that Wed evening, I was putting on my shoes to go for a walk after dinner.  Senator came over and alerted me.  I backed away and said "I don't think so"  thinking that this was a false alert for a low.  Then I started wondering if maybe my insulin beat my food to my blood stream and I really was low.  So, I checked.  I was 184.  Was he alerting to my high?  By then it was too late to reward him and I still wasn't convinced that he could learn that quickly.

After our walk, I did some more training with the high scent, but went to the next level to see how he would do.  I hid the scent in different rooms in our house.  He found them quickly.  I had Gary hide the scent in a more difficult place.  He found it right away.  I hid the scent in my pocket and he found that.  He did have it.  Amazing!  He learned the high scent in less than 24 hrs with only two training sessions.  I had read that the high scent was easier to train, but I never would have believed this.

I have been continuing his training with both the highs and the lows for a few weeks now.  He really does have it.  I'm still not sure how I want him to distinguish them.  I've been having him alert me by pawing for both.  I say good high or good low depending on which scent I use or am experiencing.  What a smart puppy!

He's got it!

Senator has been consistently alerting to my lows the last few weeks.  We've been testing him in a variety of situations, and he has performed very well.  I still use the scent pads that I made when I was low to test him and  I have had Gary do this occasionally as well.  He gets it every time and usually quite quickly.

I've had some lows recently too that he has caught.  They happen most frequently right after a walk and I can usually feel them.  When we get back home, I make it a point to get near him and he senses the low an alerts me every time. There have also been some instances now where I didn't know that I was low and he told me.  One time, I was getting ready to leave for work and he alerted me.  When I went to check, I was at 63.  Not a good number to be driving to work with.  Just yesterday, he alerted me when I got home from work and was greeting him.  This time, I really doubted him because I had eaten a protein bar on my way home in preparation for a walk.  I was really disappointed when he alerted me, but decided that I should check just to be sure.  I was at 71.  Its really amazing.

Interestingly, he doesn't false alarm alert.  It surprises me because his alerts always lead to lavish treats and praise.  I really don't understand this, but will accept it.

I've recently noticed that when he alerts me, his eyes are big and very round.  Usually his eyes are quite slanty, so it is very obvious.  The pawing that is the true alert is now only for his benefit.  I already know when I see those eyes!

Senator's first real scent alert

a pinnacle!

Senator had his first alert to my natural low blood sugar scent. It was so exciting.

We had just returned from a walk and I was sitting on the floor taking off my shoes.  I noticed that he was looking at me with a questioning look in his eyes and beginning to approach me.  I was thinking "Oh no, a false alert!" and I'm sure I was starting to back away.  He approached me and alerted.  I tried to ignore him.  Then, I got to wondering......Could I be low?  I tested my blood sugar and it was 74.  It was a real alert!  I praised and rewarded him big time!  He had accomplished what we had been working so hard for.  All of our working with the scent pads and the scavenger hunting that we had him perform had paid off.  He really was learning and could recognize the scent.

This was a huge hurdle and hopefully the beginning of something really great!

A visit from Dee Bogetti

I was able to have a professional diabetes alert dog trainer come to my home and evaluate the progress that we had made so far and help us to get on the most efficient path to success.

Dee Bogetti is a professional dog trainer with an impressive record of success with the training of diabetes alert dogs. I hired her to come to our house for a 2 hour lesson with Senator and me.  It was a nice combination of educational and enjoyable.  I think that we both learned a lot.  Margaret was there too and I was very pleased with how impressed she was with Dee's knowledge.  Margaret was able to know some more specific questions to ask which helped too.

We started with a demonstration of what we had been working on.  I had Margaret hide a loaded scent container in the living room while Senator and I were in the kitchen.  Senator went right to it.  Dee wanted to hide it next.  So Senator and I went back to the kitchen and Dee hid the container in the dining room on the middle shelf of a little table.  I didn't know where it was and ended up standing right in front of it while Senator was looking.  He found it quickly.  So this ruled out my body language in directing him to the container.  I was actually surprised by how well he did this.  Having it off the floor was an extra barrier.  He usually has more trouble with anything off of ground level.  So, with this, Dee was convinced that he could distinguish the low, and I was more so.

She recommended that we stop hiding the containers throughout the house and begin hiding them only on people.  Also, that we should continue to find the highest level of treats to reward Senator with.  And to do a more extreme celebration when he did find the low and give the alert.

Both she and Margaret laughed when they saw that I had loaded the container with 3 little pads.  They say one is enough and since then I've found that to be true.

We talked for awhile about training in general and both Dee and Margaret think I should be working on some more commands with Senator.  Its good for his future needs and just for his development.  So, I've been working on a command for "front" to have him come and stand in front of me with eye contact.  We talked about heel as well and I'm working on some refining aspects of that.

It was interesting watching Senator perform with these two experienced trainers.  Its amazing how he responds to their commands and body language with such precision.  There really is an art to it.  Its very inspirational.

In the meantime, I've been working on all of these things.  Senator's alerts have progressed phenomenally.  He can tell right away when I walk into a room with the scent container in my pocket.  He had alerted to one low that I had after a walk.  When we returned home, I knew that I was low so I sat him down and blew in his face.  With no verbal cues, he alerted.  It was a really exciting thing to see!

Scent progress

At our next training session, we worked on hiding the scent on people.  John held the scent container in his hand and sat on a couch.  Senator found it on him right away and pawed him as the alert.  We were all just stunned.  Margaret was able to come this time and I think that she was as amazed as I was.  Next we hid the scent on her and he did come to it right away, but didn't paw her.  He looked around the room some more and came back to her and even put the container in his mouth, but no pawing.  We tried it with me and he did well.  We practiced this a few times and then called it a day so as not to wear Senator out.  We are doing this off leash and it is very interesting to watch him while he is "working".  Its very obvious that he is searching by scent and not sight.

We also went to a nursing home to complete Senator's therapy dog testing.  He passed and I have sent in his information to the agency.  When I get that back, I will continue his socialization training with a vengeance.  I'm afraid that he's getting behind on some of his skills.

The next weekend, I was away for 4 days for JDRF government day.  It was lots of fun for me, but I hated missing that many days straight of training.   When Gary and I did some training the next day, we could tell that Senator was rusty.  We tried doing the training downstairs this time and he didn't seem to know what to do.  We went back upstairs where we had done most of the training before, and he seemed to snap back.  We did a few more trials and ended on a good note.

Since then, he has done fine.  We are still using the little plastic containers that seem to work well.  We have been putting all four of them in separate corners of the living room and I stand in the center of the room.  I try extremely hard not to give Senator any body language cues, but fail sometimes.  I want him to find it so much!  But, the few times I was able to hold completely still, he did come to me and alert after finding the loaded container.  It was very encouraging.  This seems to be our best test now.  We try to do this in other rooms as well, but the smaller rooms make it more difficult to determine if he is on the exact right container when he alerts.

I have had a few lows where I have tried to test him with.  The first one, I was at about 68.  I had Gary bring him into the room where I was sitting on the floor.  He didn't seem to get it. He did give me a long look in the eyes, but no alert.  I blew in his face and he didn't react.  I then tried to say "find it", but he just began to look around the room.  I went back and tested, and I had gone back up to 78 which maybe was too high for him to detect.  I had been using lower BGs for testing.  More between 55 and 70.  So, maybe he just wasn't determining this as low?

The next time, I was at 65 and tried again.  This time when Senator came in to me, he again looked at me and paused.  I blew in his face and he gave me a tentative paw (alert).  I said "yes" and gave him a good treat.  I'm afraid to get too excited, but it does seem like a good start.

Focusing on the low scent

On my next meeting with John, we decided to limit the target scent to only my low scent.  I had collected a few samples while I was around 60.  Food was only used as the reward this day.

We began as we had before with Senator searching for the low scent which was hidden under a container with holes poked in it.  We used 3 others as decoys and spread them around the room.

The first part again was to have Senator smell the low scent in a separate container.  I would say "find it" and he would begin to search it out.  He did this well.  We tried this a few more times and he could always find it.  Putting it on a chair as opposed to the floor was again more difficult and required additional prompts.

Then we tried hiding the scent with me.  We kept the containers on the floor, but I just held the gauze soaked with the low scent in my hand.  He was able to find it in my hand.  We did this a few times with me changing where I was sitting each time.

Next, we had two other people join us to add more of a distraction.  We repeated this sequence with John leading Senator by each of the people and then to me last.  He would find the scent and perform the alert behavior.

I tried hiding behind a door so that he couldn't see me.  I left it open about 4 in. and had the scent in my hand in the opening.  Senator was able to find me and the scent.  He pushed the door open and did the alert behavior.

Senator was seeming a little agitated.  He kept wanting to go up the basement stairs.  So we stopped there and let him out for a romp.  Afterwards, we went to a nearby nursing home for an outing.  John observed Senator as part of a fulfillment for a Therapy Dog certification.  Senator was able to calm down and behaved well.


At home, I am continuing to use only the low scent.  I put the scent into some little plastic containers that I found that Gary had drilled some holes into.  Gary hides the scent in the living room or hall while I distract Senator in the kitchen.  He seems to be doing well with this.  His alert signal is good too.  We practice this every evening.

I've had a few lows during this time and I've tried to use them to continue the training.  The first few times, I would blow on Senator's nose and give him treats.  The last time I tried this I decided that having the treats within reach was too distracting for Senator.  He was concentrating more on them.  So I have stopped that approach.  The next time, I made a gauze pad from the low and put it into one of the containers.  Then, I had Gary start him off by smelling that and saying the "find it".  I sat in the living room.  Senator came directly to me.  I blew on his nose and he alerted.

Its still hard to say how much of this he is "getting"  and how much is just his natural inclination to come to me anyway, but it sure looks good.  If nothing else, its good practice behavior for when it really does click with him.

More scent training

I had the good fortune to meet someone with some experience in the scent training of dogs.  He has spent most of his time training dogs as hunting dogs and just recently is getting interested in training dogs in ways that will benefit people.

I had a training session with John and he showed me how he begins training dogs for scent work.  His approach was very similar to what we had been doing with Senator with a few additions.

We worked in stages.
John began with holding a treat in both hands.  He had Senator sniff the treat in his right hand and said "find it".  When Senator found it in his left hand, he would click and treat.  He repeated this several times so that Senator would learn this command.

Senator was on a leash for the rest of this.

Next, he put the treat about 20 feet away in the room.  Again, he would have Senator smell the treat in his hand and said "find it".  Senator would look around the room and find the treat.  Again he would click and treat, but not let Senator have the treat that was the target.

Then, we put the treat under a plastic container that had holes punched in it and repeated the above routine.  We did this several times.

Then we put 2 or 3 decoy plastic containers in the room.  These didn't have any treats under them.  He led Senator through a path that would go by the decoys and end with the loaded container.  At this point, we added the alert behavior which is to paw me.  When Senator found the treat loaded container, he was told to sit and then to "touch". He was treated at the point of the touch.  He learned this remarkably quickly.  Now, when he found the treat, he would immediately sit and paw John.

Next we added different smells for Senator to find.  We used peanut butter on little pieces of paper as the target.  We began from the beginning and went through all of the above steps.  He adapted to this new scent well.  Next we tried cheese.  Both were in squeeze containers and the paper was gun cleaning paper and was about one square inch in size.

I tried this next.  I wasn't quite as smooth as John, but eventually caught on.  I wasn't quick enough to stop Senator from getting some of the bait treats to eat.  I think that I wasn't as confident that he was actually smelling the target.  Its hard to say when he smells it and when he is near it.  I think thats the skill in this, is knowing when Senator "gets it".  John could tell when Senator was on the scent path.  He would say "he's working now".  It was interesting and I could begin to see when he was doing this.

We tried putting the loaded container on a chair.  This was more difficult and Senator needed some additional cues to find it.  Then we went outside and tried this routine in the back yard.  Again, it was a bit more difficult.  We didn't use the containers this time.  We just hid a treat, had Senator sniff one just like it and hunt for the hidden one.  He did well with this too.

It was a very enjoyable and interesting day.  I think we wore Senator out.

Discriminating the low scent

After spending many sessions having him smell the low scent that I had put onto the cotton ball, I decided to go to the next step to see if he could distinguish its smell from another that I created.

I made the second smell by making a cotton ball with a normal glucose scent on it.  I didn't check to see exactly what it was, but it was not low.  Fortunately, I can now feel the lows. :)

Mr. I'm not allowed on the couch!

I started as before, having Senator smell the low in a pill container and clicking and treating him as he did this several times.  Then, I introduced a second pill container with the normal scent cotton ball.  I'd like to say that he went right for the low scent, but it was about 50/50.  He did learn that he only got a treat for smelling the correct container eventually.  I did this several time in a row and then moved both containers to another location.  At times he would get stymied and just stare at the containers.  I tried prompting him by saying "find the low".  Sometimes this helped get him going again.  If it didn't, I removed the decoy and we practiced with rewards for smelling the low scent container again for awhile.

I have been doing this for two days now, and will continue with this for awhile.  At least until I am convinced that he can discriminate the smell, or until I think he doesn't understand or can't discriminate.  Then, I would go back to just having him sniff it for a reward for awhile again.

I've been waiting for a vest that I had ordered to get here to take him on outings.  It finally came on Thursday, so on Friday, I took him to our small and deserted mall.  Its perfect for him because I can control when he comes into contact with other people.  We can walk in the mall for awhile until he feels acclimated, and then I take him into a few of the stores.  There is a sporting goods store that we like and that is where we spend most of the time.  Yesterday, as we turned a corner int this store, he started to bark.  There was a picture of a duck on the wall and I'm pretty sure thats what was bothering him.  So after a quick retreat, I started getting him back closer to it.  He would look at it and growl sometimes, but eventually I got him next to it and calm.

I can tell that its been awhile since our last outing.  It took longer to get him acclimated and he wasn't paying quite as close attention to me as he had been.  We will be doing many outings in the near future.

Learning alert behaviors

I decided to use a nudging behavior as an alert signal.  I thought it would be useful because it is something that would be more obvious than a nose push.  After a few attempts of trying to get him to touch me with his paw, I realized that I didn't know how to train this. So, I decided to try to get him to shake, thinking that this would be something that he would use his paw for.  None of the other things I trained had included using paws before.  He picked up on it quickly.

I asked my friend Margaret who is a trainer at the ADAI about this.  She suggested using a coffee can lid as a target for Senator to paw.  I tried this and it has really helped.  I started with the foil insert of the lid because it would be easier to hold against my skin when he was ready for that.  At first, I put it on the floor and rewarded him for touching it.  After a few minutes he got the hang of it.  The next day, I put the foil insert against my leg and said "touch" to get him to paw me.  Again, he caught onto this quickly.  The next day, I eliminated the foil insert and just said touch.  He pawed my leg.  I repeated this many times.  Then, I started to switch positions.  I had been sitting indian style on the floor for most of these sessions.  I tried kneeling and then putting my back to him.  He did well with this.  When I stood up, he was either distracted or confused, because he couldn't make this adjustment.

After a few more days of repetition, he can now paw me standing.  Its really cute because he seem so awkward doing this.  Lifting his paw must not be a natural response for him.

Now I'm realizing that I'll have to be careful about not let him paw me for any other reason.  Just recently, he has pawed me a few times as I was putting on my shoes to take him outside.

The Beginning

Senator and I are beginning a project together.  I am going to see if I can train him to become a Diabetic Alert Dog, DAD.

I obtained Senator from the Assistance Dogs of America, Inc. where I am a volunteer foster.  I had him for about three months and was working on the various commands and behaviors necessary to become a service dog.  He was doing well and I was really enjoying both the training and having such a nice dog.  Then, when he reached one year of age, he was given an evaluation by the program trainers.  He did well with his learned behaviors, but failed the xray part of the evaluation.  He was diagnosed with hip dysplasia.  It was a shock for me.  I knew it was a possibility because it is a problem with Golden Retrievers, but it was still unhappy news.

In the meantime, I had become interested in Diabetic Alert dogs.  I had begun reading all that I could find on how to train them.  My goal was to spend the winter reading and learning, and then ask the ADAI if I might try out my knowledge on a dog who was a flunky from their program.  Little did I know that I already had that Flunky!

So, I decided to keep him and just get an early start.  Most of what I had been reading is from the site diabeticalertdog.com.  There are videos and posts of various dog/trainer teams of the techniques that they are using. It's been very helpful.


I began the training recently and decided to document it all here as a good place to keep things organized.

As of now, my strategy is to train him to
1. Recognize the scent of low blood sugar.
2. Alert me that he has recognized the scent.

3. Find the scent hidden on my body and alert to it.

4. Recognize the low scent generated by my body when my blood sugar is low.

5. Move on to high blood sugars and do the same thing.


I collect the scent by catching my blood sugar at about 70-75 and soaking a cottonball in my mouth.  This sample can be frozen and saved for awhile.  They are good for about 3 days after thawing.

Here is the link to a video that I watched before starting this.

I collected my first sample when I was at 72.  I froze it in a test strip container until I was ready to begin.

I tried the scent work for the first time on 12/30/11.  I thawed out the cotton ball and put it into different containers and had him sniff it.  I clicked and treated for each sniff.  He seemed to catch on to what he was being rewarded for, but I can only hope that it wasn't just the sniffing, but the smelling that he was associating with the treat.  I'm going to continue as if its the smelling for awhile, and then try to use more than one scent to make him distinguish scent from sniff.

I use a glass, a bowl, a baggie and an envelope to vary the container.  I learned on the second try to be quick with the bowl sniff.  He ate the cotton ball!

I tried it again the next day(12/31) with a fresh cotton ball.  I found myself at 76 and made a new sample.  This session went very similar to the first.  I was more conscious of having his nose farther into the baggie, glass, or envelope to be sure he was having a good chance to smell the scent.

On 1/2/12, we had our third session.  I used the second cottonball from the 76 bg sample.  He seemed enthusiastic about doing this training.  I did the scent training for awhile and then we worked on his basic commands afterwards.

I have been repeating this basic process for several days now.  I am trying not to get into too much of a routine, but its hard to vary this much.  I'm worried that he is just going through the motions and not really learning about what he is smelling.  I still think its good to have him associate smell and reward  as many times as I can.  At some point, I'll have him make a choice between smells to see if he can pick out the low scent.