Welcome to my obsession: scent work. About six years ago I enrolled in a nosework class at a local training center. Within a few months I had entered a trial. After a few more months, I started teaching the class. And everything snowballed after that. I love training and trialing and teaching and judging scent work, but I also love discussing and writing about it. It is an endless well of interesting behavior which illuminates learning (both human and canine) and the ways in which the dog perceives the world.
For those of you who don’t know: scent work/nosework (I use
these terms interchangeably) is a sport where dogs have to find a q tip that
has been scented with a drop of essential oil on it that has been hidden in a
defined area, and communicate the location of the q-tip, or “hide” to the handler. It is based on the way real
life detection dogs (i.e. bomb seeking dogs or dogs who find illegal drugs)
work and it is taught in a very similar way.
Which of These Things is Not Like the Other?
The thing that continues to amaze me about nosework/scent
work is the way dogs learn about
target odors, and keep learning, sometimes in spite of their handlers.
A few weeks ago, I was volunteering at a trial, helping set
boxes for an AKC HD Novice class. In this class, the dogs have to find and
“alert” on a box with a glove or sock in it that has been “scented” with the
handler’s odor. There are two rows of five boxes and in addition to the box
with the handler’s glove, there is a box with a glove that the judge has
scented.
In my experience, It’s not unusual for this class to have a lot of NQs or failures. It’s a relatively new program, that does not have the history behind it that training for oils does. And this day was no different. The first five or six dogs either alerted on a blank box (with nothing in it) or the judge’s glove. The judge (who is a good judge and wanted people to pass and be happy) was getting frustrated.
“I just don’t understand this…the dogs are just hitting on
my glove. I didn’t even scent that glove strongly, it’s been sitting in a
ziplock bag, untouched, for a couple of
weeks…”
I’ve been there. When I am judging, also want people to
pass. But in my experience folks just don’t understand how many repetitions it
takes for a dog to really understand what it is we want them to look for. And I
think people don’t expose them enough to another person’s glove. I have
sometimes asked around at trials about how folks are training, and many were
just exposing them to their glove and rewarding the dog, figuring that would
send the message: find my odor. But in my experience, unless the dogs are given
the opportunity to sniff some other person’s scent article and not get rewarded
for it, they will often assume that they are supposed to alert on any person’s
scent article.
Also, many of the dogs have a much longer reward history
with the essential oil q-tip in a box, and that is what they are
expecting to find. When they don’t find that, they tend to alert randomly, OR I
suspect that they look for any box that smells different.
Because, after all,
that is what we generally start out teaching: we have empty boxes and we
have a box with a scented q tip in it. Only one box that smells different and
we reward the dogs for alerting on the different one.
So, the dog is out in the search, walking around, sniffing boxes because he knows that’s what
the handler wants, and is puzzled because he usually gets some oil smell
somewhere but he’s not finding that. But he is still sniffing, and then he
notices that two of the boxes do smell somewhat different, but which one to
alert on? Well, maybe this one is different enough, and so he alerts on it.
I suspect that the
judge’s glove is the one that smells the most different to the dog. After all,
the handler’s odor is super familiar. And so, in the dog’s mind, the judge’s
glove is the logical choice.
On this particular day, while I was watching the dogs
search, I started thinking about the compare and contrast method that I often
see the dogs using when introduced to buried hides. Initially, most dogs don’t
alert on the hide buried in the sand (or water) because it’s different. It’s
not accessible, but it also has the added ingredient of sand. It’s different enough
so that most dogs will not alert on it.
Generally, after
being rewarded a few times on the buried hide, the dogs start to comparison
shop, when faced with 3 or 4 boxes. You can see them show interest in the hot
box, but instead of alerting on it, move
immediately to the next box and then go back to the hot box. They are clearly
thinking,” hmmm, this one is different, yes, compared to this one, definitely
different”….sometimes they will go to a third box and compare that one also,
before eventually deciding that the buried hide is different, but it is enough
like the regular odor to alert on. It is different enough, in the right
way.
This is an elaborate game of “which of these things is not like
the others” (like the Sesame Street song) and I always love to see this
pattern, because it demonstrates that the dogs are really processing things,
trying to figure out the problem. I wish I had video of this, it’s fascinating
to watch.
While watching the class, I saw this very same thing start
to happen. The judge happened to put the hot box out next to the box with the
judge’s glove. (In this class, the judge puts the hot box in a different
position every time)The dog who ran this configuration was lucky and had a
handler who was following the dog’s lead. The dog sniffed the judge’s box, then
went to the handler’s box, then went back to the judge’s box before finally
sitting down in front of the handler’s box.
“Alert?” asked the handler
“Yes!” replied the judge,
happily circling “Qualified” on
the score sheet.
“yaaaay!” the happy handler rewards her happy dog and hustles
out of the search area.
After this, the dogs started to qualify more. I don’t know
if the judge had noticed the change, but
she did start putting the hot box next to the judge’s box more often.
The dogs whose handlers were letting them lead, often did
the compare and contrast pattern before alerting. A few of those got it wrong, but many of them
got it right. The dogs whose handlers were working a “pattern,” that is, just walking
down the line working the boxes sequentially, and not letting the dogs make the
decision about where to go, didn’t have the opportunity to try and compare and
contrast the two boxes directly. They would sniff the judge’s box, notice that
it was different, walk with the handler to the next box, notice that one was
different also, but because the handler was leading them along (often too
quickly) down the line, the dog did not have any opportunity to return to the judges box or the handler's box, and by the time the handler walked him around again, he had kind of lost the thread of what he was looking for.
And, as often happens, some dogs were just clearly clueless
about what they were doing out there. Generally those were the ones who would
just alert on a random box.
Sometimes we and the dogs just need some more time to work
things out. This is the beauty and joy of nosework. Given time, things will come
right in the end.
Above: the HD set up, with the competitor's box (C) and judge's box (J) next to each other
