THE OTHER END OF THE LEASH

Several readers have asked about the new product called “Dognition” that promises (quoting from the website), “You’ll learn your dog’s cognitive style by playing fun, science-based games — an experience that gives you the insight you need to make the most of your relationship with your best friend.” I was curious myself, given that the force behind the product is the work of scientist Brian Hare, whose relevant claim to fame is his research on the ability of dogs to inherently understand a person’s pointing gesture. I’ve argued that this claim needs more research, as does his suggestion that the long-standing relationship between dogs and people has resulted in the evolution of special communicatory skills in dogs (especially as regards to pointing). However, I love that his work has helped to energize research on canine cognition. (If you want to read more about whether dogs can inherently understand the pointing gesture of a person, see Do Dogs Inherently Understand Pointing Gestures? and  An Update on Pointing Gestures and Dogs.)

Dr. Hare has since come out with a new book, The Genius of Dogs: How Dogs are Smarter Than You Think, and a product called the Dognition Assessment Toolkit. I haven’t read the book yet, but was intrigued by a set of games he has developed that are supposed to increase one’s understanding of their dog’s personality and cognition style. The perfect cabin-fever reliever perhaps? Even better, the website explains that all the results will be aggregated and used as an example of “citizen science.” It says: “By gathering this data we can begin to understand more about all dogs, much more quickly and on a broader scale than if scientists had to conduct this research themselves.”

I signed up Willie, Tootsie and Katie’s dog Leo at $59 each, passing up the $129 offer of the toolkit + one year’s annual membership. Katie and I began the games last week, and finally finished with Tootsie yesterday. I say “finally” with intention: I want to tell you that the games were incredibly fun and that I learned oodles that I didn’t know about Tootsie. Alas, here’s the bottom line: Sometimes the “games” were fun, sometimes they were tedious, sometimes they were mildly aversive (more on that later) and I’m afraid I didn’t learn much about Tootsie that I didn’t know. I had planned to finish the games with Willie today, but realized that rather than looking forward to it I was… what’s the right word? “Dreading it” is too strong, but it felt like a great burden that I would have to slog through. Then I realized that I didn’t need to finish the games with Willie at all; I already had learned what I needed to know to make my own evaluation of the project for this post, and didn’t have any expectation that I would learn anything especially useful about Willie that I didn’t know before. Before I continue, let me be absolutely clear: Others might find the exercises great fun and the eventual evaluation extremely useful in improving their relationship with their dog. But all I can give you is my honest assessment of my own experience, for whatever it is worth.

Here is a bit about how the program works: There are five categories of games, labeled Empathy, Communication, Cunning, Memory and Reasoning.  Before playing the games one fills out an extensive questionnaire. I’d give it kudos for being thorough, but many of the questions should be answered with “I don’t know” (which, good for them, is always an option). For example, you are asked if your dog “understands” the cue to sit. Without testing Tootsie a la Ian Dunbar’s game, and asking her to sit while she is already sitting or while lying down, I can’t answer that question. I can answer if she does sit when asked, but that is an entirely different question (which was also asked, I should note). However, many of the questions were easy to answer, like “Is Tootsie friendly to other dogs?” One could answer “All the time, Some of the time,” etc. I choose “Some of the time” for Willie often, given that his behavior is so context dependent.  After the questionnaire, you begin the games. With a few important exceptions noted below, they are explained very clearly. If you decide to play, be aware that you’ll need someone to help with all of them, a fact not made clear before you begin and that threw me off when I first tried to get started by myself one afternoon.

Many of the games involve putting a treat on the ground on one side of you or the other. The treats might be on the ground, inside a cup or under a piece of paper. Your dog has a variety of ways, over the course of the tests, to choose the “correct” side, based on your pointing toward it, showing your dog that it is in the cup before turning the cup over, etc. Each of the games that require a dog to “choose sides” do a good job of establishing first whether your dog has a side preference. (Both Willie and Tootsie more often went to my right side (their left), which fits with the knowledge that most dogs are “left pawed.”) Katie and I did quite a few of the games together, either with her as the helper for me and Tootsie, or vice versa for her and Leo. The biggest snag that we ran into was a discrepancy between what the dog did and how we were asked to record it.

Here’s an example: You define “left and right” with 3 sticky notes, one directly in front of you, one each to the right and left of you on the floor. With a helper holding your dog, you might point to a cup with a treat hidden underneath it on your right side. The instructions tell you to score your dog as “retrieving the treat” if you pointed to the treat on your right, and your dog went between the sticky notes in front and to the right of you. Thus, passing between 2 sticky notes on the ground on the same side as the treat meant your dog had “retrieved the treat.” (The word ‘retrieve’ was confusing to us…’chosen’ or ‘focused on’ would be more helpful.) But here is what sometimes happened: Imagine I placed 2 cups on the floor on either side of me, one with a treat, and pointed to the one with the treat while Tootsie watched me from about 6 feet away. Let’s say that the cup with the treat was on my right. Tootsie was released and wandered through the sticky notes to my right, but paid no attention at all to the cup with the treat underneath, and passed behind me to vigorously sniff at the (empty) cup on the left. Based on the instructions we were told to score that as “retrieved the treat,” but her behavior indicated she was focused on the cup without anything in it.

Other games require you to say something to your dog and then stay stock still and silent for up to two and a half minutes. This was the part that Katie and I found a bit tedious. Standing silent and motionless while looking directly at your dog is so unnatural I found it unnerving. (In one test you stand still and silent for 1 minute, then 1.5 minutes, then 2 minutes and then 2.5 minutes. Katie and I were both ridiculously relieved when it was over.) One could argue that science is not about having fun, but is about designing and conducting good experiments. However, anyone who has taken Experimental Psychology 101 knows that the behavior of the experimenter must be carefully controlled in any study. For the results to be lumped together and analyzed, either only one carefully trained “handler” would have to be the one cuing the dogs in exactly the same circumstances, or a few well-trained handlers would have to undergo extensive training and testing to ensure that their behavior was consistent. Thus, it is hard to know what the results of an aggregate of hundreds or thousands of dog owners doing these tests in their own homes would actually mean.

The mildly aversive section occurred when I was to put a treat on the ground, say “No” or “Leave It,” and then wait for an entire (endless) 90 seconds without moving or speaking. This was not a problem with Tootsie, who doesn’t know a Leave it cue, and who trotted toward the treat and gobbled it up. However, Willie does know Leave It, and I wish I had a video of his face staring at mine for 90 seconds after being told Leave It when I placed a dried turkey heart on the ground. That 90 seconds lasted at least an hour… he looked so confused and unsure after about 10 seconds that I could barely maintain my stance. Next you are instructed to put the treat down, say Leave It and turn around so that your back if facing your dog. Willie waited 1 minute and 17 seconds and then padded over and ate the treat. Next you are to hold your hands over your eyes, and this time he waited less than 20 seconds. I have never been so thankful to have a dog disobey a Leave It in my life. Both Katie and I (she acting as recorder and dog holder if needed) darn near cheered when Willie took the treat, so that we didn’t have to wait another endless 90 seconds. I suspect that reading this one could argue: “Really Trisha. Ninety seconds? Seriouslyy, how long is that really?” Answer: When you have to stand still and stare at your dog for that period of time, it feels much longer than one might think.) (Cautionary note: After those games I worked on Leave It with Willie, reinforcing him for avoiding a treat on the ground for 2, 4 and 6 seconds; the games being the perfect way to ruin a well-trained cue if you don’t do some clean up work afterward.)

The results? We finished all the games with Tootsie (3 of 5 game categories are done with Willie) and immediately got her results. Tootsie was classified as a “Stargazer,” whose “unique genius lies in the mix of strategies that she uses to approach daily life… She certainly has a wild, wolf-like side that is especially useful in the environment of the rugged individual.” Labeling Tootsie, the 7-years-in-a-puppy-mill Cavalier King Charles Spaniel who wants nothing more than to eat (anything) and lie in one’s lap, as “wolf-like,” leaves me close to speechless. Her “empathy” scores were “off the charts,” even though it was Willie who yawned after I did and not Tootsie, yawning after seeing another yawn being indicative of empathy). Tootsie’s results also indicated that she is “highly collaborative;” I’m guessing that is because she did indeed focus on the food I pointed to in some of the tests. But categorizing her as such is a stretch, I’d argue. Tootsie is pretty much all about food, while Willie is often called “Trisha’s mood ring.” Willie appears to care deeply about how I am feeling, while Tootsie appears to care deeply about how quickly she can get the food out of my pocket. Of course, I could be wrong about my assessment and the tests could be more accurate, but I wouldn’t bet the farm on it.

The one result that I found surprising, and thus interesting, was that Tootsie did indeed focus more on the treats on the ground if I pointed to them (but not when the treat wasn’t visible ). Willie, who has learned exactly what an outstretched and pointing arm means, appeared to pay no attention. I suspect that pointing to something in the distance, with one’s arm almost horizontal to the ground, is a very different signal than pointing almost straight down toward the ground, with one’s arm just a few inches away from the torso. That got me thinking that perhaps what we call “pointing” could actually involve several different visual signals that should be considered independently. (And one more pointing note, skip if all this pointing stuff is tiresome to you!: Dognition actually has you both point AND turn your head toward and look at the object in question. Thus, pointing, and gaze are confounded here. That’s a technical point, but an important one to those who study animal communication.)

At one point the evaluation interprets Tootsie’s inconsistent responses in a series of tests as “switching back and forth” between strategies, a behavior labeled as “impressive flexibility.” Another way to label this is “having no strategy at all.” I’m just saying. Reading their evaluation of Tootsie reminded me of schools in which all the students get A’s and blue ribbons because, well, they are there. On the plus side, it appears that the designers of Dognition are working hard to clarify that there are no ‘right or wrong’ answers, and that every dog is using a different kind of cognition strategy to make their way in the world. That’s important, and I credit them for it. This is not about whether your dog is a Good Dog or a Not So Good Dog, but about finding what’s going on between those furry ears of hers. Learning more about your dog is a wonderful thing; however my guess is that most people who are motivated to spend the money to play these games with their dog, and get through all 2.5 to 3 hours required, already know more about their dog than a product like this can teach them.

A few notes if you decide you want to play: 1) Don’t even think about doing this all at once. You and your dog will be toast if you do. The website is well designed to allow you to pause, and wisely encourages you to do so. We did Tootsie’s tests over 4 different sessions. 2) Round up an assistant before you start, you absolutely have to have one for almost all of the games. 3) You need to be right beside your computer to play. Laptops and iPads work great. 4) Take your own notes about your dog’s responses. You don’t get a summary of what they did (“followed point 2 out of 6 times, for example) and I’d love to see the actual data. 5) Careful of the treats: Your dog could end up getting a lot of them. Tootsie basically ate the equivalent of two dinners during the last session, even when she didn’t “get” the treat much of the time.

Have you played it yet? I’d love to hear if you have and what you thought of it. I truly wanted to write a positive review myself;  the Advisory Board on this project reads like a Who’s Who in canine cognition, including Dr. Juliane Kaminski, Dr Richard Wrangham, Dr. Adam Miklosi, Dr. Laurie  Santos and Dr. Josep Call as well as Dr. Brian Hare. I know some of these scientists and not only admire their work, I like them personally. But good science is about evaluating the facts, and it would be disingenuous of me not to be honest in my assessment. Let me know what you think.

MEANWHILE, back on the farm. We’re melting! Finally finally finally the spring thaw has begun. It may be under freezing at night, and there is still plenty of snow and ice on the ground, but there are also patches of bare, spongy ground, and the air is rich with the songs of Redwing Blackbirds and the guttural calls of Sandhill Cranes.

The lambs are due to start dropping any day now, so Jim and I are busy in the barn. We had a lovely, long weekend at a friend’s cabin up north last weekend, and came home refreshed and ready for spring. I’m finishing up my Contemplative Photography course, which I’ve absolutely adored. Here are two of my photos for the the class, a Beech leaf and a scene from up the hill, just yesterday. Enjoy.

When I was doing my PhD research in the 1980′s I wanted to see how a dog’s brain responded to different types of sounds before and after training. This was in the 1980′s, and the only method available for animals was to do Evoked Potential tests, in which simple, tiny electrical sensors were glued to the scalp. This allowed one to measure activity in the brain in an extremely general way: What kind of brain waves does one see in the Temporal versus the Parietal lobe, for example, after hearing or seeing specific stimuli?

Evoked Potential tests are logistically easy to do on adults: You shave tiny areas of the scalp, glue on the sensors, and ask the subject to stay still for a set period of time. The research subject mustn’t move, because muscles move through electrical stimulation in the brain too, so any movement confounds the brain’s response to external stimuli. But how do you do it on dogs?  How would you keep them still? Absolutely motionless? And here I was in Zoology, not neurobiology, so how would I pull this off?

I felt lucky to be at a large research institution like the University of Wisconsin in Madison, and so I reached out to neurobiologists in other departments. “How does one do Evoked Potential tests on dogs?” I asked. “No problem” I was told. “It’s not that difficult.” “But how do you keep the dogs from moving while you’re doing the test?” I asked. I received a blank face and a silence as an answer. “I don’t understand.” the expert informed me. “What do you mean?” I repeated my question. His answer, after another pause, was “We just don’t bump the tray.”

Slowly, with growing horror, I began to understand why he was confused by my question. I hadn’t asked “How do you do Evoked Potential on live dogs?” The dogs they used for their research weren’t alive; they were killed and the tests were done immediately after their death. Not what I had in mind, needless to say. I left his office and found a researcher (Dennis Molfese at Southern Illinois University) who did Evoked Potential tests on human infants. We did use small dosages of Acepromazine to quiet the dogs, but also did lots of training, and had few problems keeping the dogs still. We ended up with good results that showed dogs paid more attention to short repeated notes when linked with recalls than long, continuous notes.

You can see then, why I am thrilled to read that researchers have now pulled off doing fMRIs on happy, healthy, live dogs. Scientist Berns and others in Atlanta worked with owners to train 2 dogs to sit motionless in an MRI machine for 10 seconds at a time. If you’ve ever had an MRI, you know that this requires far more than just lying still. For reasons once explained to me while I lay frozen with fear in a tight, dark tunnel, the magnetic part of the MRI creates an ear splitting series of pops and bangs and groans that is hard for many people to sit through. And yet, they pulled it off with dogs, and got some interesting results.

“The task was designed to determine which brain circuits differentially respond to human hand signals denoting the presence or absence of a food reward.” And sure enough, when a visual signal was given that meant food was coming (not unlike a clicker stimulus), the ventral straitum of the caudate cluster was more active than when a visual signal meaning “no food coming” was seen by the dogs. Activating the ventral straitum causes the release of dopamine, the “Oh Boy!” neurohormone mediating happy anticipation. It’s the same neurochemistry that occurs when you get your dog’s leash and she begins to dance in anticipation of a walk.

The results are interesting, but not earth shaking, given they were exactly what was expected. What IS earth shaking is that scientists have found a way to study what happens inside a dog’s brain in kind and compassionate ways. This is huge…. the amount of information we could derive from this is immeasurable. I have long said that our relationship with dogs provides us with research opportunities unlike any other, and here’s proof of it.  Stay tuned… the next couple of decades of research and our understanding of other animals are going to be absolutely amazing.

MEANWHILE, back on the farm: The lambs are thriving, the dawn chorus of bird song floods into the farmhouse around 4:30 am, but who could mind such a beautiful way to wake up?

And although the weather has been challenging of late, (hot, cold, wet, no rain, too much rain, you name it), the flowers continue to lift me up and make me happy. Here’s an Iris I planted last year, first time seeing it bloom:

When clients, eyes shining, would tell me how smart their dog was, I’d often respond: “I’m so sorry.” Of course, I said it with a big smile and we all laughed about it, but the truth is, smart isn’t always what we want. Most of us want dogs who are “smart” in that they learn what we want them to learn at lightening speed, but we don’t want them to use their cognitive powers against us. Some dogs do just that; I swear I can see smoke coming out of their ears as they try to figure out how to train us to do their bidding. I remember one dog, a brilliant Standard Poodle, who appeared to spend much of her day trying to figure out how to “beat the system.” Others, the more biddable ones who came hard wired to want to work as a team (far more rare than the former!), not only rapidly learn learn the cues we teach them, they are delighted to perform them just to be part of the social group.

Clearly, “smart” means different things to different people, and in dogs as in people, there are different kinds of intelligence. This was emphasized in a new study by Bradley Philip Smith and his colleague Carla Litchfield in Animal Behavior (Vol. 80, No. 1, 155-162) comparing the problem-solving ability of domestic dogs to that of dingoes.  Dingoes derived from domestic dogs, but have been independent for so long they are believed to represent a wild-type canid more than a companion animal.

The study asked dingoes and dogs to solve a simple problem known as the “detour task.” In this case, a bowl of food was visible behind a transparent barrier. In order to get the food, the animals had to move right or left, detouring around the barrier to get the food. The dingoes figured it out in an average of about 20 seconds, while the dogs, bless them, were more likely to paw, whine, bark and look at their humans to “fix it.” I find this especially interesting, because I have had to teach all my working dogs the cue “Go around.” On a farm you often find you and your dog on opposite sides of a fence, and when they are young they seem to need help understanding that sometimes to go forward you have to go sideways first (to get through the gate). They learn it rapidly, although some learn it A LOT faster than others. Wolves, by the way, are adept at figuring this out for themselves, performing more like dingoes than dogs.  Even if they have been raised by people, they don’t look to us to help them solve problems like dogs do.

However, dogs are better at problem-solving that includes a social relationship. You’ve no doubt heard about the studies in which dogs are better than wolves or chimps at following an arm pointing toward something, usually hidden food. (I should note that Pam Reid, at the BFF conference, stated she has the same skepticism I do about the reports that young puppies automatically follow a point. In our experience, young dogs just sniff your hand if you extend your arm and point toward something. However, they seem predisposed to learn to follow a point, which is important and significant.) It makes such sense that domestic dogs would be better at social communication with humans than wild-type canids, while dingoes and wolves are better at problem solving on their own. As has been argued (including on this blog on 2/2/2009) dogs are excellent tool users… it’s just that we are the tools being used!

A question for you: I’m starting to work on my talk, an Introduction to the APDT Canine Cognition Symposium in Atlanta, and of course am doing a lot of thinking about well, thinking. Here’s my question:  If you could ask scientists to do research on canine cognition and your dog’s mind, what would you ask them to try to figure out? What do you wonder about? Do you wonder if your dog’s memory is like yours? If she understands death? If he can count? I’d love to hear your thoughts.

MEANWHILE, back on the farm: Chaos reigns. We’re having several of the old Birdseye Maple floors refinished, and so the house is torn asunder for several days. It makes everyday routines into adventures–you’ve all been there I’m sure. But the weather broke though for a few days, and it actually felt cool (ooooh, cool!) yesterday. Sheep, dogs and humans are all extremely grateful.

And speaking of grateful, I am grateful for your kind responses to my last blogs. I will indeed continue to interweave information about training and behavior with with my own experiences with my dogs. I’ll admit that sometimes it is a struggle–we are hard wired to pay more attention to the negative than the positive and it takes a goodly number of thoughtful and kind replies to counteract the that say:  “I’m going to stop reading your blog because you just pass dogs around as if they were old shoes.”

Here’s my old shoe now, the wonderful Mr. Will, taken a minute ago, dozing by my feet as I write.