Are We Asking Our Dogs the Right Questions?

Sometimes the most interesting lessons about animal behavior happen when we are not training at all. My dog Vaya is almost always around me. When we train, she is obviously there. But even when I am busy with administration, answering emails, developing new training systems, or writing, she is usually somewhere close. Except for those moments when she suddenly disappears. And after looking around, I often find her in another room. She has discovered that she can use a chair to climb up to the windowsill. There she stands, quietly looking outside, watching the world.

I love watching her do this. Not because it is a particularly impressive behavior. Not because I trained her to do it. And certainly not because there is a reward waiting for her. I love it because it reminds me of something we sometimes forget as professional dog trainers. Our dogs have their own lives, interests, motivations, and ways of experiencing the world. Their world is much bigger than sleeping, eating, sniffing, and performing the behaviors we teach them. Vaya is not waiting for my next command. She is exploring her environment in her own way. And that makes me think. What is she watching? What attracts her attention? What information is she gathering? Why does she choose this particular place? I can make assumptions, but do I really know? Probably not. And that brings me to a question that has fascinated me throughout my career. Are we humans actually asking our dogs the right questions?

One thing I regularly tell my students is that training animals is not only about using your brain. You also need to use your heart. And I really mean that. During my career working with operational dogs, developing new training methods, and learning from some extraordinary animal trainers and scientists, I have become increasingly convinced that good animal training requires both. Your brain helps you understand learning theory, classical conditioning, operant conditioning, reinforcement schedules, stimulus control, and the scientific principles behind behavior. It helps you design experiments, collect data, develop protocols, and evaluate results objectively. But your heart reminds you that the animal in front of you is not a machine. It is a living organism with its own biological history, emotions, motivations, preferences, and limitations. An animal that experiences the world differently from us. And perhaps most importantly, an animal that does not necessarily understand the questions we are trying to ask. I have always believed that science and empathy should strengthen each other.

For me, using your heart does not mean replacing scientific evidence with feelings. And using your brain does not mean ignoring the animal's individuality. We need both. Because sometimes the biggest limitation in our training is not the dog. It is the way we have designed the training.

When I see Vaya looking out of the window, I often think about a remarkable scientific paper published in 1961. The Misbehavior of Organisms, written by Keller and Marian Breland.

The Brelands were pioneers in the practical application of operant conditioning. They had extensive experience training many different animal species and had worked within the scientific tradition established by B.F. Skinner. But something interesting happened. They discovered that animals did not always behave according to the predictions of their conditioning procedures. For example, raccoons trained to place coins into a container sometimes began rubbing the coins together and manipulating them in ways that interfered with the trained behavior. The animals were not simply failing to learn. Their natural behavioral tendencies were influencing what they had learned. The Brelands described this phenomenon as instinctive drift. Their observations helped demonstrate that the biological characteristics of an animal can influence the outcome of conditioning. What I find so inspiring about this paper is not simply the scientific discovery. It is the attitude behind it. They observed something that did not fit their expectations. And instead of blaming the animals, they questioned their understanding. That is a lesson I believe every animal trainer should remember. When an animal does something unexpected, our first reaction should not automatically be to correct the behavior or increase the reinforcement. Sometimes we should simply observe, reconsider our assumptions, and ask another question.

Let me give you an example from my own experience. Hard surface tracking. For many years, I have been fascinated by the ability of dogs to follow human scent across surfaces such as asphalt, concrete, and other environments where traditional tracking becomes extremely challenging. I have seen dogs demonstrate remarkable capabilities. I know from experience that dogs can work with scent information on hard surfaces. But I also know how incredibly difficult it can be to develop a reliable, repeatable training protocol for this type of work. And this is where I start questioning our approach. What exactly is the dog detecting? Are we expecting the dog to follow deposited human odor? Disturbed environmental material? Microscopic particles? Individual scent signatures? Or combinations of information that change depending on the surface, weather, and environmental conditions?

Perhaps we are sometimes combining several different scent phenomena under the single label of tracking. And perhaps the dog is using a different strategy from the one we think we are teaching. Think about that. If we do not fully understand the information the dog is using, how can we be certain that our training protocol is teaching the behavior in the most effective way? We might be reinforcing the correct outcome while misunderstanding the process that produced it. Or worse, we might be trying to teach a strategy that does not fit the way the dog naturally processes the available information. I am not suggesting that learning theory is wrong. Quite the opposite. I am suggesting that we need to understand the variables controlling behavior before we can make the best use of learning theory. Perhaps the question should not always be: "How can I train this dog to track on asphalt?" Perhaps we should first ask: "What information is actually available to this dog on asphalt, and how can I create conditions that allow it to use that information reliably?" That is a very different starting point. And it may lead to a very different training method.

Another fascinating example comes from detection dog training. For years, I have been interested in substances with extremely low vapor pressure. Some materials release so little of the parent compound into the air that conventional chemical analysis can struggle to identify a meaningful vapor signature. Yet dogs can sometimes be trained to locate and recognize materials containing these substances. How is that possible? There are several scientifically plausible explanations. The dog might not be detecting the parent compound itself. It might be detecting associated volatile substances, degradation products, impurities, or compounds originating from the manufacturing process. In other situations, microscopic particles or other forms of material transfer may play a role.

And there is another possibility that particularly interests me. Perhaps certain chemical signals are present at concentrations or in forms that our current measurement techniques, sampling methods, or experimental designs do not adequately capture. That does not mean dogs possess some mysterious ability that operates outside the laws of chemistry or physics. It means we may not yet understand every part of the detection process. And there is an important difference between saying: "Our equipment did not detect a measurable chemical signal."

And saying: "There is no chemical signal available." Those are not necessarily the same conclusion.

In fact, recent research has continued to investigate this very problem. A 2026 study examining potassium chlorate, a substance with negligible vapor pressure, identified microscopic airborne particles that could potentially contribute to canine recognition.

This does not prove exactly what dogs perceive in every detection situation. But it demonstrates why looking beyond conventional vapor measurements can be valuable. For me, this is where science becomes truly exciting. Every time a dog demonstrates a reliable capability that we cannot fully explain, we have an opportunity to investigate. Not to invent extraordinary explanations. But to design better experiments.

During my career, I have met many excellent trainers and scientists. I have also met people who seemed absolutely convinced that they already understood what dogs could and could not do. That worries me. Not because scientific knowledge is unimportant. Scientific knowledge is essential. But there is a difference between understanding what has been demonstrated and assuming that everything outside our current understanding must be impossible. Of course, extraordinary claims require convincing evidence. And anyone who knows my work understands how strongly I believe in objective testing, careful control of variables, and blind or double blind experimental designs.

The Clever Hans effect alone should remind us how easily human expectations can influence what we believe an animal is demonstrating. But scientific discipline should never become an excuse to stop asking questions. When a dog appears to do something unexpected, I do not want to immediately believe it. I want to test it. And if the result survives proper controls, I want to understand it. That is the difference between curiosity and wishful thinking.

We should not accept every remarkable performance as proof of a new capability. But neither should we dismiss a potentially important observation simply because our current explanation is incomplete.

Modern dog training is full of techniques, protocols, equipment, and systems. I develop many of these myself. I strongly believe in structured learning, measurable behavior, carefully designed reinforcement, and objective evaluation. But sometimes I wonder whether our focus on controlling behavior causes us to overlook other valuable information. We spend enormous amounts of time asking our dogs to perform behaviors.

And then we measure how successfully they respond to our instructions. There is nothing wrong with that. It is a necessary part of professional training. But how often do we create situations where the dog can show us something we did not specifically ask for?

I sometimes think we are so focused on the answers we want from our dogs that we forget to consider whether our questions make sense from their perspective. This does not mean abandoning structure or experimental control. It means using observation to develop better hypotheses, which we can then investigate systematically. Perhaps the dog is giving us valuable information long before we recognize it.

One of the greatest influences on my career has been learning from trainers who understood that animals are not simply objects to which we apply conditioning principles. They are active participants in the learning process. My experiences with Bob and Marian Bailey helped reinforce this perspective. They taught me the importance of precise observation, understanding behavior, questioning assumptions, and designing training procedures around the animal rather than expecting the animal to fit our expectations. And after all these years, I am still learning. I still develop new training protocols. I still make mistakes. I still discover things that challenge my previous understanding. And I hope that never changes. Because the moment I believe I have nothing left to learn from a dog is probably the moment I should stop calling myself a trainer.

And so I return to Vaya. Standing on a chair, looking out of the window.

Just a dog interacting with her environment. I do not know exactly what she is experiencing at that moment. Perhaps she is watching people passing by. Perhaps she is interested in movement, sounds, or something I have not even noticed. Perhaps standing there has become an enjoyable part of her daily routine. I do not need to invent an explanation. The observation itself is enough to make me curious. And that curiosity is valuable. It reminds me that Vaya has a world of her own. A world that overlaps with mine, but is not identical to it. A world full of information that I may never experience the way she does. And perhaps that is one of the most beautiful things about working with animals. They constantly remind us how much there is still to discover.

I often tell my students that training animals requires both your brain and your heart. Your brain gives you the tools to understand learning, analyze behavior, and test your ideas. Your heart gives you the patience, respect, and curiosity to recognize that the animal in front of you might be experiencing something you do not yet understand.

The humility to accept that our current knowledge is incomplete. The humility to recognize that sometimes the animal may be responding to information we have overlooked. And the courage to question our own training methods when the results do not match our expectations. Maybe the next major breakthrough in detection dog training will not begin with a new piece of equipment, a complicated scientific instrument, or another training technique. Maybe it will begin with a trainer watching a dog do something unexpected. And instead of saying, "That is not possible," asking: "Why is the dog doing that, and what am I missing?"

Because sometimes the most important question is not whether our dogs can learn what we want to teach them. It is whether we humans are intelligent, curious, and humble enough to learn what our dogs are trying to show us. And sometimes, all it takes to remember that is a little brown dog standing at the window.

Thank you, Vaya. You keep reminding me that being a good animal trainer is not about having all the answers. It is about never stopping asking questions.