“It builds precision.” “It teaches systematic searching.” “It develops independence.” “It improves odor discrimination and gives you a beautiful final indication.”
These are the arguments used to promote brick wall training. And I understand the appeal. A dog searches a row of bricks, investigates small openings and gives a clear indication. The handler delivers the reward. It looks organized, controlled and convincing.
But after more than 25 years of deploying dogs in special covert operations, I ask a different question: what does this exercise prepare the dog to do when the training picture disappears? Because that is where our training has to prove its value. I am increasingly uncomfortable with the hype around brick walls. The promises sound good, but we need to examine what the dog actually learns, how we verify it and whether that learning meets the requirements of the job. A useful exercise needs a defined purpose and evidence that it delivers.
The promise of better odor discrimination deserves particular scrutiny. Placing distracting materials around a target sounds straightforward. But that assumes we know which locations contain target odor and which do not. With repeatedly used porous bricks, that assumption can become questionable.
Target odor can remain after a training aid has been removed. Handling can transfer it between locations. Depending on the substance and conditions, material may retain and release odor over time. Yesterday’s target location may still provide relevant information today.
Moving the hide does not automatically reset the wall. Imagine a dog investigates a previously used cavity and detects residual target odor. The handler knows the current hide is elsewhere, so the dog receives no reinforcement. A moment later, it finds the fresh hide and gets rewarded. From the handler’s perspective, the dog has found the correct location. From the dog’s perspective, one presentation of target odor produced nothing while another produced reinforcement. Repeat that pattern, and we risk teaching concentration discrimination: disregard weaker presentations and respond to stronger ones. That is a serious concern for operational detection. Research on odor concentration has shown that dogs can learn to suppress responses to lower concentrations while continuing to respond to higher ones. Applying that finding to a contaminated wall is an inference, but it is a credible training risk we should address. I am not claiming every brick wall is heavily contaminated or every dog trained on one loses sensitivity. I am saying that, without contamination controls, we may reward a result we like while overlooking the lesson that produced it.
For me, weak available odor is valuable information. It may be the first sign of a concealed source. I want the dog to investigate it. Of course, some operational tasks require distinguishing a current source from residual odor. That is a legitimate objective. But it needs explicit criteria and careful validation. A stronger odor presentation is not a universal guarantee that the source is current, accessible or operationally significant. We cannot let accidental contamination design that lesson for us.
There is another assumption I regularly challenge: if a dog detects a strong training sample, it will naturally detect much weaker presentations. That transfer must be demonstrated. In a controlled study, DeChant and Hall found that training with progressively lower odor concentrations improved detection thresholds compared with continued training at the original concentration. The study used a particular odor and laboratory task, but its practical message matters: sensitivity to weak odor deserves deliberate training.
A wall full of previous odor placements is not a substitute for that process.
Then comes the promise of precision. I often see precision defined as a dog placing its nose at a very small opening. Smaller holes are presented as greater difficulty and therefore greater achievement.
But the size of a hole tells us little about the odor challenge reaching the dog. A small opening can release a detectable odor stream. A larger opening can present very little available odor. Concealment, airflow, temperature, source characteristics and surrounding materials all contribute to the search problem.
Operational precision means locating and communicating relevant odor information under the conditions of the task. Sometimes the source is accessible. Sometimes it is behind a panel, inside a vehicle or beyond the dog’s reach. The useful sampling point may be displaced from the physical source.
I want the dog to follow the odor information available, investigate appropriately and communicate clearly. Requiring an exact nose position can become an obstacle if we mistake a trained posture for proof of better detection. A beautiful indication tells us about the indication. We still need to assess the search that led to it.
The argument for systematic searching also needs context. I value coverage. An operational team must search methodically enough to avoid leaving important areas unchecked. However, a repetitive sequence of inspecting every cavity is only useful if it supports the deployment requirements.
Many operational searches involve large surfaces, numerous objects and limited working time. We have to manage the dog’s energy and attention across the whole task. If a dog spends several minutes examining a short piece of wall, I want to know what those minutes achieved. Was it resolving a genuinely difficult odor problem? Or performing a familiar ritual of checking hole after hole? Those are different learning outcomes.
I do not advocate rushing dogs past relevant information. Speed without sensitivity is useless. Equally, prolonged searching without useful coverage is not automatically thoroughness. I want efficient scanning, followed by focused investigation when the dog encounters information that warrants it. The balance depends on the mission. A detailed forensic search and a rapid security sweep have different requirements. We should define those requirements before allowing an apparatus to shape the search behavior.
Independence is another attractive promise. The handler taps a neutral brick, gives a starting cue and steps back. The dog continues searching. Visually, that looks independent. But distance from the handler does not prove independence from handler information.
In many wall search videos, the reinforcement appears immediately after an indication, or even after a subtle change in behavior at the intended location. That can suggest the handler already knows where the hide is. Fast reinforcement alone cannot establish what the handler knows. Precise timing is valuable, and we should preserve it. My concern is the information surrounding that timing. A handler who knows the answer may look at the target, slow down near it, alter body position, prepare the reward or become still when the dog approaches. These changes can be unintentional. The dog may learn to use them. This is the Clever Hans problem: an animal’s apparent success can depend partly on subtle human cues that the person does not realize they are giving.
Research by Lisa Lit demonstrated that handler beliefs can influence detection team outcomes. That does not prove any particular video shows cueing, but it gives us a strong reason to control for it. If handler knowledge repeatedly supports success from the beginning, we may build dependence while celebrating independence.
A reliable detection dog needs substantial experience in double blind scenarios. Neither the handler nor anyone interacting with the team during the search should know the target location or whether a target is present. Reinforcement can still be accurate through a properly designed verification process. We do not need to choose between good timing and honest assessment. Double blind work also develops the human side of the team. It requires the handler to observe, tolerate uncertainty and make decisions without an answer in their head. That is operationally relevant independence.
The promise of generalization deserves the same attention. Moving the wall into another room, taking it outside or changing its height introduces useful variation. But the wall remains a familiar apparatus.
A dog can become successful across several settings while still depending on that recognizable search picture. I use structured equipment too. Controlled apparatus can help isolate a skill and make learning easier to observe. My standard applies equally to walls, wheels, lineups and other training tools. Success on the equipment is a starting point. Transfer has to be tested. Can the dog perform on unfamiliar vehicles, luggage, rooms and surfaces? Can it detect targets presented differently? Can it maintain performance when the expected arrangement is absent? We need to measure that transfer instead of assuming it.
We should also ask how often the wall contains no target. If every search ends with a rewarded find, the exercise can create an expectation that an answer must be somewhere. When searching produces no reward, the team may continue until an ambiguous behavior is accepted. Blank searches help us practice a different outcome: complete the assigned coverage and finish without a target indication. That requires confidence from both dog and handler. It is much harder to demonstrate in a short video, but it belongs in the foundation of reliable detection work.
Finally, there is the attraction of the perfect final response. I prefer a clear, stable indication. A sit can be extremely useful. Avoiding biting, scratching and unnecessary contact matters too. However, indication criteria must fit the task. A dog may encounter odor from an inaccessible source or in a position where sitting with its nose at a cavity is physically impractical. We need clarity that survives variation, rather than a picture that only looks correct on the apparatus. This is why I question the hype.
Social media makes a neat sequence easy to admire. It rarely reveals the contamination history, previous attempts, blank trials, handler knowledge or performance in unfamiliar environments. Before adopting the exercise, I want answers about sensitivity, false indications, coverage, search time and transfer. I want to know whether the team performs when nobody nearby knows the answer.
A brick wall may have a place within a carefully designed program. But its value depends on a specific objective, controlled odor presentation and demonstrated transfer. My responsibility is to prepare dogs and handlers for the conditions they will face. That means protecting sensitivity to weak odor, developing efficient coverage, preventing dependence on human cues and repeatedly testing the team under realistic uncertainty. The environment will not arrange itself into convenient holes. It will not tell the handler where the target is. It will not guarantee a find.
Our training must prepare the team for all three realities.
The research discussed supports the concentration learning risk, the benefit of explicit low concentration training and the influence of handler beliefs on detection outcomes.
