Did We Ever Teach the Search? - A Detection Dog Training Thought Piece

Every now and then I find myself questioning something I’ve taken for granted for years.

This is one of those moments.

It started with a simple question:

In our Detection dog training systems, did we ever actually teach the search?

At first glance, the answer seems obvious. Of course we did. We spent hours teaching dogs to search. We hid tennis balls, built independence and rewarded distance. We developed dogs that would confidently leave the handler and work large areas.

What else could that have been?

But the more I thought about it, the less certain I became.

Like many trainers of my generation, I learned within systems that taught the search first. Or at least that’s what I believed we were teaching. Looking back, I’m not convinced we were.

We taught dogs to travel, leave the handler, go from A to B and cover ground confidently and independently.

A to B

If the dog encountered a significant amount of target odour during that journey, everything changed. It switched from travelling to working odour, solved the plume, located source and indicated.

The system worked. In fact, it produced some outstanding operational dogs.

The operational requirements of the day demanded independent dogs capable of working well ahead of the handler. This was particularly important in explosive detection, where stand-off distance was an obvious advantage.

This isn’t a criticism of those systems. I trained in them, taught them and believed in them.

Those who trained during that period will probably remember dummy drops. I certainly do.

The intention was to get the dog to travel the distance we wanted into a search area: the length of a box on a route search or the length of an HGV on a vehicle search.

Sometimes I’d walk, but normally I’d perform.

I’d throw the tennis ball into the air, catch it, dance around, wave my arms and chirp constantly to the dog. The whole performance had one purpose: to convince the dog that I’d hidden its precious tennis ball somewhere out there when, in fact, it was still in my pocket.

Then I’d return to the dog and, to sell the lie, spread my fingers wide to prove both hands were empty.

I now affectionately refer to this as “Jazz Hands”.

While I was performing, the handler was exciting the dog and keeping it focused on me and my tomfoolery. When released, the dog would explode towards the last place it had seen me.

At the time, I believed I was teaching the dog to search.

Looking back, I think I was teaching it to travel to where it believed the reward had been hidden.

There’s a difference.

As long as there was plenty of available odour, that distinction didn’t matter very much. The dog travelled until it encountered enough odour for the nose to take over. It then worked to source and indicated.

The system succeeded.

But what happened when the odour wasn’t generous? What if there wasn’t enough available odour to stop the dog in its tracks and only the faintest trace existed?

That question stopped being theoretical for me during an assessment.

We failed and had to retest. I also had to live with the fact that the “company trainer” hadn’t passed.

The dog hadn’t failed because he couldn’t solve odour, indicate or remain motivated. He failed because he never acquired the odour in the first place.

That forced me to ask whether I had actually taught the dog to search or simply taught it what to do once it found odour.

That failure stayed with me for a long time. It still sits with me.

Large odour forgives poor searching. Small odour punishes it.

Our solution at the time was simple. We tasked the dog into areas we thought it had missed. The dog would go out and we would bring it back.

“Check that corner.”

“Search under those seats.”

“Go behind that cabinet.”

The handler gradually directed the dog into the places where experience suggested odour was most likely to exist.

Again, it worked, but other questions began to form.

How many extra miles was the dog doing? What if I didn’t have to task it into odour? What if the dog wasn’t simply travelling until I redirected it, but actively searching every step between A and B?

I didn’t want it waiting for enough odour to interrupt its journey or for the handler to point it somewhere else. I wanted it searching continuously.

Those questions changed the way I looked at almost everything.

I realised I was already backchaining every other detection behaviour.

I taught the reward first, followed by the mechanics of the trained final response. I say “mechanics” deliberately because, at that stage, I am only teaching the dog how to perform the behaviour, not when to perform it. The odour cue comes later.

Once the mechanics are established, I introduce target odour. Only then does the trained final response acquire meaning.

Everything builds logically from the behaviour before it.

I was backchaining every detection behaviour except the search itself.

That eventually developed into what I now call the First Corner Principle.

The first iterations came from my work with High Assurance Search Dogs in the Army. Those dogs had to search steadily on the way out because the consequences of missing could be catastrophic.

People often assume the First Corner Principle is about room patterns, reward placement or getting the dog to run to the first corner.

It isn’t. It’s about teaching the dog to search for odour.

Initially, the exercise is simple. The dog repeatedly finds a very small amount of target odour in the same place. Chair line-ups are particularly useful for this.

The sequence is straightforward:

Search the first item. Find odour. Perform the trained final response. Receive the reward.

Then repeat it. Again and again and again.

Very quickly, expectation develops. The dog arrives at the first item expecting to solve the same problem it solved seconds before.

During one reward period, I change the game. I move the dog to an identical item or replace the first item containing odour with one that doesn’t.

The dog offers the trained final response out of expectation. It gives me a small false indication. It makes a mistake.

I want that moment, that mistake to happen. I have built the exercise to get there.

Nothing happens. There is no reward, and now the dog has a problem.

“It was here…”

The dog checks.

It then begins searching nasally for where the odour could possibly have gone and finds it on the second item.

The dog has learned something valuable. It must engage its nose from the start. If the odour isn’t on the first item, it must continue searching until it finds it.

For me, that is the moment the search truly begins: the moment the dog learns to use its nose to hunt for odour.

Today, I don’t want a dog travelling from A to B, hoping to encounter enough odour along the way and relying on me to direct it towards the faintest traces.

I want it searching thoroughly from A to B.

Searching A to B

If it encounters significant odour halfway through that journey, perfect. The journey ends immediately and the dog begins answering a different question:

Where is the source?

But if the odour is weak and only the faintest trace exists, I don’t want the dog waiting for me to task it into odour. I want it searching already, with every step, every breath and every metre.

So, did we ever teach the search?

I think we taught dogs to travel. We taught independence and distance, and we taught dogs how to solve odour once they encountered it.

What I’m less certain about is whether we deliberately taught them to actively acquire odour.

That question changed my training.

For me, searching isn’t simply moving through an environment. It is the dog continuously asking:

“Is the odour here?”

And then:

“What about here?”

Only when the answer becomes “yes” does the second search begin: the search to source.

I’m not convinced the old systems were wrong. They solved the problems they were designed to solve and produced some exceptional operational dogs.

I’m simply not asking the same question anymore.

Today, I’m less interested in how far a dog will travel than I am in what it is doing with its nose at every step of that journey.

If I could go back and watch my younger self dancing around a training field, juggling tennis balls and finishing with Jazz Hands, I’d smile. Those methods made me the trainer I am today.

But I’d also tell him:

“Tom, you’re teaching the dog where the reward might be. One day you’ll realise it’s far more important to teach the dog how to look for odour.”

Perhaps that’s the search I should have been teaching all along.

Tom Moir

ScentMasters UK

Trainer. Coach. Student of searching.