For anyone who may just be joining us, in the last hour we talked about what I’m doing up here today. 

It’s not to wow you with stories of how good a shooter I am or how cool I used to be. Frankly, all of that would probably be pretty disappointing, not to mention boring.

Instead I’m here to share some things I’ve learned about training that I initially stumbled upon, almost by accident because I have spent a lot of time, both in the military and as a security contractor, working in environments where we didn’t have everything we “needed” to train effectively.

We talked about what’s wrong with the training industry, by which I don’t mean tactics, techniques, or people.  The problems, at least at a systemic level, aren’t with instructors or what they teach. Instead, the problems are structural in nature.

I’ve been working on this for nearly 20 years, and I’ve distilled it down to two main issues.

First, we don’t train people in the same way that their brains learn.

Second, we don’t have a good way of measuring success in training.

Keep those two things in mind, and let’s shift gears for a moment to talk about the architecture and function of the human brain.

Why is this important?

Is it because everyone needs to turn into a psychology and neuroscience nerd to teach shooting and tactics? 


It’s important because for us, and I’m collectively referring to us here as instructors in this field, our mission objective is to successfully put information into our student’s brains.

In other words, if we frame this in “tactical”, mission-planning terms, the brain of the student is literally the terrain that we fight on as instructors.

In an operational setting none of us would choose to go work in a place where we didn’t understand at least the basic terrain, or environment.

Allow me to suggest that we shouldn’t do it as instructors either.

Just like in operations, your chances of succeeding go way, way up if you know and understand the terrain you’re fighting on.

So, what I’m going to here is present a very simple terrain model of how the brain works.

We aren’t going to get into the biological function or physical structures of the brain, instead we’re going to frame it as an information system.

I am a former navy guy who has ended up in a number of unique situations where I had to run training in less than ideal conditions, started looking for a better way to do things, read a lot about neuroscience and psychology and wrote a book about what I learned. I’m not a neuroscientist.

So, I’m obviously not going to lay out the entire, detailed function of the human brain here.

In fact, nobody could actually do that.

Even with all of our modern technology, brain research really is still in its infancy, and we are just now beginning to have the tools to start trying to understand it.

I’m going to do is instead provide a model for understanding the practical, systems-level results of the processes that make-up brain function.

If you want to see the underlying science, it’s laid out in the bibliography of the book Building Shooters.

In our basic model there are three memory systems, which, if we use a computer analogy, you can think of as three separate hard drives.

They are short-term memory, long-term declarative memory, and long-term procedural memory.

From the perspective of function, short-term memory works kind of like RAM, or Random Access Memory, in a computer.

For our purposes, this means that, while information comes and goes from this place, nothing gets stored there permanently.

Short-term memory is also really small.  If we laid it out in comparison to the rest of the brain, it might look something like this. 

The next system is declarative memory. This is what we normally think of when we think of education and learning, because most of that is designed to get information here.

Again, using the computer analogy, this is like a 10 TB hard drive in the brain.  It has lots and lots of storage space.

What’s unique about declarative memory is that it stores memory that is accessed consciously. In other words, you have to get what’s in it on purpose.

The last system is procedural memory. Much like declarative memory, this is a huge hard drive with virtually unlimited storage space.

However, the big difference is that procedural memory is a hard drive that only stores information that is accessed unconsciously. 

In other words, you can’t really get to it on purpose, it is information that just comes out without you thinking about it.

Now here’s the thing, one of the very unique things about procedural memory is that it is the ONLY one of the three memory systems that you can get information out of during the stress response.

I’m sure all of you are familiar with the basics of the physiology of critical incidents. Surprise, massive amounts of acute stress, shocking events,

All of these things cause the body and brain to release a series of chemicals into the brain and bloodstream that change how things work.

This is sometimes called the sympathetic nervous system, and has some unique effects such as shaking, slowtime, and auditory exclusion.

Again, I expect that most people in here are intimately familiar with these things, mostly from first-hand experience.

One of the big “ah-ha” moments I had when researching this stuff is that the chemicals that dump into your brain during times of extreme stress actually act as a kind of “switch” – they trigger access to procedural memory and also preclude access to the rest of the brain.

So, from our perspective as trainers, let’s think about what this means to us.

I mentioned earlier that most of our educational research and method is intended to guide students towards getting information into their declarative memory systems – so they can be tested on it.

I actually would argue that most of our education and training structures are really designed more to make the information available to the student than they are to facilitate learning—then the student can learn it on their own if they put in the effort.

But, either way, the vast majority of what formal education and training programs do is try to put information into declarative memory.

That might be fine if you’re an accountant or analyst or something. For us, here’s the problem.

Even if the student learns everything, can perform all the skills flawlessly, and never forgets them, it might not matter all that much if they stored the information in the wrong memory system. 

Sure, it works on the qual…

Everybody might even think this person is a really, really good shooter.

But still those skills might not be available in a gunfight.

Why? Because they’re stored in the wrong part of the brain.

The lesson here is simple. If we want it to happen during a critical incident, we need to put it in procedural memory.

Let’s stick with a computer analogy.

When new information, by which I’m referring to just about anything—skills, knowledge, etc.—when new information goes into the brain the first thing that happens is that it hits a filter. You can think of it kind of like a computer’s firewall.

It’s really important to understand that getting through this filter really is our first challenge as instructors—and that we fail to accomplish doing this all the time.

Anybody ever have a student not learn? Or that apparently learned something different than what you thought you taught them?

This might be part of why.

Once we get the information we want through the filter, it’s now in short-term memory.

There are a couple of things that are really important to understand about short-term memory.

We already know that it’s small, it has a very limited amount of storage space.

It ALSO doesn’t store anything permanently.

You can have information in there, you can work with it while it’s in there and do stuff with it, like pass a qualification for instance, but this doesn’t mean that information is going to be retained.

So, it’s really small and nothing stays permanently.

What this means, is that if I’m going to put something new in, I first need to either push out something that’s already there or, alternatively, overwrite something that’s already there—which might end up corrupting both sets of information.

The last thing we need to know about procedural memory is that it appears to be compartmentalized.

This means that just because it can hold, say, 10 pieces of information, doesn’t mean I can pick any 10 pieces of information and put those in. 

Maybe I’ve got 10 slots for information  here, but I actually only have one space that will store visuomotor skills.  And one that will store “hand/tool” skills.

Let’s assume that’s true.

This means that I can only put one visuomotor skill and one hand/tool skill into the brain at the same time. 

I still have plenty of space in short-term memory, but I can’t put any more of these types of information into the brain.  – There isn’t any place for it.

Now – that is a really high-level overview of what’s sometimes called cognitive infrastructure.  In more of our language, it’s a map or a terrain model for the brain’s learning functions.

There are two other things I want to cover here and those are the two sides of the fundamental principle of neuroplasticity.

That sounds really “sciencey”, but basically it just means brain change. As in actual, physical changes being made to the brain.

For a long-time it was thought that the adult brain couldn’t change.  We now know this to be untrue. Our brains are always changing, and what we do as instructors, especially as instructors in this field, has the potential to make permanent changes to our students’ brains. 

Because of that, it’s important for us to at least understand the basic concept involved – which is commonly called Hebb’s Law.

The first part of Hebb’s Law is the most commonly known and is stated “Neurons that Fire together, Wire Together.”

This is fairly intuitive, especially for shooters, and means that the more we use a neural circuit, the stronger that circuit is going to get.

There is a flip side to this though.  That is, “Neurons that Fire Apart, Wire Apart.”

To really understand what this means, for us, let’s look at a very simple example where we are going to vastly over simplify signal movement in the brain. 

Again, this isn’t exactly how it works, but it will illustrate the concepts at work.

Here are two neurons with a wire between them to carry the signal.  In the brain, this wire is called an axon.

If we’re connecting this circuit for the first time, think of it like using bare wire with no insulator on it.

We can get the electricity through that wire, but we need to use a lot of voltage to do it, it shorts out sometimes, and we still get a weak signal on the other end.

However, the more times we use this circuit, the brain starts realizing it’s important, and wants to make it more efficient.  So, as a result, the brain starts putting insulation on the wire.  In real life, this insulation is a fatty substance called myelin.

After a while, this entire wire is thickly coated with insulation.  Now the signal moves through it really easily, using a much lower voltage, and you still get a good signal on the other end.

Now, that’s great if just I want to use this circuit over and over again.


What if I want to tie something else into the circuit?  What if this circuit (existing, insulated circuit) is my draw stroke—from buzzer to trigger press.

Now, here’s another circuit that needs to communicate with it.  This one is decision-making.

I can’t just get another signal, like decision-making, plugged into this circuit now, even if I put a lot of voltage behind it.  There’s too much insulation for the signal to get through from one to the other.  They are now “wired apart” because they were not used to together.

So, the general principal is, if we train these two things separately, even if we get really good at both of them, we’re going to have a really difficult time connecting them together later, especially under stress out in the field.

So, recapping, first is our terrain model of the brain.

The first thing that happens when we try to send information to the brain is that we hit a filter.

Once we finally get through the filter, we put the information into short-term memory.

And what do we know about short-term memory?

It’s really small, it’s compartmentalized, and it doesn’t store anything permanently.

This means that if we want our students to LEARN, it’s not enough to get the information into short-term memory.

We need to make sure that the student’s brain transfers it to long-term memory, without it getting corrupted or lost.

As instructors in this field, we also know something else.

We know that if we want the information to be accessible during a critical incident, we need to put it in the procedural memory system.

And, we also know that if we use and practice things together frequently, they will wire together effectively in the brain.

BUT, if we always use things separately in training, we are actually going to physically disconnect those things in the student’s brains – even if those things need to get used together out in the field.

So, how do we apply all of this?

During the research for Building Shooters I was able to identify 12 things that impact this process of transferring information into long-term procedural memory, which is often called consolidation.

Some of these things have a positive impact, some a negative impact.

They are explained in detail and the science is referenced in the book.

Here, I want to just talk about a few key points.

First, when we first present information to a student, it’s very difficult to know whether or not it will get through this natural “filter” in the brain. 

Maybe it does, maybe it doesn’t. 

What I’m going to propose to you today is—don’t worry about it. 

Just expect they didn’t learn it the first time and then teach it again.

Literally teach the subject to the student with NO EXPECTATION that they will learn or retain it the first time they see it.

The purpose of this presentation isn’t to teach them.  And it’s not to waste your time. Rather, it’s to let their brain’s filter recognize the information set when it sees it again, so it lets it pass into short-term memory.

So, principle one, PRIME first. Teach with no expectation of retention.

Second, account for the limitations of short-term memory.

The fact of the matter is that no matter how much information we are capable of teaching in the course of a day, the student’s ability to learn is limited by their brain function.

Consider this.

When it comes to brand new material, you’re usually maxing out the student’s ability to learn in about 20 minutes, give or take.

Think about that for a second.

If you’re maxing out the short-term memory system’s capacity in 20 minutes, what are you doing for the rest of an 8-hour training day?

Our basic model tells us that you’re corrupting, overwriting, and replacing stuff.

This helps us get the information out as instructors, but it sure doesn’t help the student get the information in.

Certainly, when we are overwriting and corrupting most of what we teach people, especially the fundamentals, we aren’t setting them up for success. We’re doing the opposite. 

So, allow me to propose a radical notion. Let’s not do that anymore.

To avoid any confusion, let me clarify something.

An 8- or more hour training day can be really valuable to a student—in the later forms of learning that occur AFTER the information is in long-term memory.

But, that type of training structure is not very useful for getting the information INTO long-term memory.

So, we prime first. Then, we teach—just as much as we can put into short-term memory.

When we do, we do a lot of repetition so we as instructors can assure it’s the right information and so the brain knows the information is important and is worth storing—wiring it together,

Then we stop, and we let that information transfer into long-term memory.

One of the really interesting, and counterintuitive, things I learned when I was doing the research is that an awful lot of this process happens subconsciously,

And it can only happen when you’re NOT USING the information,

AND some of it can only happen when you’re sleeping.

This means that if we want people to learn, after we’ve put the information successfully into short-term memory we just need to leave it alone for a while.

Most of the science says about 24 to 48 hours.

We need to just let the students’ brains work through the biological processes necessary to move the information to long-term memory.

After it’s been moved? 

Well, now we can’t corrupt it in short-term memory anymore, so we are now free to re-use this space with new data.

Finally, the last technique we’re going to talk about here is a concept that scientists call interleaved training. You can also think of this as chaotic learning.

This doesn’t mean that the training session is in chaos, or completely unstructured.

What it means is that the more of the brain that is involved when you’re using a neural circuit that’s already in long-term memory, the more connected and important to the brain that circuit becomes.

For us this is important for two reasons.

First, it improves the chances that the information is placed into and being used in the procedural memory space. 

Remember that this means that it will be accessible during a critical incident. 

Second, using these types of training methods allow us to build the same neurological machine in training that the students will need to use in the real world. 

In other words, we can wire together the stuff the students will need on the street—and AVOID wiring apart important things like decision-making and shooting.

Now let’s shift gears and talk about that for a second, because it brings us to our second fundamental failure, which is how and what we measure.

Building the right machine.

What “machine” do we use when we do most of our firearms training and qualification?

Let’s consider it.

During most training and qualification we get an audible stimulus (let’s face it, even with turning targets you typically hear the actuator first), then we go perform a pre-known skill sequence. 

Pre-known.  What does this mean?  It means we’re going to the conscious access memory system.

Then we stop and reset for the next iteration or yardline.

Does this resemble what happens on the street?

Not so much…

On the street we have to combine declarative knowledge (context) and stimuli in working/short-term memory (context) before we event start. Leading up to a use-of-force then, there’s a combination of previously unknown audio and visual (combinations of object recognition and motion detection) feeding to pre-cognitive and cognitive processing centers. This is then combined with context, fed to decision-making centers, linked with skills in procedural memory, then looped back to dynamic input from the visual (predominantly motion detection and object recognition) system and audio system, which in turn loops back through the processing and decision-making centers and back to a variety of skills in procedural memory.

In other words, it’s kind of like we teach and qualify people to use a riding lawnmower in the back yard, in preparation for driving a suburban out on the beltway during rush hour.

Does building and measuring the first “machine” to prepare for and determine readiness to use the second machine make sense? 

Does it help us assure good operational outcomes?

We aren’t just failing to build the right circuitry in the brain. If we train A LOT this way, we are actually insulating the shooting-skill-circuitry to a point where it might be really difficult to connect it to the rest of the relevant brain functions.

Allow me to suggest that this doesn’t help us on  the street.  Allow me to further suggest that we should be both building, and measuring the second “machine.”

Here’s another question. 

We all have limited resources.  What percentage of those resources do we put into the first “machine?”

For almost everyone in this room, I’m willing to bet the answer is “most of them.” 

It might even be “all of them.”

So, when I talk about the two fundamental failures, this is what I mean. 

Our delivery method doesn’t match up with the reception method at the other end.

We also aren’t even looking at the right machinery when we’re trying to figure out whether or not it’s working.

Earlier I mentioned an 800 lb gorilla in the room that we never talk about when there are discussions about operational performance of law enforcement (or others who carry guns).

This is that gorilla.

We use a tool that doesn’t work to prepare the wrong machine to achieve a standard that doesn’t mean anything.

Then we wonder why we sometimes don’t get the results that we want.

When things go wrong, it’s tempting for us as armed professionals (in any sector of that industry) to either point fingers or to point out limitations we have in terms of resources. 

There may be valid points along those lines…but I submit to you that there are much more fundamental problems—structural problems—that exist with respect to our systemic training and qualifications.

There’s good news here though. 

The good news is that we can fix it. We really can.

Before we break, let’s take just a minute to quickly look at what both I believe and hope is going to be the future of the training industry.

First, we’re going to start basing the design of our training on the idea that what we’re really trying to do is create a series of neural networks. 

While we can’t predict everything, we do pretty much know what we want those networks to be.

We know where in the brain we want the networks to be. And we know how to put them there, effectively and consistently.

As a result, our training design and training structures will become more of an engineered training system that targets specific information to specific parts of the brain—by design—and that matches the delivery with how the brain receives and stores information. 

Once we get information into the brain, we will then enhance and connect the relevant neural networks together—building the right neural “machinery” for the job.

On the measuring side, I believe we’re going to see significant change in two areas.

First, we’re going to stop measuring—by which I am referring to qualification—machinery that’s not operationally relevant.

As a result of this change, we’re going to stop putting our focus of training effort on developing, the wrong machine. 

This doesn’t mean we stop teaching or measuring fundamental shooting skills. 

It means that we will start qualifying people by measuring them in ways that matter.

Instead of testing the wrong machine, we’re going to start building and measuring what I refer to as the “Real Response” that is required in the field.

In terms of results…

Once the industry as a whole starts making these changes,

I think we’re going to start seeing greatly improved outcomes both in the field and in training.

And we might even see our overall costs and resource requirements go DOWN as well because what we do now is so inefficient that making the changes may actually end up saving us time and money.

While all of this stuff is still pretty early in terms of implementation, I do have a few things I have permission to share.

One is from the DCI Academy in South Dakota. 

About 9 years ago, when all of this was still very much in its infancy, they started looking at restructuring their training program.

I had the opportunity early on to speak with their advanced training coordinator and they have incorporated a lot of the things we’ve talked about here today in terms of training design, as well as many other changes.

Before the changes, they averaged about a 12% failure rate in their firearms program.  Now it’s down to less than 1%.

There’s another police academy, this one is in Arizona,

About two years ago they started implementing some changes to their academy based on these basic concepts we’ve discussed here. They made just three simple changes.

First, they started priming and teaching most of their firearms skills in the classroom, dryfire, before taking students to the range.

Second, they intentionally made an effort to reduce the stress in firearms training by not sharing the qualification scores with the students. 

They never even told them what a passing score was. 

As an aside…this is based on some science that we haven’t talked about here today suggesting that the chemicals released into the brain by stress can often prevent effective learning.

Third, they stopped emphasizing “zones” on the targets, and just told the students “do not miss the target” before each qualification.

That academy historically has seen about a 30% failure rate on qualifications in their firearms program. 

Not as in 30% fail to pass the course, but about 30% of the total qualification courses fired during the program are failed.

During the academy they ran after implementing just these small changes, the number of failures during the entire program went to zero.

Zero qualification failures by any student during the entire academy.

This stuff works, not because it’s based on artificial concepts, some pseudo-scientific mumbo jumbo, what policy makers would like to have happen, or some kind of mystical voodoo or secret sauce that I or anybody else invented.

It works because it’s based fundamentally around how the human brain works

As trainers, is the terrain we work in, and as operators, is the control system for everything we do.

As a result, once we start actually considering how the brain works into what we do…our outcomes can’t help but improve.

Thank you and that concludes today’s prepared remarks.

Subscribe to our Newsletter

Share this post with your friends

"Easily one of the more important books of our time when it comes to preparing police, military and armed civilians for armed lethal combat."
-Kenneth Murray
Author, Training at the Speed of Life
Co-Founder of Simunition