Interview: Robert West, creator of the COM-B model, on the three forces behind every human behavior

In this episode, MichaelAaron and Richard sit down with behavioral scientist Robert West, creator of the COM-B model, to explore the three forces behind every human behavior: capability, opportunity, and motivation. They discuss why behavior change so often fails, how environments shape decisions more than we realize, and how businesses and governments can design more effective interventions.

Transcript

Robert West: [00:00:00] When we're looking at people's behavior, all our
attention and our focus is on the person, and therefore we think, "Oh, g- okay
that's where the source of the behavior, what's causing the behavior lies."
interestingly though , when we're looking at our own behavior, particularly
when our behavior is not necessarily up to a standard that we would wish it to
be, we're looking outwards.
We're not looking at ourselves, and therefore we're much more likely to
attribute the causes of behaviors, and particularly failures, to external events.
MichaelAaron Flicker: That was a clip from Dr. Robert West, professor
emeritus at University College London, and one of the world's leading experts
on behavior change. He's best known for developing the COM-B model, a
simple but powerful idea that for any behavior to happen, three things need to
be in place: C, the capacity to do it; O, the opportunity for it to happen; and M,
the [00:01:00] motivation that drives the choice in the moment.
In our conversation, we went much deeper than just unpacking COM-B. We got
into how it can be a tool that analyzes the major factors that drive behavior
change, and also how this COM-B system can be used to help bring
stakeholders and leaders along to support your behavior change initiative. With
over 900 scientific papers to his name, he spent decades studying what actually
drives human behavior, both at the individual level and at a population level.
It was a truly fascinating conversation. And now, here's our episode.
Welcome back to Behavioral Science for Brands, a podcast where we bridge the
gap between academics and practical marketing. Every week, we sit down and
go deep behind the science that's powering some of America's most successful
brands.
I'm MichaelAaron Flicker.

Richard Shotton: And I'm Richard [00:02:00] Shotton.
MichaelAaron Flicker: And today we're sitting with Professor Robert West,
one of the world's leading experts on behavior change and the science of
motivation. Let's get into it. Professor West, welcome to today's show. We're so
excited to have you with us.
Robert West: Thank you for having me.
I'm looking forward to our conversation.
MichaelAaron Flicker: Richard and I have been very excited about our
conversation. And to ease you in to the episode that we have today, we thought
we might start with a few quick-fire questions to warm up the conversation.
And before we started the show, this one was... We were almost leading into it.
Dr. West, what do your friends and family members think about your passion
for behavioral science?
Robert West: Okay. Yeah they are I think like most of the-- my family, they're
mildly skeptical. We're a rather skeptical [00:03:00] family, I have to say. But
they put up with it. And in fact with my three sons, we work together closely.
I work with my son Jamie, and I work with my other two sons on websites and
stuff like that.
Richard Shotton: Family business, fantastic. Moving on. What is something
that you once believed strongly, but now you've changed your mind on?
Robert West: Oh, that the world is a fair place. I would love to believe that the
world is just and fair.
Unfortunately, my experience over the decades has led me to believe that it's
seriously not.
MichaelAaron Flicker: More on this later, friends. Outside of the COM-B
model, which behavioral science model do you have a lot of respect for?
Robert West: I think that my all-time favorite [00:04:00] model is I guess it's
not really called a model, but people will know it, anyone who knows anything
about psychology, and it's operant learning theory.

It's it's operant conditioning. I think it is incredibly powerful. It's, it's-- At its
very s- heart, it's around the ef-effect of reward and punishment. But what I
think has been so important is that in developing the model, the the scientists
working on it have figured out some really important details about how reward
and punishment work, and how you can make them most effective.
Richard Shotton: And then if you had a magic wand, what preconception
about behavioral change would you erase from people's minds?
Robert West: That it's all common sense. That there's nothing to see here
because we all know how people behave, and all our common sense ideas are
perfectly fine, despite getting [00:05:00] it wrong so often.
Richard Shotton: Yeah. We, w-we did an interview with Dan Ariely
reasonably recently, and he said something fascinating, which was he used to
present his experiments, and then afterwards people say, "Oh that was obvious.
Doesn't everyone know this?" So what he started to do was he would present the
methodology of the experiment, ask people to predict the results, and of course
they were wildly wrong.
It's only common sense after you've been told the results.
Robert West: Absolutely. I'm a big fan of his, actually. I think that he's got he's
got some terrific ideas, and he expresses them really well. And I have in a sort
of informal way done that myself with people.
And what's really interesting, and he will have experienced this themselves, is
when you go, "Oh I'm s- actually, that's not the case," you get two responses.
One is, " you're wrong. It is the case despite the fact that the evidence points in
that direction." And the other one is, "Oh [00:06:00] yeah.
I knew that, yeah, already." So it's very hard to defeat this particular problem.
MichaelAaron Flicker: Final question. What insight from behavioral science
do you think is criminally underused?
Robert West: Wow. I would say probably that the environment in which we
operate has way, way more influence on our behavior than we give credit for.
And unfortunately, this has been something which actually psychologists have
somewhat been guilty of over the years but are now really, I think addressing

rather better than we did. If you look at the sort of psychological theories, they
really focus on what goes on inside people's heads.
When in when you really look at the variance [00:07:00] accounted for, to use a
technical term the environment, both the wider environment and your
immediate environment plays so much greater role than we give credit for.
Richard Shotton: Why do you think people underestimate that so much?
What-- Is there a theory you've got?
Robert West: There is a theory about it actually, which is called the
fundamental attribution error, which I think is rather good. Which essentially its
argument is that when we're looking at people's behavior, all our attention and
our focus is on the person, and therefore we think, "Oh, g- okay that's where the
source of the behavior, what's causing the behavior lies."
Interestingly though, when we're looking at our own behavior, particularly
when our behavior is not necessarily up to a standard that we would wish it to
be, we're looking [00:08:00] outwards. We're not looking at ourselves, and
therefore we're much more likely to attribute the causes of behaviors, and
particularly failures, to external events.
But when we're studying behavior and when we're even as laypeople we it's the
focus of our attention is on what's going on inside someone's head, so that's
where we think the causes must lie.
Richard Shotton: Yeah. So if we're driving down the road and someone cuts us
up, we believe it's 'cause they're a rude person, an idiot.
But if we made a mistake driving, we put it down to this particular stress we're
under or the interplay of the l- the light or something on the windscreen, so
yeah.
Robert West: Absolutely. Yeah. Absolutely. And and not only do, especially
when it's something bad that someone does, not only do we make a, what they
call an internal attribution, we attribute it to the person, we think it's something
about them that's what they call stable.
It's inherent. They are that kind of person. They're the sort of person who does
that. And we [00:09:00] make-- we don't make sufficient allowance, not only
for the external factors influencing the behavior, but to the instability of the

internal factors. We think if someone has a belief or if someone has a desire,
this characterizes them as people.
We don't give enough attention to the fact that our desires go up and down and
our beliefs come and go.
MichaelAaron Flicker: I can't help but ask, based on that jumping off point,
going back to our quick round, quick-fire questions, you used to believe the
world was fair. You now think it is a whole lot less fair. How does that connect
to your life's work in behavioral science?
How does that connect to what you've, y- what you've focused on over the
years?
Robert West: I think the connection arises from the the fact that the more you
study behavior and what causes [00:10:00] it what leads people to do things and
be successful or unsuccessful, and I think this is something that famously
Woody Allen has commented on, the more you realize that, oh my goodness
luck has played a huge role in it.
And so when in our society and I think some societies do this more than others
and not wishing to cast any aspersions, but I think the US society tends to do
this more than, let's say many countries in Europe You say if someone's
successful, it's 'cause of them and what they did and how they are and they
deserve it.
And if someone's unsuccessful, it's because, it's 'cause of their failings and and
it's and they gotta buck up. Now, you can see why although that is it's not a, it's
[00:11:00] not a justified position from a scientific standpoint you can see why
it's part of how a society can operate.
Because one of the things you want to do with- within a certain person's
capabilities is to motivate them to maximize the the outcomes given those
capabilities, and so put more effort in and so on. The problem lies in the fact,
then in the fact that when because there are huge limits to that, and because the
our circumstances so much dictates what we achieve even little things can make
huge differences that it causes a lot of stress and a lot of punishment social
punishment and other kinds of punishment to people which is unnecessary and
unhelpful and actually to the detriment of society.

I think it's the [00:12:00] more you understand the causal networks that lead to
things like human behaviors, the more you think, "Wow," there but for the grace
of God," as it were.
MichaelAaron Flicker: In our podcast, we talk about a wide range of things,
but we also talk about the application to commercial implications.
So we understand, we have an enthusiasm to understand behavioral science and
human behavior, and then we ask the commercial application question. Hearing
you talk about that makes me think about what is a company's role in driving
not just people to buy their products, but to try to use all of the media that they
spend or all of the impact they have on reaching lots of people in a more
positive way.
Y- if you have the microphone to change human behavior, [00:13:00] based on
what you just said is there a positive thing you can do to also make a good
outcome for society? I, it just, it strikes me that for people who have large
budgets and lots of money, both can happen.
Robert West: Yeah. My value system is that I think that the things that matter
most to me are human wellbeing and and long-term prosperity and the kind of
environment in-- our habitat, the environment in which we operate.
And other things that other people value, money, profit, whatever are, are--
would always in, in my value system be secondary to that. And what I think is
quite interesting, I d- I mainly deal with the public sector and governments and
NGOs and so on. [00:14:00] But when I work with communications agencies
and so on who are working for the private sector, one of the things I find quite
interesting is that the stereotype is of this idea, " we just gotta make money," it's
all about shareholder value and so on. But when you dig a bit deeper into what
motivates people to do something, it's often a passion for the thing they're doing
whether it's a startup i- in tech or whatever or it could be anything, some new
invention or it could be something very prosaic, but a passion for something
they're doing.
But also, most people do actually want to do some good, and so I think that one
of the things I would say is that you harness that, and in the old cliché you do
well by doing good. So essentially one of the thing-- Humans are quite
interesting because we're-- i- in the sense that [00:15:00] we we often put up
with a lot of nonsense about things, but we-- I think we can detect authenticity
when we see it, largely.

And if you project authenticity because you are authentic in what you do,
because you do care about your customers, you do care about the wellbeing of
your staff and you're not just pretending in order to meet some bottom line,
which in the end comes and goes.
But if you do care, it comes through, and with luck, because luck plays a lot of
a big role, as I've been saying that will then translate into a successful business,
I think.
Richard Shotton: There's an amazing book on that you, your phrase doing well
by doing well. I love John Kay's "Obliquity," and he has lots of different case
studies around different companies, some of whom [00:16:00] try to prioritize
maximum profit in the short term, others who had a genuine love, let's say it
was a a drug company for the-- a longer term aim.
And what he shows again and again is actually optimizing for the long term
ends up generating the best profits. Now, if you're solely focused on
maximizing money tomorrow, it can often quite quickly blow up.
Robert West: I think that's right. It doesn't always happen. The world is a very
uncertain place, and sometimes you can get very rich very quickly by being a
complete heel
And not caring about anything. But then what have you got? The-- it's s-so
what? Whereas if you've created a successful company that's doing a good job,
that where you've got satisfied customers and where your workforce are happy
and fulfilled, great. That's a much better feeling.
MichaelAaron Flicker: Yeah. And we talk about [00:17:00] building brands,
and so beyond the product itself, the brand is that layer of what-- exactly what
you're talking about, Dr. West. You're-- We're talking about that extra layer of it
stands for something, it means something beyond just the product itself, and it
occupies that space in the buyer's mind.
I think to-- that's really a lovely way to think about it.
Robert West: Yeah I think you're right. The brand is a layer of authen- ideally,
of authenticity on top of what the product is or the services that you're
delivering.
MichaelAaron Flicker: We have asked you to come here today because you
have helped create what is called the COM-B model.

Can you talk to us about what is the COM-B model? And let's start there. Let's
get a base layer before we go deep.
Robert West: Yeah. The COM-B model is a general and very [00:18:00] bird's-
eye view model of behavior that was developed by myself and Susan Michie
my wife and busi- and colleague as a kind of model to help guide people to
devi- devise behavior change interventions, that is to influence behavior.
So it's very much targeted on a s- on a very specific goal, and that is to predict
what is the likelihood of a particular behavior occurring in a particular situation,
in a particular population or individual or group of individuals in certain
circumstances. And the essence of the model is incredibly simple.
It is just blindingly obvious common sense, which is possibly why it has
become popular. And it is this. It is that in order for any of us to do anything at
any moment in [00:19:00] time, we have to have three things in place. One of
them is that we have to have the capability to do that inside our heads and our
bodies.
We have to have the physical capability. We have to have the strength, the
stamina, whatever it might be the sensory abilities and so on, all of that. We
have to have the psychological capability. We have to know what the thing is,
how to do it. We have to have an understanding of what the benefits are and so
on.
And the mental... Or sometimes the mental resilience to be able to put up with
the things that are getting in the way. So that's capability. That's one of the three
pillars of behavior. The second pillar is opportunity, which is the environment
in which you exist must afford you the the scope to do the thing that you're
doing.
Very obviously, if you if you imagine... Sometimes I like to imagine what Jimi
Hendrix would have been [00:20:00] doing in the 15th century without electric
guitars. Would he have been a terrific lute player or whatever? Maybe not.
Maybe he'd have been in the fields, I don't know or milking cows.
But his wonderful talent specifically At the electric guitar with a 100-watt
Marshall stack would not have manifested itself. So the opportunity and we talk
about two kinds of opportunity. The physical opportunity, which is the
resources, the equipment, the time, the money the spaces and facilities and so
on, and the social opportunity, which is the norms that exist within society, the
things that constrain our behavior what you c- what you're allowed to do, what

you're not allowed to do, what you what your friends do, the social cues and
social support and so on.
So you've got capability, you've got opportunity, and then the third pillar
[00:21:00] of behavior is a little bit different, and that's motivation. That's the M
in COM-B. And the concept behind motivation is that essentially, given what
your opportunity is to do and what your capability is to do, motivation
determines what you actually do.
So motivation's job is to prioritize particular behaviors over others. That's the
job of our motivational system in the brain, is to say, "Of all the things I could
be doing right now, I'm gonna do this." and within motivation as has been
identified by many psychologists, and not just psychologists, but philosophers
and anthropologists, you can essentially break it down into what we call
reflective motivation.
That's the sort of the head, as it were, your clever part of your brain, which is
going, "Oh, I've worked out the right thing to do, and I'm gonna do this because
it's a good idea." [00:22:00] And automatic motivation, which is your animal
brain, if you like, which is your feelings of desires and wants and needs and
urges and your habits and instincts.
And one of the propositions within the motivation side of it, which is elaborated
in a theory called prime theory, which I developed before the COM-B model,
but essentially the proposition is essentially it doesn't matter what you think you
should do If you don't actually feel that you want or need to do it, and then have
an impulse to do it.
In other words, your shoulds have to work through wants and needs in order to
influence behavior. In i- in ancient philosophy, this was characterized by the
concept of the rider and the horse, sometimes the rider and the elephant.
Essentially, the rider, the clever part of your brain, has to get the horse to do the
stuff that it needs to do.
You can't [00:23:00] just say, "Oh, I'm gonna do this," and then it happens. So
one of the really important insights, I think, over the last decade or so within
behavioral psychology is this understanding that y- yes, it is important to get
people to think that something's a good idea and to think that it's worth it, but y-
but you've then got a bit of work to do quite often to turn that into something the
person genuinely wants to do or genuinely feels the need to do in the
circumstances in which it could occur.

So those are the three pillars, and the other sort of side of the COM-B model,
which adds a bit of complexity to it, is that all these different components
interact with each other. So capability influences motivation, opportunity
influences motivation, and so on. So what starts as a i- incredibly simple model,
you can elaborate into a sort of systems model, a sort of network [00:24:00]
model.
And you can then use this, as we've been doing with some research over the
past few years, into building more elaborate quantitative models of population
behavior, which can make quite fine-grained predictions.
Richard Shotton: Yeah. So y- you've taken us through the overview of the
model and the three parts, and that's sounds like they're brilliant in theory.
Moving to practice, let's say I know you're very interested in encouraging
people to quit smoking. You're a marketer, you're at the Department of Health,
and you are work on this problem of quitting smoking. H- how would COM-B
work through it i- in practice?
Robert West: In practice, what it does is it directs you to ask questions of the
data that then you can use to build your interventions.
And in the case of smoking, you find something very interesting when you do
that, and that is that [00:25:00] the motivation side of things is great for getting
you to the starting gate for making a quit attempt, and after that makes precious
little difference. A lot of people say, " in order to stop smoking, you've really
got to want to stop smoking."
In a way because i- in order to make a quit attempt, you've got to want to stop
smoking. But what we see after that, once someone's started to make the quit
attempt, all the things that make a difference to the success of that quit attempt
have got nothing to do with motivation. They've got everything to do with
capability and opportunity, particularly in the case of let's say countries like the
UK and the US where we have medications available to help people stop
smoking.
You've got Chantix in the States it's called, and Champix here, or Varenicline or
cyt- and then Cytisine is another one. But essentially you've got some
medications, and you've got smoking [00:26:00] cessation support programs,
and those support programs, they don't really focus on your motivation. They
don't give you you've got to buck up and really want to stop smoking.

That boat has sailed. That job has been done. What they do is they focus on
what are the barriers that are preventing you putting that motivation into effect?
What is your capability? So when you're a government agency looking at how
you can reduce smoking prevalence among by by getting people to quit
smoking, you've got a two-stage job.
You've got your motivational stage, which gets people to the quit attempt, and
then you've really got to put your effort into the capability stage which is
helping to make those quit attempts succeed.
Richard Shotton: Do you think that idea of motivation being like a necessary
but not sufficient condition, do you think enough people recognize that?
It feels like a lot of people who try and change behavior jump straight to
motivation- [00:27:00] and forget about these other drivers.
Robert West: Absolutely. Absolutely. That is that is one of the most common
mistakes I think that's made is to assume that what you're dealing with is a
problem of motivation.
But it doesn't always happen. Sometimes it's the other way around. For
example I used to do quite a lot of research into road safety and driver behavior.
And when I started, the assumption was that novice drivers who crash their cars
quite a lot, hence the high insurance premiums that the problem was lack of
capability.
It's common sense. They've just passed their test. They don't really know how
to drive. Actually, not so much. It turns out that the biggest predictors of crashes
amongst young drivers are much more to do with motivation than they are
capability. It's not the less skilled drivers who are crashing their cars, it's the
drivers who care less about the consequences of [00:28:00] crashing to
themselves and other people.
There's this actually, it's quite a strong relationship between sort of social
deviance and and crashes to the extent that murderers, for example have
something like 10 times the crash rate as other people, and not just 'cause they're
fleeing the scene of the crime.
Richard Shotton: That is amazing.

I didn't know that. That's amazing. I feel very... I feel more proud now of my
crash-free record now. Yeah. I can... I'll use this in court if I'm ever accused of-
Absolutely ... of murder. Yeah.
MichaelAaron Flicker: As we are starting to go deep, I want to just take a pull
back question and ask, I'm listening to this episode, I'm learning about the
COM-B model.
I'm saying either I've heard about it before, I didn't understand it this deeply, or
I'm excited to know more. How do you help people take this base insight of C-
O-M equals B and actually start [00:29:00] to apply it in their work? So if
you're excited about it, but you want to start to think, "How do I start to go
deeper?"
Dr. West how do you talk to people about starting to take it from understanding
the basis of the idea to using it?
Robert West: Yeah, so there's a number of different approaches depending on
who the audience is. And as I alluded to earlier, is at one level, it's a very
common sense model, which is something that when we're talking to
policymakers who are not behavioral scientists, they've got no particular
expertise not even necessarily a great deal of interest in behavior or behavioral
science.
There is a level at which you can interact with people in that position, which
will help them a little bit to gain a better insight into the problem that they're
trying to solve, which may be a behavioral problem, whether it's to do with
getting people to self-isolate during COVID, if they've got [00:30:00] symptoms
or whatever, or stopping smoking or alcohol or whatever it might be.
Or it... literally anything. We get peop- requests from all government
departments and lots of different industries. It's a general model. So there's,
there is a level at which you can with as you can probably tell with the
discussion we've been having, that people can go, " yeah, okay, I get that." And
then what they, what it can lead to is them questioning their assumptions, like
the assumption it's all about motivation. So actually, is it all about motivation?
Maybe what we need to do actually is to provide a more facilitatory
environment for them. Maybe that would be the thing.
So even at that level, there's an opportunity there to work with policymakers and
clients to s- to, to raise the game a little bit. But then what's most interest to me
is the working with people with a bit more expertise who [00:31:00] actually are

developing behavior change strategies either as policies or that it might be tools
like an app or something like that, or a service that you deliver to people on a
health insurance program or whatever.
Now when, with that, then what we've done is to develop a, an online tool
which people can use called the CACTUS tool. And people can find it on the
COM-B website. Basically, it's C-O-M hyphen B org the COM-B website, and
it's there. It's a free tool for people to use. And what I would say about the tool
is that y- you, you can just use the tool and go, "Okay, I think I know enough
about it, and I'm gonna go through this process where it says what's the behavior
you're interested in?
What's the population or the group? What's the what's the setting that the
intervention is in?" And then you go through your various COM-B targets. Do
we need to raise awareness [00:32:00] of the behavior? Do we need to improve
how people, know how to do the behavior, et cetera, et cetera. All the
capability, opportunity, and motivation things.
It will then prompt you for what we call behavior change techniques. It'll
identify behavior change techniques that will be most relevant and so on. And at
the end of it, it comes up with a a sort of an output which you can use if you
wish to and we do this 'cause it's quite interesting, as a prompt for one of the
large language models.
I use Claude myself. That is a structured prompt which then helps the language
model to go, "Oh, okay, I get what you're trying to do." And it'll usually do a
pretty good job. It's not the end of the ro- the end of the process, but it gives you
a starting point for the strategy. So the CACTUS tool is something-- is a tool
that we've been developing.
It's still under development. We're gonna be developing a website version. And
I think that it, it probably is usable by people with a [00:33:00] certain level of
expertise without support. But what we often do is we do a bit of hand-holding
with people to go, "Okay, let's go through this.
Let's take each of these targets. Let's see what the evidence is to support it so
that we can arrive at a better kind of a conclusion." So that's just one way of
approaching it as well. And then at the most technical level what we do is we
use it to build what we call systems models, which we then use as the basis for
in, in one of the projects I'm involved in, what we call an agent-based model,
which is a a sort of mathematical model for mo- modeling population behavior,
and parameterizing that with data to make fine-grained predictions of what

would happen if we for example, in the UK, raised the price of e-cigarettes by X
percent.
What is our best prediction about what impact it would have?
MichaelAaron Flicker: I think that really opens the door to talking about
[00:34:00] real-world examples where the government has used COM-B to
encourage positive behavior change. I think we're just starting to get into that,
so maybe you could talk a little bit more about, in your personal experience
some of the great outcomes by using COM-B in this way.
Robert West: I-- yeah, so obviously health behavior is a key area. It's one that I
mostly worked in and COM-B has been was, has been widely used in the
government's comms strategy around it. It was, it's also be- it was also used...
I'm gonna say it was used, it was used in the sense it was proposed for our
government's response to COVID.
But as we all know, the UK unfortunately, rather like the United States, didn't
cover itself in glory with its political response to COVID and we ended up with
a lot more deaths than we need have done. [00:35:00] But at least the scientific
advisory committee to the government was saying These are these are things
you should do."
And an example of where we failed actually, as a where I think we could have
done a lot better with a, in hindsight, pretty straight simple concept from
COVID is the government made the ki- made the assumption that when people
had symptoms or had been exposed to someone who's infected with COVID,
the reason why they weren't self-isolating was motivation.
It was because they were bad people, and they needed to be punished and and
then all told that they should do this. And all the evidence that we were getting
with regular surveys at the time was actually it wasn't that motivation wasn't a
discriminating factor between people who were and were not isolating.
It was opportunity. It was particularly around the issue of being able to
[00:36:00] actually make ends meet, because the... If you self-isolated, you were
gonna take a massive financial hit, and it was gonna be really problematic. And
so if you need to feed your family, et cetera. It's really obvious and basic when
you think about it.
The government were told about this. They didn't actually for whatever reason I
have my own views about what the reasons were, but for whatever reason, they

didn't take that on board, and we ended up with a lot more infections than we
need have done and a lot more cost to the economy than would've been the case
if they had made an appropriate response in terms of facilitating the opportunity
to engage in this desired behavior.
So those are, that's th-those are a couple of examples. What I'm observing now
working with companies that provide services, behavior change services to local
populations, is that typically now when governments put out briefs to those
[00:37:00] services to say What are you gonna deliver and how much is it gonna
cost and what results are you gonna have?"
They usually require them to do it in terms of COM-B. So that must be a bit
annoying for people if they don't like the model, but it does seem to be
something that's quite embedded.
MichaelAaron Flicker: What about in businesses? Have you seen an example
where a brand has used COM-B to make its marketing more effective?
Robert West: I believe I have, but I don't think I'm in liberty to comment. Yes I
think there are fairly large brands who have and are using the model not least
because people working in those companies are from the sort of school at UCL
of co- the school of COM-B as it were.
So they've brought it with them. But I [00:38:00] think i-i- I think I-- My
understanding is I don't know the world of business terribly well only from what
I the people I actually deal with. But I think that it is a general model of
behavior. And I really hope My interest in this is not to make money, it's not to
it's not to make companies rich or whatever. It's to improve h-humankind. And
to the extent that it, it achieves that, then that's good. But I like any tool, it can
be used well or badly I think one of the unlocks for me regardless of the people
listening and where- what they might do day in and day out, all of us have a
passion for using communication or using our station to improve human
behavior and to improve [00:39:00] all of our lots in life. And y- one thing that I
had not considered in the way that you presented it was that it gives us an ability
to talk to a more generalized audience.
MichaelAaron Flicker: You talked about policyholders. I'm thinking about
stock shareholders, stockholders- executive boards that want to understand the
issue to a level of specificity, but only to a certain level. And so the model gives
us the ability to go deeper- where we need to, but also stay more general to open
people's minds to the opportunity that it may not just be the most commonly
assumed problem that's stopping the behavior change.

Robert West: Yeah, absolutely. And I think what I hope, what I like to think is
that by providing this sort of bird's-eye view, it does, it, A, it covers all the
bases, so it makes sure that you're not missing out on something because a lot of
the psychological theories are really quite narrow, and they focus on some
aspect of motivation [00:40:00] very often, or sometimes capability, sometimes
opportunity.
But it goes, hang on a minute let's look at this from a bird's-eye view and see
where it l- where the land lies. But then if you're interested, if you have the
motivation, capability, and opportunity to dig deeper, there it is. There's this
other level of granularity underneath it, and you can keep going down like a
fractal, down into the weeds of it at, to any level.
And if you're let's by some chance working in a company as a a, as a behavioral
advisor or in some behavioral capacity, and you're familiar with theories like
operant learning theory or the theory of planned behavior or the health belief
model or self-determination theory or the, any of the sort of about 100 relevant
theories or subjective expected utility theory or prospect theory, Danny
Kahneman's one if you if these are your g- your gig, as it were, [00:41:00] you
can see how those link up with the bigger picture and how if that's where you
think you need to put your effort, you can see how you can nest that within that
broader framework and not get a bit, not get too blinkered in thinking that, oh,
what we've got to do is positive framing or negative framing or any of these
other kinds of concepts.
Richard Shotton: Up till now we've very much focused on Com-B, but you've
also created the the behavior change wheel. H-how do those two things
interlink? What's the W- where do they mesh together?
Robert West: COM-B was specifically developed as the hub of the behavior
change- wheel. The paper in implementation science that kicked this all off
which was published in 2011 was a paper about the behavior change wheel, and
Susan and I developed the wheel with COM-B at its heart [00:42:00] as a piece.
And the i- the idea is that COM-B is a model of behavior. The behavior change
wheel is a sort of l- large scale model of a framework, not a model, a framework
for behavior change. And so in, in the "Behavior Change Wheel" book and all
the implementations of COM-B it's it's really about finding a process in which
you say you do what we call a behavioral diagnosis.
What needs to change within capability, opportunity, and motivation to get the
behavior in the direction that you're wanting to go? Now, you can then map that

onto, in the behavior change wheel the ri- the sort of ring that goes around that,
which is what we call intervention functions, of which there are nine broad
ones.
You can educate, you can persuade, you can incentivize, coerce train you can
regulate or restrict, you [00:43:00] can model you... and you can envir- do
environmental restructuring. And so there's a mapping of the COM-B targets
onto these broad intervention functions. And then having done that, you can
say, "Okay, how do these link up with much more granular and specific
behavior change techniques to deliver those?"
And then the outer ring of the wheel is what we call implementation options,
which is, okay How are we then gonna make that real? And the options are
things like, if you're a government legislation, regulation, fiscal policies,
providing a service and so on. And communication obviously. So it-- what you
can-- what we do with the Behavior Change Wheel is to give people a kind of
path from the problem specification, what is it you're trying to achieve, all the
way through to a, the finished product, as it were, in terms of a [00:44:00]
behavior change intervention.
And the key thing really that we wanted to achieve with that and we didn't just
make it up obviously. We, we-- although if it had been me, it probably would
have done it. But Susan insists on- She kept you on the straight and narrow ...
Looking things up- Yeah ... and checking things out.
But we did a review of all the frameworks that were out there at the moment at
that time, and we wanted to put them all together to get the best of them. But
essentially, what we, what you end up with is a way of l- of connecting your
intervention development with the science of behavior change- and the practice
and the evidence. It could be that-- Let's say you're developing a comms
campaign to achieve some goal. Now it may well be that there's a lot of
evidence in the scientific literature to help you, but actually it's quite likely
you're gonna have to do some audience research to find out specifically with
this problem, [00:45:00] what what the audience is all about.
And so the COM-B can provide you with a way of structuring your audience
research so that you make sure that you identify what the real targets are. You're
not just firing a shotgun at at the target.
Richard Shotton: So would it be fair and maybe too simplistic to say COM-B
is very much about the diagnosis of the problem.

It tells you what needs to change, and then when you move to the Behavior
Change Wheel, it's what are those tactics and solutions and interventions.
Would that be fair?
Robert West: That's exactly right. That wouldn't be simplistic at all. Phew.
Richard Shotton: I try and avoid too many simplis- simplistic parts of it in a,
per podcast.
And then-- So that's a brilliant, I think setup of the difference between COM-B
and Behavior Change Wheel. The other thing is, you mentioned it was back in
2011 when COM-B was created. You've now [00:46:00] created COM-B 2.0.
Why did you feel the need for a kind of refresh and what are the major
differences?
Robert West: So there's a bit of pushback about naming it COM-B 2.0. Ah,
okay. Good to know. Yeah. Good to know.
Richard Shotton: We can edit that out. No, not at all. It's got the, it's got the,
it's got the official brand now. Because it's work in progress. Okay. So
essentially I I, obviously, as we get more and more experience with COM-B and
new information becomes available, and Susan and I and others have been
doing other research into things like what we call ontologies, which are a sort of
scientific foundation for conceptualizing behavior and so on.
Robert West: Then a lot has happened since 20 2011. And and what I wanted
to do was to pull that together into an updated COM-B. Whether it's called
COM-B 2.0- Okay. Yeah ... Depends on my colleagues as much as [00:47:00]
me. But it will be an it, it will be an updated version.
And there's a couple of key things that really came out or several key things.
One is that the model that I've talked to you about is really the way we've been
thinking about it is like an atomic version of behavior. It's behavior of a person
in the moment, here and now, right?
At this poi- point in time. At this point in time, do they have the opportunity,
capability and motivation to engage in this behavior? And that's the starting
point. But for a lot of the behavior change tasks that we're faced, that's not what
we're interested in. We're interested in populations. We're interested in groups,
in companies, in teams, in aggregates of people.

So what we've done is to separate out the let's call it the atomic COM-B in the
moment, the individual at this time, [00:48:00] versus the aggregate COM-B,
which is a statistical model. It's a model of what is the likelihood in a population
of this behavior, given the distribution of these capability, opportunity, and
motivation factors.
And because you're abstracting away from a sort of very mechanistic, causal
kind of model in the moment, a process model, into a, what we call a
dispositional model, a sort of more an epidemiological type model, then that
abstraction that takes you down a particular modeling route. So when we look at
capability and in a population, we're looking...
Some people will have high capability, some people have low capability, and
so on. So it'll be a... So we're modeling a distribution, and when we're
intervening to change capability, we're interve- we've got to make decisions
about how we want to influence that distribution. Do we [00:49:00] want to
shove the tail of the distribution up a bit?
Do we want to shift the mean, and so on? So it moves you from a sort of
process s-somewhat mechanistic type view to a more epidemiological view of
the behavior. And that then I think provides your, the users of the model with
something they can really get their teeth into because these are parameters that
you can actually measure and you can specify in a population.
So in our smoking data, we collect data every month on the population of the
UK, and we're looking at things like quit attempts and quit success. And so we
can look at the frequency, the the prevalence, the incidence. We can look at the
average level of motivation to quit and things like that.
And so you... So with the aggregate model, you've, you're connecting with that
very what for behavioral scientists is a very [00:50:00] natural measurement
process. So that's one thing, atomic versus aggregate- Yeah ... COM-B. The
other probably critical thing is that we-- whereas when we wrote the original
model, we gave examples of the kinds of things that we meant by capability and
opportunity.
But over the years, this has crystallized into a a set of I was gonna say 26, but
yesterday it went up to 27. And it, because it's a movable feast this is a model.
It's not reality. It is a model of behavior rather that, that's used for a particular
purpose. So how you chunk up your various components is dependent very
much on your use case.

So we, what we've done is we've crystallized on X number of capability,
[00:51:00] opportunity, and motivational factors, which provides the basis for a
more granular COM-B assessment or COM-B diagnosis. And we've linked that
in to a specific set of intervention functions and behavior change techniques.
So it's So the updated COM-B is much, has a much more granular layer to it
than the original one did. And there's one other thing that I'll mention that may
or may not make it to the final broadcast , and that is the link with ontologies.
Ontology is a word that most of your listeners probably have never heard of, or
if they're philosophers, they'll know that it's the sort of science of knowledge
and understanding and how we see the world.
But ontologies are computer science artifacts that basically underpin everything
that you do when you go on the web. And they are ways of classifying the
world and relationships between things in a [00:52:00] very scientific, specific
way so that you know what you're talking about. So when we interact with large
language models, because we're using language and language is very flexible
and often ambiguous and so on, the language models are dealing with language,
and so they make mistakes, and they as we know and a lot of the sort of really
clever science that we'd like to do with behavioral, behavior change you can't
really do with just putting in prompts using natural language into these these
models. So if y- But if you can create ontologies with very distinct classes, very
precisely defined, with unique identifiers, then not only does the human know
exactly what you're talking about, but so does the computer.
And the computer goes, "Thank you. Finally," now I can give you an answer
that's gonna be more [00:53:00] s- useful to you." So what we are doing with
Combi and with the rest of the work we're doing, is pulling it together with our
computer science and AI friends into a sort of a more, shall we say, a more
rigorous scientific model of behavior that I think we don't know yet, but I th-
The hope is that it will enable us to interact much more productively with the AI
tools that we have available.
MichaelAaron Flicker: It's
Robert West: wildly exciting.
MichaelAaron Flicker: Wildly exciting to say the least. And I think it just
gives everyone listening a chance to understand not just how to apply
behavioral science, but to think more deeply about the academics that can
power getting to new ideas and bigger ideas, because I love what you said, Dr.

West, you can go deeper. You... A- However deep you wanna go, there's a-
another layer that you can continue to explore. And in all the [00:54:00] time
that Richard and I have been working together on this, that's what you find
about people who are passionate in this field. They like to go deeper. They
wanna talk more and learn more.
So I think it's a great prompt for us all.
Robert West: Yeah and I think that the thing to bear in mind is it's always work
in progress. Science is always work in progress. And so people get frustrated a
little bit when with me sometimes when I sort of mess around with models and
concepts and and they s- something that they thought I was saying a week ago is
no longer true.
Yeah, I'm sorry, but that's how it goes, yeah. And I try to, I try not to be too fl-
erratic about it. But but it... Really important thing to understand about science,
and it's so true of behavioral science, is that is that w- this is a perception of the
world.
It's our own way of representing an incredibly complex world. We... The reality
is what it is, and our perception of it is [00:55:00] what that is. And that can,
that, that can-- You can come at something from a lot of different perspectives,
and it's not that one perspective is necessarily right or wrong, it's just it is more
or less useful for whatever purposes you're currently putting it to.
Richard Shotton: Yeah. I love the idea of knowledge being in a state of flux,
and it brings to mind that amazing... I suppose it, half of these quotes are
apocryphal, but that quote from John Maynard Keynes when he was criticized
for coming up with a different point of view than a week earlier, and he turned
around and said, " when the facts change, I change my mind.
What do you do, sir?" and I've always thought that's a great comeback to anyone
that criticizes a lack of consistency.
MichaelAaron Flicker: Yeah, that's true. Yeah. Dr. West, we wanted to give
you a chance to leave our listeners with a takeaway, either of the COM-B model
of your work or of people that want to use behavioral science for good.
What would you like to leave [00:56:00] folks with as we're coming to an end
of our conversation?

Robert West: I think the thing that, the overarching thing that we can all
always do better is to be reflective about our assumptions. And one of the things
the COM-B model does is it helps you to challenge your assumptions about
what's driving a behavior or what would be needed to change a behavior.
And whatever we think is fine, but we should all be good and I'm gonna use the
word Bayesian here, w- in terms of Bayesian statistics, we should all be good
Bayesians, which as you were saying, Richard, just a minute ago, around when
the evidence changes, I change my mind. Yeah. We should never...
With Bayes' theorem people can look it up if they're not familiar with it, when
you get to a probability of one, there's no chance of changing anything It's not
possible to change or zero. So we should always be somewhere between one
and the [00:57:00] o- one and the other. And we should be ascribing an
appropriate level of confidence to our beliefs, which should change as the
evidence comes in.
And that's fine. That's how it should be.
Richard Shotton: I think it's very easy to believe let's say someone had been a
smoker and they have changed their behavior. It's very easy to think, " I-- it was
all about motivation for me," or, "It was all about psychological capability for
me, therefore, that is the thing I need to change on everyone else."
But that, I'll conf- find false consensus effect is a very dangerous myth, and
often other people require a very different solution. So make sure that you work
out whether your own situation is representative or it's a kind of aberrant belief.
Robert West: Absolutely. And pay attention to the data.
And data beats experience. I'm sorry in terms of, i-in terms [00:58:00] of what
people usually do, experience beats data, especially my mum who says, " I see
more people smoking now, therefore smoking prevalence must be going up."
And that's not what the data says. Data beats experience.
I'm sorry, but it does.
MichaelAaron Flicker: For those that wanna learn more about your work, Dr.
West, that may want to hear more of what you're talking about, learn more, can
you tell us about where they can find more about you and maybe any projects
that you have coming up that we can all help stay up to speed on what you're
doing next?

Robert West: The COM-B model, which we're working on a lot, and we've got
this, a new version coming up, is a-accessible via com-b.com-- web-- .co.org
website C-O-M hyphen B.org. There i- the in the AI work we're doing which
also links with [00:59:00] that is also on our website.
This is actually work that is fun- has been funded over many years and is now, I
think, coming to a pretty good state in terms of being usable. And that is
humanbehaviorchange.org. It's the Human Behavior Change Project. And then
there's books that I write with my son Jamie, who's a writer and performer.
And there's one called "Energize," one called "React," and To be coming out at
a bookshop near you at some point later this year will be one called Reflect: The
Science of Real-World Decision-Making. And this is actually a labor of love s-
dating back to about 1979, where I think I finally got to a point where my views
on it are sufficiently settled.
And the idea behind that, and it is just very briefly, is that there are some
wonderful models of decision-making, as we all [01:00:00] know. But what's
really struck me working in the field for so long is that they're all quite
compartmentalized and mostly relate to decision-making in laboratory situations
that don't connect that well with the real world.
And when you put them into the real world, often you get very different kinds
of results. And so we wanted to really look at the literature and experience of
people making decisions in real world settings and the enormous variety of
ways in which people do that and how they go wrong. I've become quite an avid
follower of various YouTube channels like Mentor Pilot, which is, for those of
you who are interested in that kind of thing, it's a fantastic channel.
Very instructive about real world decision-making and where it goes wrong.
And so I've tried-- we've tried to put all that understanding into a framework in
this new book, which will be coming out later in the year.
MichaelAaron Flicker: Thank you for sharing that. For all of our listeners
today, if you [01:01:00] found the conversation we had with Dr.
West engaging, please share it with others and comment, and follow our pages
so we can reach folks more just like you. Until next time, I'm MichaelAaron
Flicker.
Richard Shotton: And I'm Richard Shotton.

MichaelAaron Flicker: Thank you, Dr. Robert West, for joining us today. It
was a lovely conversation.
Robert West: Thank you. It's really my pleasure.
Advertisement: Behavioral Science for Brands is brought to you by Method1,
recognized as one of the fastest-growing companies in America for the third
year in a row, featured on Inc.'s 5000 list. Method1 is a proudly independent
creative and media agency grounded in behavioral science. They exist to make
brands irresistible, helping people discover products, services, and experiences
that bring moments of joy to their lives.
As behavior change experts, Method1 creates emotional connections that drive
true brand value for their [01:02:00] clients, focusing primarily with indulgence
brands in the CPG space. Find out more at method1.com.