What is Behavioral Science?

A Primer for Anyone in the Business of Behavior Change or Support

Jeff Brodscholl, Ph.D
Greymatter Behavioral Sciences

As demand for behavioral science continues apace, it can be easy to lose sight of the assumptions people may make about behavioral science and what it can do for them. Part of the problem lies with the phrase “behavioral science” itself: If you rearrange the words and add filler, you get the seemingly obvious tautology that “behavioral science is the science of behavior”, which is not exactly a statement that screams for elaboration.

Yet, there’s quite a bit within that phrase that people outside of behavioral science might not know about that can do with further unpacking. For instance:

What are all of the phenomena that are tied to the term “behavior”, which behavioral science takes as its subject-matter?
In what way is behavioral science a “science”?
If there’s an “applied” version of behavioral science, in what way is it applied – and what is its value to me?

In this post, I’m going to take a step back and provide a basic 101 overview of what behavioral science is and what it can mean for anyone who has a stake in human behavior, whether it’s because they work in communications, policy, or the design of almost anything – a service, product, experience, or intervention – that has shaping, supporting, or accommodating people’s behavior as a key factor for success. What I discuss in this post will also be relevant to anyone whose job it is to obtain insight into people’s thoughts, feelings, actions, and other behavioral drivers as a critical step in these efforts.

I’ll then rely on the “rule of threes” to tell you:

Three things that anyone who’s interested in applying behavioral science to their work in these areas needs to know about behavioral science in terms of its scope, methods, and what makes it “scientific”
Three things that behavioral science delivers that make it valuable to turn to when developing solutions that can effectively shape, support, or accommodate people’s behavior as intended
Three watch-outs, in the form of things that behavioral science is not, which, if you know about them, can empower you to avoid the pitfalls and be a "super user" of the science

The First Three Threes: What is Behavioral Science?

If you ask people to tell you the first thing that comes to mind when they think about “behavioral science”, chances are they’ll say that behavioral science is about biases and habits, or that it has something to do with the way in which our thoughts, feelings, and actions are determined by what goes on in our brains. These answers essentially reflect a commonsense version of the first thing that behavioral science is, which is that –

#1: Behavioral science is a body of knowledge about behavior and its determinants

You can think of this as the pile of collective knowledge about behavior that’s captured in decades’ worth of peer-reviewed journal articles lining the bookshelves of academic libraries and behavioral scientists’ offices. Yes, it includes what we know about people’s biases and habits and the role the brain plays in behavior – but it also more broadly encompasses:

How we think, which includes how we interpret information, make inferences, reason and problem-solve, predict the future, and encode and recall memories for facts and events
How we learn, whether it’s through the consequences we experience with our behavior or through repeated practice
What motivates us, and how we organize our behavior in the sustained pursuit of desired objectives
What our emotions boil down to, including the types of emotions that exist, their causes, and their consequences
How we form preferences and make decisions

It also encompasses the social aspects of being human, starting with our immediate kinships and radiating outward to our interactions with others through work, community, and the broader society in which we live. Included here are the ways in which our sense of who we are is, itself, shaped by others and made possible by looking at ourselves as an object of our own attention. That means that the knowledge also touches on:

Our ability to empathize, take perspective, read other people’s minds, communicate, and use language
How social influence works, including how beliefs, attitudes, and practices are transmitted through modeling, norms, and direct efforts to persuade
What our mental representations of ourselves are about, and what role they play in our thinking, feeling, and motivation
How our beliefs, attitudes, thinking styles, and emotions are shaped by culture

Lastly, all of this knowledge is multi-dimensional, meaning, it typically involves looking at any one aspect of behavior from more than just a single angle. Sometimes, this is about connecting behavior to the way in which the brain and other key biological mechanisms operate. At other times, it's about relating behavior back to specific mental systems or processes that the brain supports, or to the structure and function of culture, institutions, and the types of social worlds we inhabit. And, often, it’s about working across these perspectives to achieve a full understanding of what’s behind people's actions and experiences.

Sound like a lot? That’s because it is! That’s a virtue, though, as it means that there’s a lot of material that can be plumbed for practical problem-solving – no small thing when you consider the ways in which any one behavior can be driven by many different determinants, or the drivers can vary by the type of behavior in question or the context in which it occurs. It’s also why behavioral science is multidisciplinary, encompassing not only neurocognition and behavioral economics, but also social and personality psychology, sociology, and anthropology, often with a certain amount of crosstalk occurring between these fields.

But what makes this body of knowledge special isn’t merely its comprehensiveness. It’s its provenance – the way it has been developed over time. Put simply, it’s knowledge that, because of the way it has been acquired, inches closer to being “bankable” than what you might get if you relied on everyday wisdom alone. It’s for this reason that the knowledge can not only lay claim to our attention, but also at times yield unexpected insights that add something new to our understanding of what makes people tick. That brings us to the second thing that behavioral science is, which is that –

#2: Behavioral science is a unique collection of methods for learning about behavior

Ultimately, behavioral science looks to build a body of knowledge that does two things:

Describe people’s thoughts, feelings, actions, and social worlds in a way that’s true to what the world of behavior is actually like
Explain those same thoughts, feelings, and actions in terms of specific causes that are, in fact, their real-world determinants

Not all methods for developing this knowledge are able to achieve these aims. Having a patient free associate about their dreams and then interpreting the metaphorical meaning of those free associations does not work to uncover the causes of schizophrenia or shed proper light on the processing of traumatic memories. It does, however, support dubious theories about psychosis being due to “refrigerator mothers” or depression being due to repressed childhood memories, which, in turn, lead to treatments with either no or even deleterious effects.

What’s needed, then, are methods that follow certain rules – rules that, when satisfied, allow us to make credible inferences from observing and interacting with people, which can then be laddered up to testable theories about people and their behavioral drivers, or to generalizable claims about what impacts what. Among the rules behavioral scientists follow are that they:

Use tools for measurement that demonstrably measure the things they purport to measure, and give the same answer when applied to the same thing repeatedly
Sample people for research in such a way that what’s observed within a sample can be generalized to the broader universe of people from which the sample was drawn
Make observations under conditions that allow us to say that, when behavior X is observed to co-occur with condition Y, it’s because Y causes X
Collect data within a context that matches real-world conditions so that what’s observed reflects what people do, think, and feel in real life
Privilege simpler explanations over unnecessarily complex ones, and formulate hypotheses that are capable of falsification

Many of these aren’t just the rules for behavioral science; they’re also the rules for the natural sciences. As such, behavioral scientists often use methods that are similar to ones found in other scientific fields, albeit modified and supplemented with procedures that are suited to the unique aspects of behavioral science’s subject-matter. Among the methods commonly used in behavioral science are ones that entail:

Collection of data from carefully-curated samples that are subject to statistical analysis and modeling
Experiments involving random of assignment of people to conditions differing in terms of exposure to certain stimuli or interventions
Self reports captured under conditions that take into account the scope and limits of introspection
Formal observations of the way behavior covaries with observed changes in the natural environment, or with variations in carefully-crafted test stimuli, which can then be used to infer the workings of the brain or mind

These methods cover a wide terrain, but they're ultimately bound together by a common commitment to rigor that's the same we expect of the sciences generally. And that commitment exists because –

#3: Behavioral science is a discipline that holds itself highly accountable to reality

Behavioral science can certainly make for a good story about why people do the things they do – but, for behavioral scientists themselves, having a good narrative alone isn’t enough. That’s because the story behavioral science tries to tell isn’t just any story, but one that strives to reflect human truths that are out in the world, operating independently of what we might want to believe about behavior and its drivers.

What that means is that, in behavioral science, the work of understanding behavior needs to go wherever data and evidence take it, even if that means landing on a narrative that, being perfectly coherent and actionable, nevertheless confounds expectations because it challenges conventional ways of thinking about people. Again, that’s a good thing: If the science were merely confirming what we already believed about people and did nothing to open us up to new insights, then why would anyone bother to spend the time and money doing behavioral science work?

Of course, this feature of behavioral science can also sometimes cause friction, as it calls for a tolerance for processes, ways of reasoning, and forms of open-endedness that can seem alien to everyday ways of doing business or making policy. But there are ways to harmonize these practices with the realities of real-world problem-solving. And the upshot is that, by making use of the science for practical purposes, one can arrive at insights and take actions that can be given their due precisely because they are developed with an eye toward a closer alignment with the underlying reality of what makes people tick.

The Second Three Threes: What is the Value of Applying Behavioral Science?

So, there are good reasons to believe that behavioral science can deliver value above and beyond what we often get from the tried-and-true ways of understanding people and taking actions related to their behavior – but what exactly does this mean in practice? Here, I’ll focus on three specific benefits that can come from applying behavioral science for organizations and initiatives that have shaping, supporting, and accommodating behavior as a key objective.

#1: Behavioral science can put your efforts on a more successful footing from the outset

There’s essentially no activity directed at human behavior that does not have behavioral insight as an early step in its development. Whether you’re developing a strategy to increase adoption of a treatment innovation, building a product that helps people develop environmentally-friendly habits, or devising a policy to encourage greater use of public transportation, it’s inevitable that you’ll need to brainstorm and conduct research to get a picture of people’s needs, beliefs, and other behavioral drivers in order to inform the approach to shaping and supporting their behavior in desired directions.

These insight generation efforts are not neutral, though. Early assumptions about the people who are to be the targets of a specific solution exert a key influence on the insight generation process: They shape the types of questions that are asked, the data that are collected, the methods used for data collection, and the way in which the data are interpreted. These influences have downstream consequences for the picture that’s developed of people and their behavioral drivers, which then often gets crystalized in personas and other collateral that are used to guide further work. And that has consequences for the strategies that are pursued and the ideas that are put on the table as solutions are developed.

It's here, though, that behavioral science can begin to add substantial value – specifically, by contributing new, evidence-based ideas about behavior at a key early stage in the solutions development process. Just as behavioral scientists have special methods for generating basic knowledge about behavior, so, too, do they have specialized tools for accessing and applying that knowledge to understanding the specific behavior of a target group of interest. These tools allow behavioral scientists to use their background knowledge as a filter, or “lens”, for analyzing what’s already known about the target group and their behavior, and to dig into the scientific literature for learnings that can fill knowledge gaps. These efforts can bring to light behavioral drivers that make sense to consider from a behavioral science perspective but that might get overlooked using more traditional approaches. Behavioral scientists can then leverage their toolkit of research methods to subject these initial ideas to further testing. They can also use methodologies such as think-aloud protocols, psychometric assessments, behavioral experiments, and real-world observational studies to surface additional behavioral drivers that would be difficult to get at with any credibility using standard survey and interview methods.

These applications don’t necessarily replace ideas about behavior that other people involved in the insight generation process might bring to the table. What they do, though, is supplement them with ideas that have a unique grounding owing to their origin in basic behavioral science knowledge and their verification through rigorous research methods. By being so grounded, they not only introduce new behavioral drivers to consider, but provide an incrementally better foundation for crafting solutions that are likely to hit their mark once they are deployed. And that has the effect of inching the entire solutions development process one step closer to success.

#2: Behavioral science supports orderly processes and methods for successful solutions development

As noted earlier, behavioral scientists often rely on experiments as a way of developing basic knowledge about what's behind people’s thoughts, feelings, and actions. Those experiments work by deliberately manipulating some feature of the world to which people are exposed and then observing the effect on behavior. Which experience is manipulated and how, as well as which behavioral outcomes are studied and in whom, is not arbitrarily determined. Rather, it’s the end-product of a process that involves:

Identifying the behavior to be understood
Reaching for a theory about the behavior, or conducting initial research to develop such a theory
Using the theory to generate hypotheses about how the behavior might vary with changes to some aspect of the world (i.e., one that can be experimentally manipulated)
Using the same theory, along with pretesting and practical considerations, to design the experimental manipulation

If that sounds a little like a process that could also be used for developing a real-world solution aimed at behavior (where the experimental manipulation = a feature of the solution to be developed), then you’d be correct. Indeed, behavioral scientists are often called upon to use their skills not just to develop basic scientific knowledge, but to develop and test evidence-based interventions that are meant to address real-world behavior change challenges in public health, environmental sustainability, and other domains. They’re also often called upon to work alongside product designers to assist in the development of products and services that have the right features to make them engaging and easy to use. And while their efforts may sometimes be as simple as a behavioral science team running online experiments to test the effects of messaging manipulations, nudges, or other maneuvers, they can also include all of the operational complexity one would expect of efforts that require multidisciplinary collaboration and multiple stakeholder buy-in to make those efforts work.

What behavioral science can offer in all of these cases is a style of reasoning and a set of methods that results in:

A consistently tight focus on the outcome that matters most: The behavior you’re trying to shape or support
A logical, orderly use of behavioral insights to inform ideas about solutions design (where, at times, the connection between insights and design ideas can otherwise be haphazard)
Reinforcement of the full set of steps, from insight to testing, that are supposed to characterize the solutions development process, but that can sometimes receive short shrift
A reliance on methodologies that keep the entire process evidence-based while encouraging a mindset of test-and-learn through behavioral experimentation

These applications don’t have to be rigid. They can give rise to activities that are executed at any level of complexity depending on the type of solution to be developed, the population to be exposed to it, practical considerations, and the degree of evidence that decision-makers will need to feel confident with the outcome. As an added bonus: Basic behavioral science knowledge can continue to inform what goes on throughout the process, not just by contributing to insight generation, but also by leveraging what the literature on tested interventions, behavioral driver manipulations, and the effects of context might say about features that would be worth including in messaging, product design, or whatever form your solution needs to take. The net result: Solutions that have a level of credibility and a greater likelihood of achieving success in impacting people's behavior than would be attainable otherwise.

#3: Behavioral science is a good match for the practical realities for success

Finally, there’s the way in which the orientation of behavioral science – not just its body of knowledge and methods, but the mindset it adopts in its efforts to understand and impact behavior – aligns extremely well with a key practical reality, which is that, if you’re going to develop a service, campaign, policy, or other intervention to shape, support, or accommodate people’s behavior, your success will ultimately be tied to how well your solution does in bringing about the behavioral outcome you seek. If your weight loss app doesn’t result in its users losing weight, those users will abandon your app and rate you poorly. If your social marketing effort to get people to abandon political conspiracy theories doesn’t reduce the percentage of the electorate that indulges them, then your efforts will likely fail to receive further funding. And if you’re a mission-based organization, the failures won’t just translate into financial costs, but will also be an affront to the mission the organization values.

In other words, success or failure in every one of these cases hinges on a very specific empirical truth – one that has to do with whether a sought-after behavioral outcome has actually occurred as a consequence of exposure to the solutions you've developed. And that means that the most successful organizations are going to be the ones that value not only keeping their eyes on this prize, but also doing so in rigorous, evidence-based ways.

Behavioral science embodies the values and practices that an organization can adopt to orient itself the right way in these efforts. It not only keeps its focus on behavior, but sees its success as following from the way it makes itself answerable to what truly drives behavior out in the world. More importantly, it takes these statements as matters of principle that inform the science's activities from top to bottom, creating the conditions that make evidence-based learning about behavior simply the "natural way of doing business". As such, an organization that adopts behavioral science has an opportunity to benefit beyond what it derives from the science’s knowledge and methods alone to what it gets from embracing the commitments to evidence-based reasoning that the science makes in its quest to understand behavior and develop effective ways to intervene with it. When the organization is already tuned this way, behavioral science becomes a natural fit for amplifying practices that already help it to be successful. When the organization isn’t tuned this way but is open to embracing behavioral science at this much deeper level, the result can be transformative.

The Third Three Threes: What Are the Watchouts?

We've discussed how behavioral science can add substantial value to any effort in which taking action with respect to people's behavior is a key component to success – yet, there are a number of sticky issues that can stand in the way of being able to unlock this value or even result in misguided applications. We’ll close out by discussing some of these complexities, specifically by making clear three things that behavioral science is not. And we’ll use that opportunity to tell you a few things you'll want to know to ensure that your use of behavioral science is set up for success.

#1: Behavioral science is not a panacea

Earlier, we said that what behavioral science can do for the development of a solution focused on behavior is move the development process closer to success. “Closer” is the key word here. While behavioral science can be a powerful tool to have in the toolkit, it is not a form of magic that can act as a guarantor of success:

The methods of behavioral science are not engines of insight due solely to their clever design
The hypotheses derived from applying behavioral science knowledge to a specific problem are just that – hypotheses that must be tested before they can be treated as something closer to gospel
Design processes that leverage the practices and methods of behavioral science, including rigorous experimentation, can still fail to yield an acceptable outcome even when followed to the letter
The use of behavioral science will fail if not supported by an organizational culture that understands the ways of thinking and doing that are at the heart of behavioral science practice

That doesn’t mean that the use of behavioral science must somehow amount to little more than an expensive shot in the dark. Behavioral science contains the very knowledge that can be recruited to minimize the risks involved in its applications. There are also processes that can be followed to ensure that behavioral science is engaged under the right circumstances. Applied behavioral scientists know about these processes, and they know how to take the steps to ensure that the science is used in ways that are appropriate and will yield maximum value. What the above issues make clear, though, is that any application of the science needs to be carried out with a certain level of open-mindedness and modesty, combined with careful, well-informed calculations around how the knowledge and methods of the science should be used in any one case. And those attitudes and practices becomes even more important to adopt when you consider that –

#2: Behavioral science is not remotely simple

The most popular ideas from behavioral science tend to be ones that make human behavior seem easy to understand – yet, there are no “three facts” about behavior that unlock all of the secrets to why people do what they do. At the end of the day, behavioral drivers interact with one another and with context in complex ways, and the result can be behaviors that take off out of nowhere, or solutions such as behavior change interventions that look great in a study but fail to work elsewhere, or that fade in effectiveness over time.

What this means is that overly generalized, paint-by-numbers approaches to behavioral science are likely to fare poorly unless the problems they are applied to are essentially well-contained and rote. It also means that, when a solution aimed at behavior is developed, it may need to be tested in ways that take direct account of the way its efficacy can vary over time, contexts, and people, depending on the costs that would be incurred from poor generalization.

The good news is that behavioral science, when used robustly, can be leveraged to address many of these very challenges. Corners of the scientific literature that are mature can often point to where behavioral drivers may interact or have boundary conditions, which can then be capitalized on for understanding and taking actions with behavior that are quite sophisticated and even more likely to succeed. Likewise, methodologies can be selected from the behavioral science toolkit that can provide the means by which the efficacy of a solution aimed at behavior is given a full, proper accounting. Yet, even after dealing successfully with these issues, there would still be others that would need to be tackled to ensure that the applications of behavioral science are set up for success. And that owes to the fact that –

#3: The practice of behavioral science is not clean and tidy

It would be nice to end this article with a picture of behavioral science as a cooperative effort oriented strictly toward the quest for behavioral truth – but, alas, such is not to be. That's because behavioral science is a messy human endeavor, fed from the top by academic scientists who, under intense pressure to differentiate themselves, are incentivized to conduct exciting work at high volume that can yield grant money, good press, job security, and, if they can win the Nobel Prize, immortality. Obviously, these incentives can sometimes result in cases of outright fraud, but they can also have other, more common consequences that can be quite difficult to spot:

Hidden pockets of weakness within the scientific literature that only become apparent once highly-publicized findings fail to replicate
Scientists who, working in their own silos, choose to define theoretical concepts, or measure variables in their studies, in different ways despite focusing on the same phenomenon, making any attempt to integrate their findings difficult or deceptive
Publication of effects that are quite trivial in size, but which make it across the finish line because they meet the conventions that govern publication decisions

These features of the science may be lamentable, but they are not insurmountable. The key to minimizing their impact, though, lies with discernment, and, if knowledge from the scientific literature is going to be used, the ability to review it in such a way that hints of flakiness can be detected, the agendas of individual behavioral scientists can be skirted, and judgments about what to leverage can be made based on what appears to be the tried-and-true. These are not matters that can be dealt with effectively when learning about behavioral science from a book or one-day workshop: They require considerable knowledge and skill, acquired from years of training and experience, to properly navigate. And that means that, if you’re going to reach for behavioral science, your best bet is to make sure you have a well-credentialed behavioral scientist at your side to guide you through the terrain and help you get the most out of what the science has to offer.

Summary

We started this post by asking what behavioral science is. As with any science, it’s a complicated discipline with pitfalls that need to be respected. Yet, at the end of the day, it’s a powerful tool to have in the toolkit when success hinges on the development of anything, from communications and policies to services, products, experiences, or other interventions, that has shaping, supporting, or accommodating people’s behavior as a fundamental objective. That’s because it’s a broad-based body of knowledge, set of methods, and evidence-based attitude that looks to understand people’s behavior, and develop ways to impact it, that takes into account the full complexities of behavior, doing so in a manner that’s deeply answerable to what drives behavior in real life. There are ways to bring this body of expert knowledge to life for practical use, and applied behavioral scientists have the knowledge and skills to make these applications possible. With their support, behavioral science can be activated and operationalized with great effect to:

Help you obtain new insights into people and their behavior that can place the design of solutions aimed at behavior on an advantageous footing;
Support a blueprint for the process of solutions development that’s rigorous, tightly focused on behavior, and leverages behavioral experimentation to drive evidence-based decision-making; and
Encourage a point of view regarding the problems of behavior and how to address them that aligns an organization’s way of doing things with what it ultimately takes to make efforts to shape, support, or accommodate behavior succeed.