More and more, as we investigate how psychology and behavioral science can be used in design, the topic of morality comes to light. Where does the ethical boundary lie between persuasion and manipulation?
Neutrality in design
When we’re hired as designers it’s our job to make decisions. It doesn’t matter what we’re creating (a sign up form, toaster, or mousetrap), we are responsible for making decisions that affect how the object is perceived and used. We conduct research and observation to better understand the motivations and mental models of our target audience. It’s with this information that we’re able to bring shape to a design and affect each user’s experience. Whether discreetly or overtly, each content and design decision biases a response.
In that sense, there is no such thing as neutral design.
The spectrum of influence
Every choice a person makes is made within the context of influence. Some of these may be intrinsic: biases formed by things like past personal experiences or societal norms. Others are extrinsic, the result of external motivation. When a person interacts with an object, the interface itself becomes a source of influence. Consider the concept of affordance, in which an object provides cues about its intended use while restricting improper use. Every time we design a system or interface that compels an individual to an action, it falls somewhere within a spectrum of influence. Most forms of influence fall into one of the following categories:
- Unintentional influence
When influence is not deliberate, are we ethically responsible for its consequences?
Sometimes influence occurs unintentionally. Any arrangement of elements within a space will provoke a response, even when intention is absent. It doesn’t matter if the objects are placed haphazardly within their environment, their unique arrangement will trigger unique responses. Antipatterns are a good example of how a user interface can influence behavior in unintended ways. When a design lacks intent, any benefit or harm to the user is coincidental.
- Open / candid influence.
Is it enough to simply be candid about our intentions, or does a greater morality come into play?
Open forms of influence are transparent in their attempt to persuade audiences. An example of this would be the suggestive arrangement of elements or objects within an environment. When we use visual hierarchy to give priority to the elements of a design, we do so in an attempt to direct the user’s flow of focus from one element to the next. When a store promotes a product in an endcap, the intention is obviously to drive sales by persuading consumers into purchasing it. Call to action content can be one of the most direct forms of candid influence.
- Incognizant influence
Does the imperceptible nature of subliminal influence make it more or less devious than overtly deceptive types?
Some types of influence aim to affect behavior on a subconscious level. These are the kind that Robert Cialdini discusses at length in his book, Influence: The Psychology of Persuasion. Incognizant forms of influence are often the most subtle, as they work to change peoples’ behavior without creating an awareness that any form of persuasion exists. Understanding concepts such as social proof, scarcity and authority allow us to get inside the minds of our audience and provoke them into action. Because people are generally unaware that they are being influenced, there’s a lot of opportunity for these techniques to be used in an exploitative manner.
- Deceptive influence
Are there any situations where the use of deceptive influence can be justified?
Deceptive forms of influences are those where the outcome of an interaction do not match the user’s expectation. Sometimes this is accomplished through a deliberately confusing interface, and other times it’s by presenting misleading information. Either way, the intent is to manipulate a person’s behavior through the use of trickery. Most of the time these forms of influence are designed to achieve business objectives at the expense of site’s users. Victims of deceptive forms of influence eventually become aware that they’ve been taken advantage of, and this can have long term negative impact on the brand.
Coercion may not technically qualify as a form of influence, since influence implies that a choice of action exists. Coercion is a forced action or an action that is performed without the consent of the user. Often, other types of unethical influence are used to trap people into unfavorable positions where they are coerced into disclosing needless personal information or purchasing more of something than they actually need. No one would ever opt for coercion, because individually we all believe that we’re the ones best equipped to choose what’s in our own best interest.
Does this mean that offering more choices is always better?
Understanding how choice affects decision making plays an important role in the ethics of influence. Choice can be broken down into three broad categories:
- full freedom of choice: the entire range of available options are presented
- limited choice: the full range of available options is edited down to suggested ones
- no choice: no option or alternatives are provided
Choice architecture is the term used to describe the curation of options we present to an audience. On one side we have “no-choice”. This is clearly undesirable as it takes away a person’s freedom. On the other side we have complete freedom of choice, where every possible option is presented. This can be equally undesirable as the sheer number of options presented may be unmanageable. It would also require an audience to have a thorough understanding of the subject in question in order to comparatively select the most favorable option. When we limit the choices that are presented to a person, the design becomes suggestive. But suggestion is by no means unethical — in fact, it’s necessary.
Influence is necessary
The majority of decisions that we make in the course of a day are not made by selecting the preferred option among a rational analysis of all possible outcomes. If we were to do this, our brains would unleash the spinning beach ball of death, rendering us paralyzed each time we faced even the smallest decision. There’s a capacity to how much relevant information we can process, so our brains rely on mental shortcuts to reduce the effort needed to navigate through the day. The brain does this by subconsciously picking up on subtle cues that trigger emotional or pattern responses, prompting us to respond impulsively to different situations. Many of these cues are the result of external influence, nudging us toward one option amidst many similar competing options. Think about choosing toothpaste (or beer) in a supermarket: our decision is ultimately influenced by the packaging and our gut feelings about the brand as opposed to a quantitative analysis of “tastes great” vs “less filling”.
Influence is complex
The light side of the force
Highly persuasive designs will appeal to a person’s sense of reason while also generating a response on an emotional level. Think about how curiosity might be used to entice a customer to find out more about your services, or how injecting bits of humor or personality into your site might make your brand more likeable. Aarron Walter, Stephen Anderson and Andy Budd have all done a lot of work describing how principles of neuroscience and psychology can be used to create more seductive, emotionally engaging interactions. When used appropriately, these techniques can compel audiences to spend more time on our sites, become emotionally attached to products or services, and walk away smitten by our brands.
Those that understand the predictable response patterns of humans also have the potential to wield that knowledge against us. The same mental shortcuts we rely on to help us make decisions can also be exploited.
The Dark Side
As we become more adept at influencing the behavior of people, we should be wary of encroaching the realm of manipulation. Manipulation could be described as being a more selfishly motivated form of influence, encouraging others into an action that suits one’s own advantage. In UX design, tactics designed to take advantage of others are called Dark Patterns, a term coined by Harry Brignull.
Dark patterns are clearly unethical.
Dark patterns are boobie-traps blatantly designed to mislead people. Examples of dark patterns include sneaking extra items (like trip insurance) into a shopping cart, disguising ads as content, or abusing users’ privacy by spamming people in their network. The site wiki.darkpatterns.org has a comprehensive list of these black hat techniques, providing numerous examples and resources.
Does the end justify the means?
Most examples of dark patterns rely on cleverly crafted interfaces or deliberately confusing interactions. While beneficial for the perpetrator, the outcome is usually detrimental to the user. Because of this, it’s easy to label them as unethical.
What if dark patterns were used in a way which resulted in a beneficial outcome for the user? Would the end justify the means?
Even in this scenario, I can’t imagine it would. Any perceived benefit would likely be nullified once a person realizes that they’ve been deliberately misled.
So what about when incognizant forms of influence are employed? Is subliminal influence considered deceptive? Is there anything inherently wrong with understanding the patterns and fallacies of the human decision-making process, and using that knowledge to influence behavior? Is this knowledge really any different than the wisdom and intuition gained through years of professional experience? Assuming that no design decision is neutral, wise design decisions would naturally take into account human behavior and motivation, right?
This leads to another question: exactly what type of behavior are we influencing?
In the case of dark patterns being used for good, the deliberate deception meant that the end did not justify the means. But in cases of incognizant influence, what if the influence exerted helps the user accomplish their goals? When you join a social network, think about how social proof, commitment or rewards might be used to motivate you to encourage your friends to sign up. Expanding your network of friends will help you get the most utility and fun out of the application. Does the end now justify the means? Although you are still affecting behavior by playing on the automatic responses of the mind, the intention to deceive is not present.
Clarity of intention and expected outcomes
If what makes a dark pattern unethical is that it deliberately misleads, then perhaps an identifier for ethical influence is the degree to which motives are disclosed and outcomes are foreseeable.
If a website user can easily predict the consequences of performing an action, and if those outcomes are inline with the perceived intentions of the site owner, then there is an honesty to the interaction. Compare this to situations where the signals are mixed: where intentions and consequences are masked or do not match the expectations of the audience. This might suggest that if the motives of the site owner are clear and the interactions are honest, then any non-deceptive forms of influence used to encourage the user into action could be deemed ethical.
As always, things are not quite that cut and dry. In the example above, the first button appears honest but is deliberately designed to mislead. The second button succeeds as an honest interaction, meaning that there is no way to misunderstand the outcome of intended user action. Yet the action itself is immoral (from the human, not Dalek or Cylon perspective, of course). When you use honest interactions to influence harmful behavior, you shift the burden of morality from the site owner and interface to the end user. But although the weight of the moral burden shifts, it doesn’t free the designer or site owner from responsibility. The problem is that by sharing some of the burden of the moral decision, each party (the site owner, the designer, and the user) can assume a bit less responsibility for the outcome. It’s kind of a pluralistic ignorance effect.
“Well, I didn’t make anyone click the button,” says the site owner.
“Hey, I just clicked a button, I wasn’t actually the one destroying the planet,” says the user.
“What? Don’t look at me, none of this was my idea, I’m just the web guy,” says the designer.
We are responsible for our creations
Pluralistic ignorance is the cause of a lot of the shameful behavior in our industry. But as much as we are to blame, we ourselves are also victim to a form of incognizant influence: Authority, the idea that we naturally tend to follow the orders of those that are established as superiors. Too often we feel a sense of no-control, and look to our bosses or clients as the voice of authority. The lengths to which we’ll go under the instruction of authority is frightening. The most notable case is the infamous Milgram experiment in which participants in a study were ordered to administer increasingly painful electric shocks to a fellow participant. The experiment is a worrisome examination of the power of authority and how readily people will neglect to assume responsibility for the commands that are delegated down to them.
If you work in a team, you’re most likely only responsible for a portion of the final project. Chances are there’s someone above you managing the project, and other designers or developers building out other sections of the site. For freelancers, you may be in the hands of an overbearing and demanding client team. It can be convenient to do your job without thinking about the bigger picture. It’s easier to be passive than to be confrontational, and it’s easy to place the weight of responsibility onto your boss or the client. But the truth is that we alone are responsible for the things we create and contribute to making. As Trent Walton writes, You are what you eat.
A good friend of mine who works at a large agency was recently confessing to me how burnt out he felt. It wasn’t the long hours or impossible deadlines getting to him.
“I’m so tired of coming up with new ways to get people to buy shit they don’t need.”
At the other end of every interface is a person trying to accomplish a goal. What role did you play in creating it, and how might their experience change because of it? Do you use design as a tool to bring utility and pleasure to people, or as an exploitative tool that promotes consumerism?
We’ve grown accustomed to using design to support business goals, focusing on influence through a strategic lens. But something changes once we examine influence under a scientific lens; design principles start to seem a bit less abstract.
With an increased awareness of the psychological underpinnings of design, we’re forced to admit an uncomfortable truth: that to a large extent, humans are just emotional and biological machines with programmable behavior.
Maybe people are like more like sheep that we care to admit. And maybe we’re partly to blame. But through these predictable behaviors our human imperfections come to light. It’s here, within these inconsistencies, that our individuality lies. It’s what makes each of us unique, and so unlike machines.
I think that’s what’s at the heart of persuasive and emotional design. It’s the key to differentiating influence from manipulation. It’s the motivation to create situations that allow for beautiful and unanticipated interactions to occur in their own ways—the motivation to design something that doesn’t trick but delights—the intention to invite your audience to share your story through their own experiences.
To whatever end we choose to use influence in design, we’ll probably never come to a consensus on where the ethical boundaries lie. At the very least however, we can strive to be more conscientious, broadening our understanding of how and why our design decisions influence others.