man in black jacket holding black smartphone




You are, unfortunately, working late at night again. You have a report due in the morning but it isn’t quite done, so you’re sitting in your home office trying to finish it.

You decide to take a short break from the report and read one of the blogs you try and keep up with. You read a post from a well-known journalist who is about to leave on a trip overseas. He writes a kind of farewell post and at the end he signs it “Bye!” On the blog page, you see an ad for the journalist’s latest book with a “Buy now” button. You click and buy his book.

Next, you’re skimming a news site and you see a headline: “Is the Fed Chairman Right?” Suddenly you realize it’s getting late and you haven’t finished writing your report. You get back to work.

Was any of your behavior during that interlude “primed”?

Priming is when exposure to one stimulus influences your response to another stimulus. In the example above, you were primed with the word “Bye” (first stimulus) at the end of the blog post you read. Your exposure to “Bye” then affected your response to the second stimulus—“Buy” that was on the “Buy now” button.

That’s not all. Your exposure to the word “Right” (another stimulus) in the news headline primed you to think of the word “write” (the last stimulus), and made you realize you were supposed to be writing your report.

Priming With Homophones

Psychologists and marketing researchers have known about the effects of priming for decades. But realizing that priming works with homophones is a newer discovery.

A homophone is a word that has the same pronunciation as another word, but a different meaning, and sometimes a different spelling.

Here are some examples:


Derick Davis and Paul Herr (2014) wanted to see if homophones would act as primes, and if so, how strongly and under what conditions. Their idea was that homophones would activate the meaning and association of the related homophone, and that this activation would affect behavior. If people saw the word “bye,” they’d be more inclined to buy something.

People Subvocalize When Reading

When people read, they subvocalize—that is, they speak words internally. These words activate memories associated with the words. When people read the word “bye,” they associate meanings such as leaving or going on a trip. But because people subvocalize and say the word to themselves, Davis and Herr hypothesized that the association for the homophone “buy” would also be activated, and that the associations would stay in memory for a short time.

Since homophone priming is based on subvocalizing, this entire effect is different for every language.)

Suppressing Homophone Activation

Research on reading has shown that a lot of the time people automatically and unconsciously suppress the activation of homophone associations. The better a reader a person is, the more she suppresses the associations.

It’s a myth that good readers don’t subvocalize

If you’ve ever taken a speed-reading course, you may have been told that subvocalizing will slow down your reading. Don’t confuse moving your lips with subvocalizing. It may slow you down to move your lips while you read, but everyone subvocalizes (no sound, no movement). In the case of the homophone effect, it’s not that better readers aren’t subvocalizing. It’s that better readers don’t have as many automatic homophone associations.

So if you’re a pretty good reader, you might be less susceptible to homophone priming—unless you’re under a high cognitive load.

The Cognitive Load Gotcha

It takes some cognitive work to suppress homophone activation. This means that as you become more mentally busy—as your cognitive load increases—your susceptibility to the homophone priming increases, too.

Let’s go back to the example of working on the report late at night and encountering the homophones. Since you’re working late and you have a deadline the next day, it’s possible that you have a fairly high cognitive load, which would make you susceptible to the bye/buy homophone activation.

Embedded Homophones Have The Same Priming Effect

In their research, Davis and Herr tested embedded homophones too, such as “goodbye” and “bye” or “good buy.” They found the embedded homophones followed the same priming activation as single homophones.

Homophone Activation Is Unconscious

Davis and Herr tested 860 participants in their homophone research. None of the participants were aware of the homophone activation that had occurred.

Order sometimes matters for homophone activation

Homophones activate each other most when both words are common words (buy/ bye). But if one of the words isn’t that common (you/ewe), then the order matters for activation. If you see “you” first, you’re unlikely to have “ewe” activated (unless you’re a sheep farmer). However, if you see the word “ewe,” it’s likely that “you” would be activated.

An Ethical Dilemma?

So here’s one of those times when designers have an ethical dilemma. Do you purposely try to get people to do something by messing with homophones? Do you put the word “bye” in your blog post or on your website not too far from the word “buy” in a different context? Do you not only sprinkle the priming homophone, but also up the cognitive load on the page so the person will be even more susceptible? Should I have even included this information in this book if it means that some people will now use it to get people to take an action that’s not in their best interest?

I’m often asked about ethics in my work, because so much of the research I talk about has to do with how to get people to take action. It’s something I think about a lot, although I don’t have a quick answer. The basic question is: “If we use this information from behavioral science research to get people to do what we want them to do, are we being too manipulative? Are we being ethical?”

One point of view is that if you’re trying to get people to do something, no matter what it is, then that is unethical. Another is that if you’re trying to get people to do something that’s good for them (eat healthier, quit smoking), then it’s OK. I fall somewhere between these two ideas.

My take on the research I talk and write about is that these effects are powerful. However, there’s a limit. Using these behavioral science influences won’t give you total control over the other person. I also believe that each designer has to decide for himself or herself where the line is between influence and ethics. Every time you design something, you have to decide where that line is.

Here’s some idea on where my line is usually drawn:

I don’t completely agree with the people who say that it’s OK to use these techniques to change behavior related to eating, smoking, or conserving energy—things that help the individual or help society—but it’s not OK to get people to buy a new refrigerator. Trying to change behavior is trying to change behavior.

I’ve been an expert witness/consultant for the US government on cases involving internet fraud, and this has given me some insight into where the line is on ethical and unethical behavior. Putting your product or service in its best light, and matching your product or service with the needs and wants of your customers—these are OK. Does everyone really need a new refrigerator? Probably not. But encouraging them to buy a new refrigerator now, and to buy it from you, is perfectly fine. Otherwise, we might as well proclaim that all marketing and advertising is unethical, which is probably an opinion some of you have!

Purposely deceiving people, giving them confusing instructions so they don’t know what they’ve agreed to, encouraging them to engage in behavior that harms them or others, or trying to get them to break the law—these are not OK.

Interestingly, homophone priming falls right in the middle of those two places on the continuum. This means that in the work that I do I wouldn’t use homophone priming to get people to take the action I want them to take.

As I said, you’ll have to decide for yourself.


  • You can use homophones to affect people’s behavior.
  • When you want to increase the power of the homophone effect, increase people’s cognitive load.
  • People are not aware of the homophone priming effect. Think carefully about the ethics of this technique. Don’t use it if you find it to be unethical.


Leave a Reply

Your email address will not be published. Required fields are marked *