Is Your Favorite Website Performing Psychological Experiments On You?

Andrew Rich / Getty Images

Facebook recently drew criticism and even outrage when users learned that the popular social media site had subtly altered the content of their news feeds in order to provoke an emotional reaction as part of a large-scale, secret psychology experiment. Nearly 700,000 Facebook members who used the site in early 2012 unwittingly took part in an experiment looking at "emotional contagion."

Apparently, Facebook manipulated the content found in the news feeds in order to look at emotional responses to positive and negative posts.

Some users experienced a flood of positive posts while others were exposed to a deluge of negative posts.

According to the results, "the emotions expressed by friends, via online social networks, influence our own moods, constituting, to our knowledge, the first experimental evidence for massive-scale emotional contagion via social networks," the authors explained in a paper published in the Proceedings of the National Academy of Science.

Users were understandably upset at what some later described as "creepy" emotional deception and even "mind control." One U.S. Senator went so far as to ask the Federal Trade Commission to investigate the incident. The social media giant later issued a public apology for the manipulation.

While many people described the intrusion as disturbing, unethical, and anxiety-provoking, it leaves many to wonder just how common such experiments might be. Could some of your favorite websites be conducting psychology experiments on you without your knowledge?

Online Experimentation Is More Common Than You Think

While it might come as a surprise to some, websites frequently conduct a wide variety of experiments without user knowledge. In many cases, these experiments center on user behavior. For example, a website might be interested in learning if changing the font or position of certain website elements leads to higher user engagement and click-through rates.

Such experiments are generally referred to as A/B tests. Some users might see one version of a website (version A), while others see a varied version of the same site (version B). The results are then analyzed to determine if altering elements of the page resulted in differences in user behavior.

It's also hard to imagine a world in which marketers don't deliberately engage in attempts to influence consumer behavior. Many retailers - both online and offline - mine huge volumes of consumer data to determine how to best influence what people buy, and then utilize that data to make marketing, advertising, and merchandising decisions. Any woman who has ever started receiving a barrage of coupons for diapers and baby formula after a visit to an obstetrician probably realizes just how readily marketers access even the most deeply personal information.

So what is it about the Facebook experiment that drew so much ire from users? One possible reason is that the experiment involved a direct attempt to manipulate the emotions of users without their knowledge - and in a way that could potentially lead to negative feelings.

Other critics suggest that the most disturbing aspect of the Facebook experiments was the lack of truly informed consent from the involuntary participants. While the social media site might argue that implicit permission was given since a research clause exists within the site's Terms of Service, the fact is that very few users ever thoroughly read TOS agreements. In order to obtain legitimate informed consent, the participants would have been informed of the possible risks, duration, and purpose of the experiment and told that they could quit the study at any time.

In an opinion piece, the Los Angeles Times Editorial Review Board condemned Facebook for the experiments. "It was unethical for Facebook to conduct a psychological experiment without users' informed consent. And it was especially wrong to do so in a way that played with the emotions of its users. That's dangerous territory," they wrote.

OKCupid Admits to Psychological Experimentation

But Facebook isn't the only website performing research that some users might find invasive or even deceptive. In one blog post, the popular online dating website OKCupid proclaimed that they regularly perform such experiments ("We experiment on human beings!").

"If you use the Internet, you’re the subject of hundreds of experiments at any given time, on every site," wrote Christian Rudder, one of the site's co-founders. "That’s how websites work."

(Disclosure: and OKCupid are both owned by the same parent company, IAC).

Such experiments can range from the subtle – such as experimenting to determine what time of the day is the best time to publish a blog post in order to get the greatest response – to the to more complex. In various experiments cited by OKCupid, users had their photos removed, had personal descriptions hidden, or were told that they had high compatibility with others users when they were actually bad matches.

Like the Facebook experiments, the reaction to OKCupid's user experiments has been largely negative. Many users argued that making relatively minor changes to elements of a website and then observing the resulting change in user behavior is not the same as willfully deceiving customers into believing that they've been paired up with good romantic matches.

So are such experiments ethical? In a post for the Harvard Business Review, writer Michael Luca suggests that while controversial among users, the Facebook experiment would likely pass muster with most Institutional Review Boards since the possible harm was relatively minor. "OkCupid’s is another story," Luca went on to write. "Because the experiment involved telling users that their compatibility scores were high when they actually weren’t, it probably wouldn’t have gotten through many schools’ IRBs unless participants were asked for their consent."

The bottom line:

While many online users might be uncomfortable or even angry at the idea of being part of an experiment without their knowledge, Rudder is right about one thing - it is a standard practice and virtually every website out there engages in some form of experimentation. From online shopping sites to educational portals, companies can learn a tremendous amount about what works (and what doesn't) by experimenting on their users.

Do users prefer Helvetica or Arial font? Are readers more likely to click links placed on the left column or the right column? What font size do people find the most readable? These are all questions that website developers and designers want to know, and experimentation can offer critical answers that can help both website owners and users. Where to draw the line, however, is a matter of debate.

One solution offered by Mikolaj Jan Piskorski, a professor of social strategy at the International Institute for Management Development, is to utilize observational studies in which information can be collected without actually manipulating users. "We use natural experiments to overcome ethical problems that arise in randomized experiments," he explained to The New York Times. "I think websites should consider more of these natural experiments even though they are harder to pull off."

Even the lead researcher of the controversial Facebook study now admits to having second thoughts. "I can understand why some people have concerns about it, and my coauthors and I are very sorry for the way the paper described the research and any anxiety it caused," Adam Kramer explained in a June 29, 2014 Facebook post. "In hindsight, the research benefits of the paper may not have justified all of this anxiety."

As for OKCupid, at least a few legal experts suggest that some of the site's experiments appear to run afoul of FTC rules regarding deceptive practices.

While we may never be aware of which websites are experimenting on users (unless they choose to disclose it), the debate over what's ethical or even legal is sure to remain a hot-button issue for years to come.

Continue Reading