Humans: The Lab Rats of the Internet

OK Cupid, a popular dating site, recently released a blog post titled “We Experiment on Humans Beings!” (Note the very fitting logo). The post proudly explains that the company has conducted a series of experiments on its users in an attempt to optimize its algorithm. The caveat is users were unaware that they were the subjects of an experiment. OK Cupid paired bad matches together and told the users that they were actually good matches to determine the effectiveness of its algorithm.

In the podcast, An Imperfect Match,  the hosts talk to the president and co-founder of OK Cupid, Christian Rudder. Rudder defends the companies actions with the in-the-end-it-works-out-better-for-everyone utilitarian argument. However, the hosts don’t buy it. They feel that informed consent is a crucial part of social science experiments. Additionally, the hosts explain that companies performing experiments seem to ignore the ethical standards and safe guards typically present in social science experiments that are necessary to protect the subjects.

Both the OK Cupid blog post and the podcast mention the outrage over similar tactics used by Facebook back in June. (SparkNotes: Facebook manipulated users’ newsfeeds to determine the affect on their moods without gaining informed consent). However, many websites, not just Facebook and OK Cupid, track user activity to gauge experience and develop performance metrics, which is why this is such a significant topic for discussion. Rudder argues that you’re the subject of science experiments merely by using the internet. This reminds me of the saying, “if you’re not paying for it, you’re the product.” OK Cupid, Facebook, and many other sites offer their services for “free.” But, is it really free or are we selling ourselves, our habits, our personal information, and our privacy? Are you willing to sacrifice these things for better performing sites? Have we entered into an assumed trade agreement without even knowing it?

The OK Cupid issue specifically has not gotten much press most likely because we have become desensitized to organizations breaching trust and privacy of its customers. Additionally, relative to larger scandals, such as Facebook, it is small – but it’s a slippery slope. Sites such as OK Cupid and Facebook have a profound ability to impact our lives, affect our moods, and influence our behaviors. As the hosts said, they are capable of immense “emotional manipulation.” Should something be done or are we signing an implied informed consent agreement just by signing up?


10 responses to “Humans: The Lab Rats of the Internet

  1. I really enjoyed your blog post and find a lot of these topics relevant to what goes on in each and every one of our lives; as we are currently on the internet reading these blogs. On one side, I completely agree that users have the right to know when they are part of an experiment due to privacy concerns. However, to play devil’s advocate, would businesses still thrive without the data of their users which allows for continuous improvement of a product of service? Is this data collection and analysis essential for a company?


    • I’m sure that those of us who have studied psychology or other social sciences have learned that one of the EPA’s five Ethical Principles for experimentation is informed consent. This has always been the one that, in my opinion, is most inhibitive to the collection of pure, accurate psychological data. I think so much of psychology hinges on how we shape our behaviors based on our expectations, so going into an experiment knowing what they’re going to do often defeats the purpose of the experiment because it inevitably affects their behavior and skews the data. Imagine if OKCupid had informed the people beforehand that their dates were bad matches. The participants wouldn’t have taken the date seriously and OKCupid would have no way to test or improve its algorithm, which is designed to help daters reliably find love in the long run. In most ethical situations I advocate full disclosure and transparency, but I don’t really think it can effectively be applied for the sake of psychological research, and the success of OKCupid’s algorithm depends on this very thing.


  2. The focus here was great. Of all things people in need have to be able to trust, its their dating sites. The idea that they could mess with peoples relationships gives them power over individual development. If they abuse this for experiments, the subjects could be hurt greatly.


    • Seriously, dating and relationships are serious and life-altering. What if someone who intentionally experienced a bad date through an OkCupid “experiment” had their confidence completely destroyed and now feel worse about themselves as a person as a result. As customers, OkCupid users do not deserve to have something so sensitive such as their love-life experimented on by a business without their knowledge or consent.


  3. How do we determine the difference between a social science experiment and in-depth marketing research? How do the ethics differ between these two? Where does anonymity come into play when using the data from your customers’ online activity?


    • The important questions. Companies use our information all the time for internal data analysis to improve customer service, product customization, pricing, marketing, etc. But where does it cross the privacy line? We willingly entrust our personal data to these organizations with the assumption it will be protected. However, instances like this prove that’s not always true. Interfering with personal relationships seems too far to me. I don’t have experience online dating (yet), but I know I would feel extremely violated if my confidential information was used against me.


  4. Would it really disrupt Facebook, or OK Cupid’s business models if they tried to get more informed consent?

    What if OK Cupid offered incentives to people who volunteered in randomized experiments? maybe their users would trust them MORE.


  5. I agree with what others are saying that this seems like a big invasion of privacy and people should have been informed. Also, I do not see why this was even necessary. Personally, I am not sure I understand why this experiment needed to have taken place. If they have a high number of successful matches wouldn’t that mean that their algorithm is good? Wouldn’t that be proof instead of the fact that they are capable of making bad matches?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s