Why We Think We're So Much Smarter Than We Really Are
... and why there may be a serious cost.
Posted August 2, 2015
"…as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns—the ones we don't know we don't know."1 — Donald Rumsfeld
Humans have always sought ways to enhance the finite capacity of our brains, in particular by relying on other resources to hold information for us. Prior to the Internet, we depended on our co-species to share the immense burden of information required to successfully navigate the world—such as which plants were poisonous and which were edible. We continually updated these shared banks of knowledge through ongoing face-to-face interactions, and as we evolved, we supplemented our human memory banks with tools such as books. When we source information from other people, and even when we rely on non-human sources, such as books, there is no confusion between what we know and what the source is providing: We are aware of the limits of our own knowledge.2
In contrast, new research has revealed that when we use the Internet, then, as Rumsfeld might say, we no longer know what we know and what we don’t.
In a series of nine experiments involving more than 1,000 participants, Yale researchers found that searching the Internet creates an illusion of knowledge, in which we conflate information that can be found online with the actual knowledge in our heads.2 Participants were told to look up the answers to some simple "How" or "Why" questions commonly entered into Google, such as "How does a zipper work?" The questions were simple enough for most people to have a general sense of the answer, and they were asked to search the Internet to confirm the details. Participants who had looked up explanations in this way later rated themselves as significantly superior in their ability to give explanations to a set of completely different, unrelated questions, as compared to a control group whose members had not used the Internet. Additionally, those who had used the Internet to answer the original questions expected that they would have increased brain activity, corresponding to higher-quality explanations, when answering the second set of unrelated questions.
Further, it turned out that accessing the Internet alone did not justify the overconfidence in one’s own knowledge. Rather, the knowledge illusion was specifically driven by the act of searching the Internet, regardless of the type of search engine used, whether the searching produced any relevant answers, or indeed any answers at all. It seemed that the very act of searching for knowledge on the Internet fools the brain into thinking we have more answers than we really do.
We already know that we significantly outsource our thinking to technology. The results of these new experiments suggest that this outsourcing habit conceals from us the extent to which we rely on external information. As technology makes information ever more readily available and accessible, the lead researchers of this study predict, the ability to assess one’s own knowledge will only become more difficult.1
By making the error of assuming we are relocating external information into our own heads, we may unwittingly exaggerate how much intellectual work we can do in situations in which we are, in reality, on our own. For example, a student using the Internet to study may only discover how little he actually knows once he is ensconced in an exam room. What seems like a relatively innocuous shift from personal to technological dependence for information could indeed have profound effects on how we make sense of daily life.
1. Rumsfeld, D.H. (2002, February 12). DoD News Briefing—Secretary Rumsfeld and Gen. Myers. [Transcript]. Retrieved from http://www.defense.gov/transcripts/transcript.aspx?transcriptid=2636
2. Fisher, M., Goddu, M. K., & Keil, F. C. (2015). Searching for Explanations: How the Internet Inflates Estimates of Internal Knowledge. Journal of Experimental Psychology, 144(3), 674-687. http://dx.doi.org/10.1037/xge0000070