Don’t follow lessons in popular Behavioral Economics books

There is now an increasing number of Behavioral Economics books, including Nudge, Thinking fast and slow and a slew of others. It may be tempting to read them and apply what they preach. But be careful, as many of them oversell hard the findings of the studies they describe.

Many BE results are coming from psychology, and to put it mildly, psychology is in the midst of an existential crisis as replication attempts after replication attempts fail. A good example is Kahneman’s insistence on priming on Thinking fast and slow, and how the whole corpus of priming results is now unravelling fast. Priming may just be an artefact after all, or in the most optimistic cases, its effects are much more context dependent than previously assumed.

Another great example is the supposed effect of harder to read fonts on cognitive performance1: supposedly, problems printed in harder to read fonts get better performance from people than easy to read fonts. This is again mentioned by Kahneman in his book.

Now after this single paper in 2007 showing an effect, a 2015 paper in the same journal2 reviews this 2007 paper and 16 other studies and concludes there is no effect.

All good you’ll tell me, science in action. Except that if you read Benartzi’s The Smarter Screen as Behavioral Economics applied to digital, then you will find several pages analyzing the effects of dysfluency (harder to read fonts, among other things). Then recommendations like these ones:

“b. Do you want people to remember what they read, as in a classroom? If so, then introduce a measure of disfluency to the content. Find ways to slow the mind down.
c. Do you want people to reflect on the information, as in the case of mortgage rates or cigarette warnings? This also calls for a level of disfluency.”

Excerpt From: Shlomo Benartzi. “The Smarter Screen.”

Great, now you have recommendations to start using hard to read fonts on your website or packaging based on 1 study out of 17 that shows people perform better when reading harder ro read fonts.

Listen to pop-academic advice or even academic advice at your own risks. Academics don’t suffer any consequences if what they publish turns out blatantly false.

  1. Overcoming intuition: Metacognitive difficulty activates analytic reasoning. Alter, Adam L.; Oppenheimer, Daniel M.; Epley, Nicholas; Eyre, Rebecca N. Journal of Experimental Psychology: General, Vol 136(4), Nov 2007, 569-576.,
  2.  Disfluent fonts don’t help people solve math problems. Meyer, Andrew; Frederick, Shane; Burnham, Terence C.; Guevara Pinto, Juan D.; Boyer, Ty W.; Ball, Linden J.; Pennycook, Gordon; Ackerman, Rakefet; Thompson, Valerie A.; Schuldt, Jonathon P. Journal of Experimental Psychology: General, Vol 144(2), Apr 2015, e16-e30.