Konstantinos KarypidisEntrepreneur, Consultant, Trainer, Speaker

Professors Dan Ariely, George Loewenstein and Drazen Prelec some years ago did a very interesting experiment. They showed six different products to marketing students at Sloan Business School at MIT. The products were:

1) Wireless trackball

2) Wireless keyboard

3) A book for design

4) A box of Belgian Neuhaus chocolates

5) A bottle of Côtes du Rhône wine 1998 (relatively cheap)

6) A bottle of wine 1996 Hermitage (more expensive)

Then, they gave the students a piece of paper with the list of products and asked them to write the last two digits of their social security number at the top of the page. After that to write the same number next to each item in the list as if it was its price, and finally to write next to each price a “YES” or “NO” deciding whether they would buy each product at that price. So if someone’s two last digits of her social security number were 82, she had to write 82 to the top of her page, write 82 next to each item in the list as its price, and then with a “YES” or “NO” whether she would buy the product at that price.

Finally they asked the students to write the maximum amount that they were ultimately willing to pay for each item in the list.

Would you believe that a random number – such as 82 in our previous example – could influence the price that the students were willing to pay for each product? In other words, could we be confident enough that highly efficient brains – as supposedly MIT students have – are not going to be influenced about the real value of six products because of an absolutely random number?

The answer is NO! We couldn’t and moreover we shouldn’t! The result of this experiment was that the students whose last digits in their social security number were between 00-19 were willing to pay less than those who had as last digits 20-39, who in turn would pay less than those whose numbers were between 40-59 and these were willing to pay less from those with last digits between 60-79, leaving those with digits between 80-99 with the desire to pay the most of all the students for the same product list. The very interesting fact was that students were rational when they were evaluating the expensive and the cheap wine (they were all willing to pay  for the Hermitage of 96 than for Côtes du Rhône 98), but both prices were influenced by the original, random number that they had written at the top of their page.

The researchers were trying to prove what is known as arbitrary coherence. The theory of arbitrary coherence says that although the initial cost of products is arbitrary (arbitrary), once it is established in our minds will shape not only present prices but also future ones as well.

Almost always, I begin my presentations in companies and conferences by asking: “How many of you believe that you are sufficiently rational thinkers?” And of course the vast majority of the audience raises his hand.

This is something that never ceases to impress me, because I personally never had in high esteem the rational skills of my brain… I mean, how can I be sure that I SURELY KNOW? Also it frightens me, because anyone who is very confident in her ability to think rationally and maximize her utility at any given situation, will not willing to change or alter her assumptions or perspectives, in an extreme world where adaptability is absolutely necessary. Isn’t this serious? Surviving in this world – as far as I concern – means knowing that whatever brought you here, probably won’t get you where you want to be in the future, which in its turn means that you must always rethinking and retesting your approaches. So how can someone effectively do that when research shows that we are highly prone to be influenced by any kind of randomness?

Unfortunately I do not have an answer to that question. I have some thoughts and some strategies with which I am trying to be as effective as possible, but – for sure – I am not sure. Maybe the only applicable approach is first to learn how to unlearn what we already think we know and which very possibly, won’t help us to surpass the present challenges, and then plan as good as we can based on scientific research and our own – or other people’s – previous experiences. Then we should just go for it. The old, good, irreplaceable “Trial and Error method”.

The key – and the most challenging I believe – idea behind the process I just described is that we must stop being extremely confident in our ability to evaluate things and to predict the future. Whenever you are extremely confident in your rational thinking abilities, remind yourself the experiment we discussed about. And think twice whether you are arbitrary influenced or not.



  1. If you want to read the paper of the described research click here. For an easier description of it, as well as other incredibly interesting behavioral economics experiments, you can read Dan Ariely’s fantastic book “Predictably Irrational: The Hidden Forces That Shape Our Decisions”.
  2. For our ability and beliefs that we can successfully predict the future, I invite you to read Nicholas Nassim Taleb book called “The Black Swan: The Impact of the Highly Improbable


* Τhis article was originally posted on eudaimonialand.com