The face as a mirror... of the pocket!

Pedro Rey


As this is my first post on this blog, let me introduce myself. I am a researcher into behavioural economics, a field that uses ideas from psychology to enrich the study of economic problems. More specifically, at present we are working on a project that – as I will explain at the end of this piece, and for various reasons – may have interesting implications in the field of medicine.

Our research aims to answer a simple question: what is the relationship between how much consumers like the products they buy and how much they are willing to pay for them?


This is a vital question is for economists, as we construct all our theories around the assumption that there is a very close relationship between the two. However, it is very difficult to prove the assumption, as that requires us to ascertain two things that are also difficult to measure. There is no clear way to objectively measure how much a consumer enjoys a product, nor is it in the consumer’s interest to tell us how much they are really prepared to pay for something. For this reason, we conducted an experiment in a theatre, the Sala Becket in Barcelona. There, we asked spectators to fill in questionnaires before and after the performance. Crucially, admission was not by conventional pre-paid ticket; rather, spectators could pay what they wanted after the show. The questionnaires provided two important measurements of enjoyment: how much spectators had enjoyed the show; and how much they expected to enjoy it before coming to the theatre. Asking audiences to pay what they want enables us to compare willingness to pay with what they really pay. Obviously, spectators could decide to pay nothing, but the production company we worked with, Sixto Paz, has been using the “pay what you want” strategy successfully for some years now. The most interesting result we have noted so far is that the amount paid depends less on how much spectators say they enjoyed the play than on the difference between how much they thought they would enjoy it and how much they actually enjoyed it. We therefore conclude that the most important thing is to satisfy expectations and that, although, in principle, one might think that it would be best to attract potential consumers by greatly raising these expectations, there also exists the risk of disappointing them. Accordingly, the greatest care should be taken when designing advertisements.

Nevertheless, we find using the method conventionally employed in marketing to measure levels of consumer satisfaction, the questionnaire, somewhat unsatisfactory. This is because the fact that we can find a link between how much a consumer pays and how much they say they enjoy a product may be due, simply, to the fact that we all try to be consistent. For this reason, we also used technology to obtain an alternative measure of enjoyment: deploying high-definition cameras, we filmed spectators during the performance, then processed the results using facial recognition software. This enables us to ascertain precisely, for example, how many times and how intensely a spectator laughs during a comedy, and how much attention they pay to the show. Obviously, people go to see comedies in order to enjoy a good laugh, so how much someone laughs can be a good way of measuring their enjoyment. We therefore decided to seek the relationship between data on payment and on facial expression.

I mentioned at the beginning that our study may be of medical interest. There are two reasons for this. Firstly, the show we used for the experiment, Lucy Prebble’s play The Effect, about a clinical trial for a new antidepressant drug. The plot is of interest to researchers, as the lead characters fall in love during the trial, and it is not clear whether this is due to the effects of the drug or whether the fact that they fall in love spoils the study, as it seems to demonstrate that the drug works, though this is not really proven. This gave us the idea of including this theme in our own experiment by conducting a clinical trial with the audience. Before the play, spectators were asked to take one of two pills: a placebo or a “real” pill (in fact, another placebo), which we said we were studying to see whether it enabled people to enjoy the show more. A surprisingly high proportion of spectators actually took one of these pills, without knowing which, and this enabled us to study whether the fact that someone is told that they have taken something that will increase their enjoyment effectively raises their expectations, their enjoyment or even how much they are willing to pay for a product. We have not yet obtained the final results, but we can note that the “supposedly effective” pill – like advertising which claims that those who buy a product will feel better – appears to generate positive effects on all three variables.

Moreover, the employment of facial recognition software may also have clinical utility. In particular, just as it is by no means easy for economists to measure a consumer’s enjoyment of a product, so doctors have difficulties in objectively assessing the degree of pain that a patient feels, and this can present problems when it comes to taking decisions about treatment. In both cases, the problem is similar: how do different people interpret the same scale? Different people may rate the same pain as a “7” or a “9”. That is why we are also conducting research to see whether we can use patients’ facial microexpressions when different levels of pain are applied to establish a new, more objective scale or link such a scale to conventional methods of assessing pain. As I hope to continue contributing to this blog, I will reveal the results in the near future. For now, I leave you with a video to give you an idea of the direction our research is taking.