So I watched this talk of Dan Ariely's with BigThink (embedded below). He covered interesting topics as always. But he also brought up my 'pet hate' when people talk about rationality and self-interest and how they are often conflated when people talk about behavioural economics in the popular press. What's the point here: a rational actor maximises the payoff or utility that they gain from taking certain actions, that is consuming goods, engaging in certain activities, etc. A self-interested individual who maximises their money has rational preferences over money and is self-interested and will therefore engage in activities to maximise the amount of money that they have, calculating the costs and benefits in order to do so. But a rational individual who has other-regarding preferences - that is they care about how other people behave - an individual who does so rationally may give some money away, may behave trustingly, and may punish individuals she sees infringing what she may perceive to be social norms. Notice that such an individual remains rational, but her preferences are not the preferences of an entirely self-interested individual who only wants to increase the amount of money that she has.

Anyway, I feel the need to bring this up because far too many people talk about rationality and self-interest interchangeably when they are not. One of the problems we encounter, for example is that you could hypothetically have irrational individuals, some of whom are otherwise self-interested and some of whom are other-regarding. We might not be able to differentiate between these individuals if their irrationality is such that they behave in ways that do not maximise what we perceive to be their preferences. But we may still have rational actors who are self-interested (also called self-regarding), other-regarding or bits of both.  So when Ariely says that trust and punishment (vengeance) are irrational he is not actually defining the problem properly, or he's assuming that the preferences of being are in fact the preferences of someone who is an entirely self-interested money-maximising individual.  I believe that the evidence indicates that most people are not wholly money-maximising and only self-interested - their preferences are structured differently.  Consequently his rumpus about rationality is a poorly constructed problem about preferences and not a problem about rationality. 

Later he is asked about companies and irrationality.  He then talks about something 'making sense'.  Logic and rationality are not the same thing.  This is a sophomoric error.  He talks about focus groups being less useful than we think they are.  If this were irrational then it would mean that they do not help companies to make profit, because rational companies go about maximising profits (or maximising share prices or some other goal).   

Ariely finally nods his head to other-regarding or social preferences toward the end of the interview when he talks about 'society' and social norms, but places nowhere near enough emphasis on it given the power he attributes to 'irrationality' and the time, content and rhetoric he dedicates to 'irrationality'. Oh well... 

Otherwise, his commentary on reward-substitution, differences in time preferences and other phenomena is interesting and apposite.  I recommend that you take a look at the video and see what Ariely has to say, but make sure that you realise he's trying to package the talk more accessibly.