Comments on Reluctance to Exchange Lottery Tickets – Introduction

Posted on August 26, 2009


As a way to bring together some of my early thoughts on Risen & Gilovich’s work on tempting fate, I am going to write down some relatively unformed comments on their first paper “Another Look at Why People Are Reluctant to Exchange Lottery Tickets”. I plan to go section by section so some of the comments may make little sense without that article in hand or may not seem to go anywhere.

R&G place their work in the context of two earlier approaches to the phenomenon of the reluctance to exchange lottery tickets. The first of these approaches explains the phenomenon in terms of anticipated regret – the regret making it feel worse to fail due to action than due to inaction. The second explanation claims that people may feel they have a better chance of success with their current ticket. R&G think that the second of these explanations has been shown to be implausible. At the same time, while they do think that regret avoidance plays a role, they think that they can put forward a third mechanism. The basic mechanism they suggest is that people think that the act of exchanging a ticket leads to its chance of winning increasing.

It is useful to observe that all three explanations work at the level of the psychological mechanism rather than at the level of why such mechanisms exist or, roughly speaking at the level of proximate explanation rather than ultimate explanation, to use the distinction from evolutionary theory. Thinking at the ultimate explanation level, one can ask whether the phenomenon of reluctance of exchange lottery tickets is a by-product, an adaptation or something else. Given that the tickets are assumed to have the same probability of winning and the cost of exchanging them is minimal, it seems hard to think of this as functional behaviour. Indeed, given that the same mechanism might be seen to work in the case of the three envelope problem, we may be dealing with potentially maladaptive behaviour. However, if that is so, are there cases in which the mechanism is functional? I believe so. Indeed, I believe that the mechanism responsible for reluctance to exchange lottery tickets is functional in most cases, i.e. that we are dealing with a heuristic and its systematic bias.

Very often, we find ourselves in situations when we have to chose between options that appear to be of similar utility. Having made a choice it is often important that we stick with it. This is because there will be costs involved with changing our minds. The worst thing would be if we kept on flipping back and forth between the available options, repeatedly paying the exchange costs.  Once way this result can be avoided is if the option we pick comes to be perceived as being clearly preferable once we have chosen it. This result might be obtained by any one of a number of mechanisms. The most significant limiting factor is that the mechanism has to allow for a change in selected option if there is new information about the comparative utilities.

These considerations do not directly eliminate any one of the three mechanisms R&G mention. They do, however, help us to think about questions that may prove to be useful, such as what is the scope of the phenomenon investigated by R&G. After all, it is unlikely that we are dealing with a special module for knowing what to do with small pieces of paper.

The more ‘rational’ of the mechanisms suggested is that of regret avoidance. Given that a change in mind is not going to objectively affect the probabilities of the different tickets winning, it might seem most reasonable to avoid repeated switching by making the result of having lost due to switching subjectively more unpleasant and letting the emotions play their cognitive role in discouraging such activity. For much the same reason, the mechanism R&G propose seems the least reasonable. As research has shown on numerous occasions, however, such distinctions make very little difference to evolution.

R&G make use of the dual aspect account of reasoning that I have already expressed my doubts about on numerous occasions. Yet, the use they make of it does not seem to fundamentally undermine their own account. There are differences between different kinds of reasoning processes and the phenomenon R&G investigate is likely to be due to processes that lie towards the intuitive, ‘innate’, illogical end of the range that has been identified in various ways with the so-called System 1.

More problematic is R&G’s attempt to link the phenomenon of reluctance to exchange lottery tickets with the phenomenon of tempting fate that they investigate in their other paper. This is not to say that I immediately reject such a connection but I am concerned that they have not made their case sufficiently and thinking about ultimate causes, as I have already started above, may reveal such a connection may be implausible.