Marc Le Menestrel
Home > Research > A Model of Rational Behavior Combining

A Model of Rational Behavior Combining Processes and Consequences

Save this article in PDF

by Marc Le Menestrel (19/10/2010)

Presentation of my Ph.D. Dissertation

Rationality is a central concept for the understanding of human behavior. In management sciences and in economics, it is often treated as the choice of the action that leads to the preferred consequence. This consequentialist standpoint amounts to considering that actions have no intrinsic value in themselves and considerably narrow our perspective. The intention of my dissertation was to propose a model where both values on actions and on consequences would contribute to rational behavior. With this dual approach, one could complement the standard consequentialist approach to rational behavior with a more explicit consideration of subjective values such as “ethical values”, “aesthetic feelings”, “cultural tastes”, “emotional concerns” or “spiritual concerns”.

The fundamental novelty that I was proposing consisted in the articulation of two types of reasons with distinctive essential characteristics. By combining a procedural and a consequential entity, we would manipulate dual types of values that would be irreducible one to the other without being fully independent. In the past decade, I have passionately developed these ideas further and consider them now as much relevant and fruitful. I briefly share below a few developments for which this dissertation acted as a foundation.
The two classical “puzzles” of standard rationality discussed in my dissertation, namely the so-called “Allais paradox” and the “Prisoners’ Dilemma have been published in a form very similar to the one presented here (Le Menestrel 2001, 2006). The time it took to publish the second piece is a good indication of how persistent one may have to be in order to find a place within the peer-review system!
In the dissertation, I propose a mathematical model where a factor for procedural concerns combines multiplicatively with the more standard consequential utility. Since then, I have worked on the axiomatic foundations for such a model, thereby providing insights as to how one could provide both a more rigorous measurement of utility and an innovative measurement of how observed behavior may depart from the maximization of it. This works shows that the combination of a multiplicative factor and a value function can model a much wider class of behaviors than those considered rational by the standard model. It has been published as Le Menestrel and Lemaire (2004, 2006a, 2006b) and Lemaire and Le Menestrel (2006, 2009).

I have used the distinction between procedural concerns and consequential motives to approach the articulation of ethical values and self-interest in my teaching of ethics in business decision-making. I found this dual approach particularly interesting to analyze the psychological aspects of ethical decision-making at the individual level as well as the strategic aspects of business decision-making at the corporate level. The general model has been published in Le Menestrel (2002) and an application to the strategies of the oil industry can be found in van den Hove et al. (2002) and Le Menestrel et al. (2002). The community of Operations Research has revealed a most interesting audience to share such a methodology, and that lead to a series of papers published in that field (Le Menestrel and Van Wassenhove, 2001, 2004, 2009; Le Menestrel 2006; Rauschmayer et al. 2009).

It is only at the end of my dissertation that I began to use the metaphor of the biased balance to explain how two types of values could combine to produce an observable phenomenon. On the one hand, there are the objects that are put on the arms of the scale and are measured with a function and on the other hand, there is the bias of the fulcrum that distorts the behavior of the balance and that is measured as a multiplicative factor. In a sense, the model proposed in my dissertation prefigures what I considered a promising approach to the foundations of science. Instead of focusing on the properties of objects and of their relations independently of the measuring device through which phenomena are observed, one should consider the articulation of both the objects and the measuring device itself, which would not necessarily be neutral and would induce departures from the expected behaviors. In this sense, the physical sciences from which standard theories of rationality originate should rather be taken as a limit case of social sciences, a restrictive setting according to which phenomena would be treated independently of the subjects that are there to observe them. As of today, this work still remains at the stage of working papers (Le Menestrel 2010).

I like to say that objectivity, rigor, formalism and models should be seen as means towards a better account of our essentially subjective, qualitative and human values. This departs from the idea that our rationality is “bounded” because we fail to follow the tenet of an illusory mechanical model. I believe that our subjective nature is more than a constraint or an obstacle to our rationality. To the contrary, I believe that we human are much more intelligent than machines or inert physical systems. Moreover, I believe that a better account of our subjective nature could help us in improving our respect of human values, our quest for social justice and our relation with nature.

Click here to download my Ph.D. Dissertation: A Model of Rational Behavior Combining Processes and Consequences.