George Ainslie, Picoeconomics: The Strategic Interaction of Successive Motivational States Within the Person (Cambridge University Press, 1992, Pp. 440)



There is a simple view of motivation on which desires are like pain-killers: they come in different strengths, and their strength determines their efficacy. That is, the stronger a desire the greater its motivational force and, when two desires conflict, the stronger one “wins out” over the weaker. This view makes it puzzling how anyone could ever exhibit “strength of will” (“weakness of will” is the traditional puzzler) and act on the weaker desire, even when it is a desire for something more highly valued than what is more strongly desired. When the will, as the agent of reason, enters the fray, an explanation of what a person is motivated enough to do must reckon not only with competing desires and their motivational forces, but also with the will’s capacity to be affected by different ones in ways not proportional to their force. However, this complication sheds little light, for if the will must bestow its stamp of approval on a desire before the desire leads to action, then, it would seem, motivational force should be measured by strength of effect on the will. And if that’s the way to measure motivational force, then, since by that measure one always acts on the desire that affects one’s will more, there is no problem of strength of will after all. But the problem seems real, however inapt the label.

George Ainslie does not pit the will against desire but instead adds a whole new dimension to the simple view of desire. He applies this enriched model to a host of fascinating psychological phenomena, including impulsiveness, addiction, compulsion, ambivalence, procrastination, back-sliding, and self-deception. Although he is by profession not a philosopher but a psychiatrist, and although his writing is tinted by behaviorist jargon, his ideas are of great philosophical interest. I would go so far as to say (note its subtitle) that Picoeconomics (micro-microeconomics) is to intrapersonal conflict what Thomas Schelling’s The Strategy of Conflict (Harvard 1960) was to interpersonal conflict.

Ainslie starts with the platitude, which appears in Plato’s Protagoras (356), that people, when tempted, tend to favor the immediate and discount the distant. But he gives this platitude a new twist and applies it in a variety of surprising ways. The basic idea is that this tendency to discount temporally distant rewards (negative “rewards” included) is built into our psychology, and is not based not merely on specific factors like uncertainty about the future, distraction from long-term goals, conditioned responses to immediate stimuli, and gratification of biological needs. Ainslie’s idea of temporal discounting of personal rewards is the intrapersonal counterpart of Hume’s conception of sympathy for others (the more distant people are, in space, time, or kind, the less our sympathy for them). The added twist is that when we quantify this temporal drop-off, we find that its rate is anything but linear--it is hyperbolic. Ainslie’s graphical representations illustrate situations in which one desire is generally preferred over the other but there is a region of time during which the preferences are reversed. For example, on Sunday afternoon one might prefer being rested and alert on Monday to watching a certain late movie; that night, as show time approaches, watching the movie becomes more and more attractive; Monday morning, being rested and alert is again valued more highly. Relying on the idea of conflicting temporal perspectives, Ainslie proceeds to make intelligible the puzzling phenomena enumerated above (unfortunately, his graphs and often his discussion do not take retrospective attitudes like regret and guilt or satisfaction and relief into account). He suggests ways in which people’s inchoate appreciation of their changing temporal preferences leads them to adopt various strategies of impulse-control and self-reward. He cleverly applies certain game-theoretic concepts to conflicts of intrapersonal interest and identifies intrapersonal versions (involving competing interests rather than persons) of such predicaments as Newcomb’s problem, the paradox of deterrence, the backwards induction paradox, and the iterated prisoner’s dilemma.

Ainslie’s insightful and often virtuosic discussion of assorted examples of psychological conflict and irrationality sometimes seems to go beyond the conceptual resources of his model. This raises the question of just how much of the explanatory work is actually accomplished by the model itself. Take the concept of second-order attitudes, on which philosophers rely in addressing the fact, e.g., that people often realize that their values are disproportional to the strengths of their desires and may attempt to act on or else knowingly resist that realization. Ainslie uses that concept (in effect, if not explicitly) in his discussion of such phenomena as procrastination, vacillation, backsliding, precommitment, personal rules, and control of attention. However, it is not obvious that his model is equipped to capture second-order attitudes, inasmuch as hyperbolic discount curves reckon only with moment-by-moment preferences. Intersecting discount curves can represent changing preferences over time (never mind the unrealistically precise numerical quantification presupposed in plotting these curves), but it is not clear how this model is supposed to incorporate not only the play of conflicting interests at each of various times but the effect, at any given time, of assessments of those interests over time. In Ainslie’s model, though not always in his discussions, the agent is situated only at some particular point in time and able then only to weight future rewards relative to that time. So how can it matter to him now how something will matter to him later? In other words, how can the model explain why should strategic considerations ever prevail over tactical ones?

The worry here is that despite the sophistication of Ainslie’s psychological hedonism, with its marketplace of motivations, it may not fully succeed in accounting for the rationality of bargaining with one’s future selves. For example, despite his emphasis on the strategem of precommitment, as when one adopts a “personal rule,” for reducing the motivational effect of impulses, he does not make clear how, on his model, it is rational to obey the rule one adopts for resisting recurrent temptations. Why not one last fling or one more bite? Even if the personal rule does have motivational force, why should it? Related to this is the problem of integrating temporal perspectives and determining to what extent each is authoritative. There is the prospective outlook (when the tempting attraction is not yet available), the immediate perspective (when the reward is imminent), and the retrospective point of view (where some combination of satisfaction or relief and guilt or regret may be expected). A related problem is that Ainslie attempts to turn his descriptive model of the interplay of motivations into a normative model of how to manage that interplay. He gestures towards a kind of biological account of how hyperbolic discounting should have been selected for (his arguments seem inspired by evolutionary game theory), but there is a threat here of recreating the naturalistic fallacy.

The worries registered above do not undermine the liberating effect of Ainslie’s model. His book may be rambling and sprawling, but it offers a whole new take on a wide variety of psychologically fascinating and philosophically perplexing phenomena. One hopes that in the near future the author will heed some of his own strategic guidelines and produce a clearer and more concise edition of Picoeconomics, preferably with a more straightforward title, like The Strategy of Inner Conflict.