The expenses of getting it ready for publication were met by Milton Fund Grant 60-220-2119-2 from Harvard Medical School. I would like to thank Elmer J. Schaefer for making me aware of the economic literature relevant to this article. He, Bernard T. Engel, and William R. Whipple made helpful comments on the manuscript, but responsibility for the assertions made here remains my own. Requests for reprints should be sent to George

Pigeons were given a small, immediate food reinforcement for pecking a key, and a larger, delayed reinforcement for not pecking this key. Most subjects pecked the key on more than 95% of trials. However, when pecking a differently colored key at an earlier time prevented this option from becoming available, three of 10 subjects consistently pecked it, thereby forcing themselves to wait for the larger reward. They did not peck the earlier key when it did not prevent this option. This is an experimental example of psychological impulse and a learnable device to control it. Although only a minority of the subjects learned it, the fact that such learning is possible at all argues for a theory of delayed reward that can predict change of preference as a function of elapsing time.

Ainslie argues that our responses to the threat of our own inconsistency determine the basic fabric of human culture. He suggests that individuals are more like populations of bargaining agents than like the hierarchical command structures envisaged by cognitive psychologists. The forces that create and constrain these populations help us understand so much that is puzzling in human action and interaction: from addictions and other self-defeating behaviors to the experience of willfulness, from pathological over-control and self-deception to subtler forms of behavior such as altruism, sadism, gambling, and the 'social construction' of belief. This book integrates approaches from experimental psychology, philosophy of mind, microeconomics, and decision science to present one of the most profound and expert accounts of human irrationality available. It will be of great interest to philosophers and an important resource for professionals and students in psychology, economics and political science.

Relative to individuals who do not have addictive disorders, drug abusers exhibit greater devaluation of rewards as a function of their delay ("delay discounting"). The present study sought to extend this finding to methamphetamine (MA) abusers and to help understand its neural basis. MA abusers (n = 12) and control subjects who did not use illicit drugs (n = 17) participated in tests of delay discounting with hypothetical money rewards. We then used a derived estimate of each individual's delay discounting to generate a functional magnetic resonance imaging probe task consisting of three conditions: "hard choices," requiring selections between "smaller, sooner" and "larger, later" alternatives that were similarly valued given the individual's delay discounting; "easy choices," in which alternatives differed dramatically in value; and a "no choice" control condition. MA abusers exhibited more delay discounting than control subjects (P < 0.05). Across groups, the "hard choice > no choice" contrast revealed significant effects in the ventrolateral prefrontal cortex, dorsolateral prefrontal cortex (DLPFC), dorsal anterior cingulate cortex, and areas surrounding the intraparietal sulcus (IPS). With group comparisons limited to these clusters, the "hard choice > easy choice" contrast indicated significant group differences in task-related activity within the left DLPFC and right IPS; qualitatively similar nonsignificant effects were present in the other clusters tested. Whereas control subjects showed less recruitment associated with easy than with hard choices, MA abusers generally did not. Correlational analysis did not indicate a relationship between this anomaly in frontoparietal recruitment and greater degree of delay discounting exhibited by MA abusers. Therefore, while apparent inefficiency of cortical processing related to decision-making in MA abusers may contribute to the neural basis of enhanced delay discounting by this population, other factors remain to be identified.

Animal studies of impulsivity have typically used one of three models: a delay of reward procedure, a differential reinforcement for low rate responding (DRL) procedure, or an autoshaping procedure. In each of these paradigms, we argue, measurement of impulsivity is implicitly or explicitly equated with the effect delay has on the value of reward. The steepness by which delay diminishes value (the temporal discount function) is treated as an index of impulsivity. In order to provide a better analog of human impulsivity, this model needs to be expanded to include the converse of impulsivity - self-control. Through mechanisms such as committing to long range interests before the onset of temptation, or through bundling individual choices into classes of choices that are made at once, human decision-making can often look far less myopic than single trial experiments predict. For people, impulsive behavior may be more often the result of the breakdown of self-control mechanisms than of steep discount functions. Existing animal models of self-control are discussed, and future directions are suggested for psychopharmacological research.

Pigeons chose, in a two-key discrete-trial procedure, between 2-and 4-sec access to grain, with the larger amount always presented 4 sec later than the smaller. As the delay between the choice and the availability of the smaller reinforcement was varied from .01 to 12 sec, all subjects reversed preference from the small-early to the large-late reinforcement. The values of delay at which preference reversed were approximately consistent with the matching law as adapted for delayed reinforcement.

Research in intertemporal choice has been done in a variety of contexts, yet there is a remarkable consensus that future outcomes are discounted (or undervalued) relative to immediate outcomes. In this paper, we (a) review some of the key findings in the literature, (b) critically examine and articulate implicit assumptions, (c) distinguish between intertemporal effects arising due to time preference versus those due to changes in utility as a function of time, and (d) identify issues and questions that we believe serve as avenues for future research. Copyright Springer Science + Business Media, Inc. 2005

Behavioral science has long been puzzled by the experience of temptation, the resulting impulsiveness, and the variably successful control of this impulsiveness. In conventional theories, a governing faculty like the ego evaluates future choices consistently over time, discounting their value for delay exponentially, that is, by a constant rate; impulses arise when this ego is confronted by a conditioned appetite. Breakdown of Will (Ainslie 2001) presents evidence that contradicts this model. Both people and nonhuman animals spontaneously discount the value of expected events in a curve where value is divided approximately by expected delay, a hyperbolic form that is more bowed than the rational, exponential curve. With hyperbolic discounting, options that pay off quickly will be temporarily preferred to richer but slower-paying alternatives, a phenomenon that, over periods from minutes to days, can account for impulsive behaviors, and over periods of fractional seconds can account for involuntary behaviors. Contradictory reward-getting processes can in effect bargain with each other, and stable preferences can be established by the perception of recurrent choices as test cases (precedents) in recurrent intertemporal prisoner's dilemmas. The resulting motivational pattern resembles traditional descriptions of the will, as well as of compulsive phenomena that can now be seen as side-effects of will: over-concern with precedent, intractable but circumscribed failures of self-control, a motivated ("dynamic") unconscious, and an inability to exploit emotional rewards. Hyperbolic curves also suggest a means of reducing classical conditioning to motivated choice, the last necessary step for modeling many involuntary processes like emotion and appetite as reward-seeking behaviors; such modeling, in turn, provides a rationale for empathic reward and the "construction" of reality.