Our capacity for intuitive decision-making evolved in a world of high stakes but relatively simple decisions. Feast or forage; fight or flight. The gains of recent technological and financial innovation have been great, but they come at the cost of increased complexity. According to the tenets of rational-choice economics, this can be only a positive development. These gains allow for the satisfaction of a maximum number of individual preferences. Anyone who has been overwhelmed by the choices they face at the supermarket or the bank, however, might suspect that the revealed-preference account is missing something.
Behavioral research has long documented evidence that we take shortcuts–known as heuristics–to simplify complex decisions. Instead of a rigorous, algorithmic analysis of a problem, we often break it down or focus on individual sub-elements to more easily compare across disparate alternatives. The exact nature of this process, however, is difficult to capture with external, behavioral methods. Economics traditionally assumes away the existence of a simplification process, but understanding how it works in the brain could prove beneficial, especially with the recent focus on choice architecture.
A paper in the current issue of Neuron by Vinod Venkatraman in Scott Huettel’s lab at Duke offers the first neuroimaging evidence that dissociates the brain processes responsible for heuristic simplification and economic choice. The researchers presented subjects with a series of complicated gambles. Each trial could have five different outcomes: two positive, two negative, and one neutral. The outcomes varied in magnitude and probability of reward across trials. After the subjects saw the possible results for each round, they got to choose between altering one of two potential outcomes.
The authors identified three possible behavioral strategies. On each round, the subjects could choose the option that either maximized potential gains, minimized potential losses, or merely maximized the probability of achieving a positive result. Rational choice economics implicitly predicts that an agent would analyze each option to determine which offered maximum utility, but, unsurprisingly, this is not what the behavioral evidence indicated. Instead, many subjects simplified the problem by consistently choosing to maximize only the chance of winning but not the magnitude of the expected outcome.
The fMRI results revealed several brain areas that were strongly related to individual choice behavior on this experiment. A gain-maximizing strategy was associated with increased activation in the ventromedial prefrontal cortex (vmPFC), while a loss-minimizing strategy increased blood flow to the anterior insula (aINS). The simplification strategy, however, increased activation in the working memory circuit between dorsolateral prefrontal cortex (dlPFC) and the parietal cortex (PPC). Whether a subject engaged in each of these strategies was strongly predicted by the functional relationship between the corresponding brain region and a fourth area, the dorsomedial prefrontal cortex (dmPFC).
The second relevant result indicated a strong relationship between a subject’s sensitivity to reward and his or her tendency to engage in probability maximization. When subjects were exposed to the results of their choices, they exhibited activity in the ventral striatum, a region of the brain dedicated to the processing of reward information. The strength of this response in individual subjects predicted whether they would take the shortcut.
One of the more interesting and counterintuitive findings in this paper is the association of the executive dlPFC-parietal system with a heuristic strategy. In many other cases, this circuit is thought to be responsible for algorithmic processing of potential outcomes. This result underscores an important lesson, however: we are not homo economicus. It is adaptive to simplify problems. It saves both metabolic and computational resources and, as Barry Schwartz has shown, it can make us happier.
The paper is a worthy example of how neuroeconomics can tell us more about how we make decisions than standard or even behavioral economics research. Our choices are not the result of some black-box process; they are the product of a functional system, and each component of that system plays an important part. Learning more about the the roles and relationships of those components will help us understand better where our decisions come from and, ultimately, how we might improve them.
Venkatraman, V., Payne, J., Bettman, J., Luce, M., & Huettel, S. (2009). Separate Neural Mechanisms Underlie Choices and Strategic Preferences in Risky Decision Making Neuron, 62 (4), 593-602 DOI: 10.1016/j.neuron.2009.04.007
[...] had an article (which I found via the excellent Nudge blog) on the Venkatraman and Huettel study I wrote about Wednesday, which is an interesting read. The piece focuses on the second result of the [...]