Feature

Looking out for ourselves from moment to moment is hard enough. No surprise, then, that when it comes to making choices for the people we will be a month, a year or a decade from now, even the best of us can falter. As Harvard psychologist Daniel Gilbert, PhD, puts it: "We're babysitters or caretakers or providers for our future selves, but we don't always do such a great job of it."

Now, a new study by New York University psychologists Michael Sagristano and Yaacov Trope, PhD, and Tel Aviv University psychologist Nira Liberman, PhD, suggests that when deciding on future courses of action, we tend to overemphasize abstract, high-level goals and ignore the concrete, low-level steps needed to reach them.

As a result, our future obligations can turn out to be far riskier, more difficult and more time-consuming than we had imagined.

The study, published in the Journal of Experimental Psychology (Vol. 131, No. 3), used a gambling task to investigate the importance of risks and rewards for decisions about the near and far future. It found that individuals who expect to gamble in the near future prefer safe bets with small payoffs, but those who expect to gamble after a delay of several months prefer much riskier bets with high payoffs.

Trope and Liberman's previous research has shown that whether we construe a future action in terms of its abstract or concrete features depends on its distance in time--the farther away it is, the more we think about it in the abstract--and that construal can have a dramatic impact on the decisions we make. This study extends those results by showing how our far-future obsession with abstract goals can cause us to focus on the desirability of an outcome while ignoring its feasibility, even when the outcome is entirely random.

"What these researchers are showing us is that the exact temporal distance of the future about which you are thinking makes an important difference," says Gilbert. "They're offering a very neat psychological explanation for a very robust economic phenomenon."

Construal level theory

Sagristano, Trope and Liberman are not the first researchers to notice that decisions can be affected by timing. The phenomenon can be seen in decisions about everything from what to eat for dinner to what school to attend, what person to marry or what career to pursue. George Loewenstein, PhD, a professor of psychology and economics at Carnegie Mellon University, gives the example of an academic who decides months in advance to attend a conference in Europe: "The higher-level idea of going to Europe is really exciting--you know, international travel--but then a week before the trip you start thinking about jet lag, about preparing the talk, how to cover your classes, getting your passport," he says.

The shift from "international travel" to "getting your passport" is typical of the difference between thinking about the distant future and thinking about the near future, say Trope and Liberman. The former is abstract, high-level and superordinate; the latter is concrete, low-level and subordinate.

"In the distant future," explains Trope, "we think about how much the outcome is attractive to us. In the near future, we think, 'Is it feasible?'"

Differences in construal level may be able to explain why the attractiveness of a trip to Europe declines as it grows nearer in time. According to Trope and Liberman's "construal level theory" (CLT), an activity will tend to look less attractive the closer it is when, as in this case, the concrete details are less pleasant than the abstract goals.

Odds now, money later

Construal level is not the only possible explanation, however. Over the years, psychologists and economists have devised a number of explanations for why humans adjust the value of future outcomes. Perhaps the academic registering for a conference in Paris expects to improve his ability to juggle multiple obligations by the time of the conference. Perhaps he misjudges his future emotional reaction to the trip, an explanation favored by Loewenstein and supported by some of Gilbert's research. Or perhaps he is simply being optimistic.

If these explanations are correct, construal level might not be as important as Trope and Liberman believe. So, they and Sagristano, a graduate student at NYU, decided to study gambling, an activity in which outcomes are uncontrollable, emotional reactions should actually work against their predictions, and optimism is--or should be--irrelevant.

In the study, subjects were told either that they would be gambling immediately or that they would be gambling after a delay of several months. They were then asked to choose among bets with equal expected values but varying levels of risk and reward: a bet with a $4 prize and a 50 percent chance of winning, for instance, versus a bet with a $8 prize and a 25 percent chance of winning.

The researchers found that subjects in the delayed group not only preferred risky gambles to safe gambles, but were also willing to pay significantly more for the chance to play them. The immediate group, in contrast, favored safe gambles. As predicted by CLT, when asked in a follow-up experiment to justify their decisions, the delayed group said it concentrated on payoffs (i.e., the desirability of the outcome), while the immediate group said it focused on probabilities (i.e., the feasibility of the outcome).

The results argue against both emotion-based theories, which predict that subjects will focus on payoffs in the near future, and experience- or optimism-based theories, which apply only if subjects believe that their own skills or motivations can influence the outcome of their decisions.

Implications

The results of the study suggest that construal level can be used to understand behaviors as disparate as buying a TV, getting engaged and deciding to gamble. In a review of CLT, in press for Psychological Review, Trope and Liberman propose extending their theory to include other kinds of psychological distance--such as the distance between self and other, between social ingroup and outgroup, and between a person and each of his or her social roles. A working parent, for example, might distract herself from the everyday office routine by thinking about the abstract joys of raising a child; later that evening, the same person, now in a new social role, might think about her career goals while changing a diaper or cleaning up after dinner.

Those differences in construal could affect decision-making just as differences in temporal distance do. However, while many researchers believe that CLT is an important contribution to the psychology of decision-making, some doubt whether it can be applied as broadly as Trope and Liberman suggest.

CLT is "a very elegant piece of research," says Loewenstein, "but in some cases I think the authors push it too far by arguing that it fully explains phenomena that have alternative, or sometimes multiple, determinants."

One question that remains unresolved is why, if Trope and Liberman's theory is correct, the bias exists in the first place. According to Trope--who cautions that any explanations at this point must be speculative--adjusting our decision-making strategies to suit the kinds of information we usually have available may just be a habit we find hard to kick.

"Usually, when we consider the distant future," he says, "we have information, or we trust information, that pertains to global major attributes of the [event], so we make overall global decisions." As the event approaches, however, we obtain low-level information that can be used to make specific plans for achieving our goals. In the gambling task, subjects had information about both high- and low-level aspects of the task at the outset, but the habit of focusing on superordinate goals in the distant future may have been too strongly ingrained to overcome.

The ideal decision-maker, says Trope, would take into account both abstract and concrete aspects of future courses of action. Focusing only on the concrete aspects of a future action can leave you obligated to do something that is easy but unsatisfying; focusing only on abstract aspects can leave you overcommitted and underequipped. "You need both the big picture and the details, the forest and the trees," he concludes.