Willpower

tempting distractions. Again and again, the decision making took a toll on the students. Compared with the nondeciders, who’d spent just as much time evaluating the same kind of information without making choices, the deciders gave up sooner on the puzzles. Instead of using their time to practice for the math test, they goofed off by reading magazines and playing video games.
    As the ultimate real-world test of their theory, researchers went into that great modern arena of decision making: the mall. Shoppers in a suburban mall were interviewed about their experiences in the stores that day and then asked to solve some simple arithmetic problems. The researchers politely asked them to do as many as possible but said they could quit at any time. Sure enough, the shoppers who’d already made the most decisions in the stores gave up the quickest on the math problems. When you shop till you drop, your willpower drops, too. On a practical level, the experiment demonstrated the perils of marathon shopping. On a theoretical level, the results of all these experiments raised a new question: What kinds of decisions deplete the most willpower? Which choices are the hardest?

Crossing the Rubicon
    Psychologists distinguish two main types of mental processes, automatic and controlled. Automatic processes, like multiplying 4 times 7, can be done without exertion. If someone says “4 times 7,” 28 probably pops into your head whether you want it to or not—that’s why the process is called automatic. In contrast, computing 26 times 30 requires mental effort as you go through the steps of multiplying to come up with 780. Difficult mathematical calculations, like other logical reasoning, require willpower as you follow a set of systematic rules to get from one set of information to something new. You often go through steps like these in making a decision, through a process that psychologists call the Rubicon model of action phases, in honor of the river that separated Italy from the Roman province of Gaul. When Caesar reached it, he knew that a general returning to Rome was forbidden to bring his legions across the Rubicon. He realized that crossing it with his army would start a civil war. Waiting on the Gaul side of the river, he was in the “predecisional phase” as he contemplated his goals and possibilities along with the potential costs and benefits. Then he stopped calculating and crossed the Rubicon, reaching the “postdecisional phase,” which Caesar defined much more felicitously: “The die is cast.”
    The whole process could deplete anyone’s willpower, but which part is most fatiguing? Could the depletion be due mainly to all the calculations before the decision? By this point, Twenge and several other researchers had been depleted by this long-running project, but the reviewers who decided whether the work could be published in the field’s top journal wanted more answers. Kathleen Vohs, a veteran “closer” who knew how to bring embattled projects to final success, took over and masterminded the project through its final stages. She designed an experiment using the self-service sales site of the Dell computer company. At dell.com , shoppers could research and configure their own customized computer by choosing the size of the hard drive, the type of screen, and a series of other features. In the experiment, participants went through some of the same processes as Dell’s shoppers (except that nobody bought a computer at the end).
    By random assignment, each participant in the study was given one of three tasks. Some were told to look at several features relevant to a computer but not make a decision. They were instructed to think about the options and prices and to form preferences and opinions, but not to make a definite selection. The purpose of this condition was to duplicate the predecision thinking without the actual deciding.
    Another group was handed a list of selections and told to configure the computer. They had to go through the laborious, step-by-step process of locating the specified features among the arrays of options and then clicking on the right ones. The purpose was to duplicate everything that happens in the postdecisional phase, when the choice is implemented. The third group had to choose which features they wanted on their customized computers. They didn’t simply ponder options or implement others’ choices. They had to cast the die, and that turned out to be the most fatiguing