Another Big Psychology Theory Fails to Replicate
The bakery at most grocery stores is a minefield. Cakes to the left, muffins to the right, pastries dead ahead, and cookies... cookies everywhere. If you escape without making a purchase, congratulations, you have tenacious self-control. Or you were just lucky. But though your wallet and waistline won't take a hit, according to a leading psychological theory, your willpower will.
Originally put forth back in 1998 by Florida State psychologist Roy Baumeister, the notion of ego depletion states that self-control is a limited resource. Like a muscle, it can fatigue with use, and needs time to recharge. According to the theory, saying "no" to sweets in the grocery store will leave you temporarily vulnerable to subsequent temptations.
In the sixteen years since its inception, ego depletion has been tested and validated in a variety of situations. Psychologists emphasize its role in many arenas, such as dieting, athletics, and consumer behavior. Some even propose that willpower can be trained and strengthened via repeated use, again, just like a muscle.
But critics aren't so sure. They note that much of the research has been done on young, WEIRD (Western, Educated, and from Industrialized, Rich, and Democratic countries) college students, and thus may not carry over to the general population. They also suggest that the effect could benefit from publication bias, the tendency to only publish flashy or positive results. It is in this light that a team of psychologists recently attempted to replicate the ego depletion effect using the two most frequently used measures of self-control in scientific research. Their results were just published to PLoS ONE.
Four groups of participants took part in the study: two diverse groups from the general population with an average age in the mid-forties, and two more groups of young adults with an average age around 20. Each group was assigned to one task, either a grip challenge in which subjects had to hold a grip machine at 70% of their maximum grip strength for as long as possible, or a Stroop Test, in which subjects viewed color words on a computer (like 'green', 'yellow,' or 'blue') that appeared one at a time in a mismatched font color (for example, 'red' may be shown in blue font) and have to press the key corresponding to the font color.
The subjects completed each of their respective tasks twice. In between the repetitions, some subjects engaged in an activity meant to diminish their self-control whilst others performed an easy control activity. The researchers then compared the scores of the subjects who performed the control activity with those who performed the experimental one.
The subjects who had performed the task meant to diminish self-control should have performed significantly worse on their assigned tasks compared to the control groups the second time around. But they did not.
"There was no evidence for significant depletion effects in any of these four studies," the researchers found.
The researchers don't believe their results are in error. The tasks they used were identical to those most frequently employed in past research, and they also used similar sample sizes. In fact, the researchers suggest their study design may be stronger, because, unlike many previous studies which found evidence for ego depletion, they used a control group and recruited subjects from a broader population.
Despite the failure to replicate, the researchers don't believe their study is enough to invalidate ego depletion altogether, just that the effect may be more limited than has previously been theorized.
Some social psychologists, like Harvard University's Jason Mitchell, have fired back at detractors, suggesting that they are impugning the integrity of their colleagues, overstating the problems of publication bias, and are likely producing "false negative" results.
(Images: AP, Nevit Dilmen)
Source: Xu X, Demos KE, Leahey TM, Hart CN, Trautvetter J, et al. (2014) Failure to Replicate Depletion of Self-Control. PLoS ONE 9(10): e109950. doi:10.1371/journal.pone.0109950