Life is uncertain, eat dessert first: Uncertainty causes controlled and unemotional eaters to consume more sweets

Abstract

Sometimes even dieters with the best self-control overindulge. Uncertain situations may undermine the self-control of even well-controlled eaters. This study was designed to test the hypothesis that uncertainty increases unhealthy snacking. Participants were either told that they would be giving a speech, that they would be listening to a speech, or that they would find out later whether they were to give a speech or not. Among participants who typically reported good control over their eating or scored low on emotional eating, participants who were uncertain about whether they would be giving a speech ate more candy than participants who expected to not have to give a speech and even those who expected to have to give a speech. Participants who reported poor control over their eating or scored high on emotional eating did not eat significantly more when uncertain. These findings suggest that, for people who are typically able to control their eating, uncertainty increases snacking more than certainty of a negative outcome.

Introduction

Even individuals who typically have good self-control may overindulge in certain circumstances. Uncertain situations can be a cue that self-control, including controlling what one eats, may not pay off in the long run. The current study is designed to test the hypothesis that uncertainty increases unhealthy food consumption.

Self-control is the process by which people pursue abstract or distal goals instead of more immediate goals and desires (Fujita, 2011). Self-control plays an important role in choosing to eat healthy foods, resisting the temptation to indulge in unhealthy foods, and in refraining from overeating (Baumeister, Bratslavsky, Muraven, & Tice, 1998; Crescioni et al., 2011; Vohs & Heatherton, 2000). The act of giving up current pleasures for future rewards is most likely to pay off in a relatively predictable environment. People who put money in the bank and eat healthily do so in part because they have some reasonable expectation that they can predict aspects of their financial situation and health in the future. Because uncertainty may reduce people's confidence that self-control will pay off, uncertainty may undermine people's likelihood of exercising self-control.

A situation can be described as uncertain if an individual lacks some important piece of information (Bar-Anan, Wilson, & Gilbert, 2009; Knight, 1921; Mittal & Griskevicius, 2014). Previous research has shown that uncertainty reduces people's likelihood of successfully executing self-control. Participants who were left uncertain about the timing of a reward were less likely to delay the reward than participants who were certain (Kidd, Palmeri, & Aslin, 2013; McGuire & Kable, 2012, 2013). In another study, participants who were given a vignette where they were asked to imagine being uncertain about what kind of pizza they would eat were more likely to choose an immediately rewarding but less healthy food (i.e., a brownie) over less desirable but healthier food options (i.e., fruit salad) than participants who were asked to imagine they'd get a specific pizza (Milkman, 2012). Participants who were asked to write about a time when they were uncertain completed fewer solvable anagrams than participants who were asked to write about a time when they were certain, suggesting that uncertainty impairs performance (Alquist et al., 2018). Healthy eating choices sometimes rely on one's ability to exercise self-control, which may be impaired under conditions of uncertainty.

Whether individuals' eating behavior is affected by a situation can differ based on their typical eating behavior. The Three-Factor Eating Questionnaire was developed to measure three different facets of eating behavior – uncontrolled eating, emotional eating, and cognitive restraint (de Lauzon et al., 2004). Uncontrolled eating is the inability to inhibit the tendency to overeat. Uncontrolled eating is associated with eating foods higher in fat (de Lauzon et al., 2004). Emotional eating is characterized by overeating in response to negative moods. Emotional eating is associated eating foods high in sugar. Uncontrolled and emotional eating are associated with higher body weight (Lindroos et al., 1997; Provencher, Drapeau, Tremblay, Després, & Lemieux, 2003). Cognitive restraint is the deliberate restriction of food consumption. Cognitive restraint does not necessarily imply success at this restraint, and research has shown that it is a not always a reliable indicator of food intake (Stice, Cooper, Schoeller, Tappe, & Lowe, 2007; Stice, Sysko, Roberto, & Allison, 2010) or body weight (higher body weight, Anglé et al., 2009; lower body weight, Boschi, Iorio, Margiotta, D'Orsi, & Falconi, 2001; de Lauzon et al., 2004; no relationship with body weight, Provencher et al., 2003). To the extent that these factors predict eating behaviors more generally, they may predict whether people's unhealthy snacking is affected by uncertainty.

There are a few possible ways the effect of uncertainty on food intake could be moderated by eating behaviors. It might be that participants who reported typically struggling to control their eating would become even more likely to eat when faced with uncertainty than participants who do not report struggling to control what they eat. In this case, the effect of uncertainty on eating would be strongest among individuals who are emotional or uncontrolled eaters. However, it was also possible that participants who reported typically being unable to control their eating would struggle to exercise self-control on the eating task across conditions (Ouwens, van Strien, & van der Staak, 2003; Van Strien, Cleven, & Schippers, 2000). In this case, the effect of uncertainty on self-control might only be observed among participants who did not report typically struggling to control their eating. If participants dispositionally struggle to control their eating, a situation that reduces their ability to exercise self-control may have no additional effects.

The present study was designed to test the hypothesis that uncertainty would be associated with greater consumption of candy than certainty. To manipulate uncertainty, participants were randomly assigned to one of three conditions. Participants in the certainty condition were told that they were giving a speech. Participants in the control condition were told that they were not giving a speech. Participants in the uncertainty condition were told that the experimenter misplaced the condition sheet and that they would be informed if they were giving or listening to the speech later. After the uncertainty manipulation, participants were asked to participate in an ostensibly unrelated taste testing study. Participants’ self-control was measured by how much candy they consumed during the taste test. Positive mood, negative mood, and perceived control were measured to assess alternative explanations for this effect.