Linking contemporary models of self-regulation to recent research on automatic attitudes, the present study investigated the impact of automatic candy attitudes, dietary restraint standards, and self-regulation resources on eating behavior. Participants were assigned to either an emotion suppression task (low self-regulation resources) or an emotion flow task (high self-regulation resources), and were then given an opportunity to taste candies. When self-regulation resources were high, candy consumption was uniquely related to dietary restraint standards (but not automatic candy attitudes). In contrast, when self-regulation resources were low, candy consumption was primarily predicted by automatic candy attitudes, with dietary restraint standards showing a tendency for counterintentional effects. These results indicate that the behavioral impact of automatic attitudes and personal standards depends on available control resources. Implications for research on automatic attitudes and self-regulation are discussed.
People are often tempted by their impulses, urges, and cravings. Because giving way to one’s immediate hedonic impulses is not always possible or advisable in the light of social or personal constraints, human beings acquired the capacity for self-control or self-regulation in a historical process of civilization ( Elias, 1939/2000 and Freud, 1930/1961). This capacity can be defined as the “ability to override or change one’s inner responses, as well as to interrupt undesired behavioral tendencies and refrain from acting on them” ( Tangney, Baumeister, & Boone, 2004, p. 275).
Obviously, not all impulses require self-control, as acting in line with one’s impulses often has no negative consequences (e.g., drinking a cup of water when being thirsty). However, in many circumstances the implications of a certain impulse (e.g., the desire to eat a candy bar) are at odds with personal goals (e.g., “I want to lose weight.”). In such cases, the resulting conflict between impulse and self-control can be described as a tug-of-war in which the stronger competitor wins (Baumeister et al., 1994, Mischel, 1996 and Muraven and Baumeister, 2000). For example, in their model of ego depletion, Baumeister and colleagues argued that the capacity for self-control resembles a muscle that may become “tired” over the course of using it ( Baumeister et al., 1998 and Muraven et al., 1998). Thus, engaging in self-regulation often depletes people’s subsequent ability to control their behavior. Consistent with this assumption, Vohs and Heatherton (2000, Study 3) showed that emotion suppression undermined participants’ success in restraining their eating behavior in a subsequent ice-cream tasting task. In a similar vein, Muraven, Collins, and Neinhaus (2002) found a decrease in the control of alcohol consumption when participants had to suppress thoughts of a white bear before. Finally, in the domain of prejudice, Richeson and colleagues demonstrated that controlling one’s behavior in interracial interactions led to impaired performance in a subsequent task that required a high level of executive control ( Richeson et al., 2003 and Richeson and Shelton, 2003).
So far, research on self-regulation has primarily focused on the control aspect of human behavior. However, the determinants of impulsive tendencies are much less clear. In the present article, we make a suggestion to fill this gap by linking the proposed conflict between self-control and impulse to recent research on automatic attitudes (for a review, see Petty, Fazio, & Briñol, in press). Specifically, we argue that impulsive action tendencies can be linked to and often are the consequence of automatically activated evaluations. More precisely, we argue that impulsive action tendencies to approach or avoid a particular stimulus are the result of automatically activated evaluations of this stimulus (Strack & Deutsch, 2004). As such, ego depletion should moderate not only the impact of self-control on human behavior. Rather, the impact of ego depletion should be twofold, such that it determines whether behavior is determined either by automatic attitudes or by personal standards. More precisely, we argue that behavior should be predominantly influenced by automatic attitudes when self-regulation resources are low, but by personal standards when self-regulation resources are high.
