Social influences are powerful determinants of food intake. Whereas some people are willing to acknowledge social influences on their food intake, others seem to actively deny being influenced by social cues. Across three samples (total n = 835), we examined factors that prior theory and research suggest might predict people's willingness to acknowledge social influences on their food intake. These included conformity, self-monitoring, sociotropy, self-esteem, empathy, and the Big Five personality traits. Conformity, self-monitoring, and sociotropy were the most consistent predictors of acknowledgement of social influences on food intake, and conscientiousness was also related to acknowledgement of social influences. Furthermore, those effects were mediated by the extent to which people believe that eating in response to social cues is appropriate. These findings suggest that people who are more concerned with, and attuned to, the social world are more willing to acknowledge being influenced by social factors. Importantly, individuals who are less willing to acknowledge social influences on their food intake may not actually be any less influenced by social cues. Failing to acknowledge social influences on food intake could have implications for people's ability to regulate their eating appropriately and also for their self-evaluations.
Social context is a powerful determinant of behavior. One domain in which social factors are particularly potent is in influencing people's food intake. For example, people adjust their food intake to model that of their eating companions, eating little when their companions eat only a little, and eating more when their companions eat more (Herman et al., 2003 and Vartanian et al., 2015). This modeling effect is one of the most robust influences on food intake. Modeling has been demonstrated in a number of correlational and experimental studies, and is observed under a wide variety of conditions. For example, modeling occurs with snack foods (Robinson, Tobias, Shaw, Freeman, & Higgs, 2011) and during meals (Hermans, Larsen, Herman, & Engels, 2012), among children (Salvy, Vartanian, Coelho, Jarrin, & Pliner, 2008), and even when participants have been food-deprived for up to 24 h (Goldman, Herman, & Polivy, 1991). Modeling persists even when the model is not physically present and participants are exposed instead to a “remote confederate” list indicating the amount of food eaten by supposed prior participants (Roth, Herman, Polivy, & Pliner, 2001).
Despite a substantial body of research demonstrating that social factors can strongly influence food intake, people typically fail to acknowledge these influences when explaining their eating behavior (Vartanian, Herman, & Wansink, 2008). For example, a series of experiments by Vartanian, Sokol, Herman, and Polivy (2013a) showed that, although participants' food intake is powerfully influenced by the behavior of the experimental confederate, participants are much more likely to use factors such as taste and hunger than the behavior of the confederate to explain their food intake. Not only did participants in those studies underreport the influence of the social model's behavior, but they were also inaccurate in their overall explanations for their food intake. That is, participants' reports of the extent to which they were influenced by the various cues (hunger, taste, or social factors) were unrelated to the extent to which their behavior was actually influenced by those particular cues. A similar failure to acknowledge social influences on behavior has also been observed in domains other than eating, such as energy-conservation behavior (Nolan, Schultz, Cialdini, Goldstein, & Griskevicius, 2008).
Are people unaware that social influences are operating, or are they simply reluctant to acknowledge those influences on their behavior? Considerable social psychological research indicates that people's explanations for their behavior may be motivationally driven. Research on the actor–observer effect and on the third-person effect has shown that people make different attributions for their own behavior than for the behavior of others. For example, people tend to acknowledge external influences (such as advertising) on the behavior of others, but typically deny these influences on their own behavior (e.g., Douglas & Sutton, 2004). Other studies have found that people can accurately predict other people's behavior but are prone to making inaccurate predictions of their own behavior (e.g., Epley & Dunning, 2000).
There is evidence that the failure to acknowledge social influences on one's own food intake might be motivationally driven. Spanos, Vartanian, Herman, and Polivy (2014) found that participants were able to accurately identify social influences on the food intake of other people, but that there were individual differences in the extent to which participants accurately identified social influences on their own food intake. Spanos et al. (2014) created the Social Eating Scale to assess people's self-reported tendency to eat in response to social cues in their everyday lives. During a social-eating situation, participants who scored high on this measure (high acknowledgers) were relatively accurate in reporting on the extent to which their food intake was influenced by the behavior of their eating companion. In contrast, participants who scored low on this measure (low acknowledgers) appeared to actively deny being influenced by their eating companion: the more their food intake closely mimicked that of their eating companion, the less likely they were to acknowledge being influenced by that companion. These findings suggest that some people are willing to acknowledge social influences on their food intake, whereas others appear to actively deny these influences.
