"Food addiction is real". The effects of exposure to this message on self-diagnosed food addiction and eating behaviour.

Food addiction is widely discussed in popular media in many Westernised societies. However, a potential concern is that endorsement of the food addiction model may cause people to perceive a lack of personal control over eating which could promote unhealthy dietary behaviours. To address this possibility, the current study investigated whether exposure to food addiction messages would, firstly, increase the number of participants who self-diagnosed as food addicts and, secondly, increase intake of indulgent foods. In a between-subjects design, participants (N = 60) read an article which either claimed that food addiction is real ("Real" condition) or that food addiction is a myth ("Myth" condition). Intake of indulgent and non-indulgent foods was then assessed in a disguised taste test and participants also completed a measure of self-diagnosed food addiction. A significantly higher proportion of participants in the Real condition self-diagnosed as food addicts relative to participants in the Myth condition (57% and 27% of participants, respectively; p = .018). Variability in intake, but not mean intake, of indulgent food was higher in the Real condition than in the Myth condition. These findings suggest that endorsement of the concept of food addiction may encourage people to self-diagnose as food addicts and thus explain their eating behaviour in terms of addiction (an external attribution). The extent to which self-diagnosis of food addiction influences actual food intake and how this might vary with individual differences and eating context remains to be determined.

Reprint of: Eating like you are overweight: the effect of overweight models on food intake in a remote confederate study.

There is consistent evidence that people model the eating behaviour of others. The extent to which people model the amount of food consumed by other people of different weight statuses has received less attention. Here we tested the effect on food consumption of exposing female participants to information about the food consumption of either normal/healthy weight or overweight individuals. Eighty female participants took part in a between-subjects experiment, in which we used a remote-confederate design and manipulated whether participants saw intake information about normal/healthy weight or overweight previous participants (remote confederates). Regardless of the weight-status of the remote confederates, participants ate more food when they believed that previous participants had eaten a large amount of food, in comparison with when they believed previous participants had eaten a smaller amount of food. These findings indicate that women may model the food intake of other women, even when they believe they are of a different weight status to themselves.

Eating like you are overweight: the effect of overweight models on food intake in a remote confederate study.

There is consistent evidence that people model the eating behaviour of others. The extent to which people model the amount of food consumed by other people of different weight statuses has received less attention. Here we tested the effect on food consumption of exposing female participants to information about the food consumption of either normal/healthy weight or overweight individuals. Eighty female participants took part in a between-subjects experiment, in which we used a remote-confederate design and manipulated whether participants saw intake information about normal/healthy weight or overweight previous participants (remote confederates). Regardless of the weight-status of the remote confederates, participants ate more food when they believed that previous participants had eaten a large amount of food, in comparison with when they believed previous participants had eaten a smaller amount of food. These findings indicate that women may model the food intake of other women, even when they believe they are of a different weight status to themselves.

Influenced but unaware: social influence on alcohol drinking among social acquaintances.

BACKGROUND: Drinking partners may be influenced by each other's alcohol consumption. However, these effects have only been shown in artificially created social pairings and typically among same-sex young adults. Here, we test whether similarly strong influence effects occur among "real" pairs of social acquaintances (friends and partners) and whether people are aware of this influence on their alcohol consumption. METHODS: Forty-six pairs of social acquaintances aged between 19 and 60 years old participated in a between-subjects experiment, in a semi-naturalistic bar laboratory setting. One member of each pair (the confederate) was randomly selected and asked to consume only alcoholic (alcohol condition) or soft drinks (nonalcohol condition), while the pair completed a game together in a bar setting. The other participant (naïve) was unaware of these drinking instructions. Postconsumption, we measured the extent to which naïve participants believed that their partner had influenced their own drinking behavior. RESULTS: A large effect of condition on alcohol consumption was observed, χ² (2) = 15.8, p < 0.001, Cramer's V = 0.59, whereby the number of alcoholic drinks selected by naïve participants in the alcohol confederate condition was significantly greater than in the nonalcohol confederate condition. The majority of naïve participants (81%) also tended to be unaware that their partner had influenced their alcohol consumption. CONCLUSIONS: Social acquaintances are influenced by each other's alcohol consumption and may not be aware of this influence on their behavior.