Poor diet is the leading cause of non-communicable disease globally [1,2]. It is responsible for approximately 11 million deaths and 255 million disability-adjusted life years (DALYs) annually, a number which is growing [1]. There is scientific consensus regarding the importance of diet in the prevention and treatment of chronic illness [1,2,3,4,5]. However, most interventions designed to change dietary behavior are of short duration and may be ineffective in achieving long-term change in individuals [6]. Therefore, innovative approaches to long-term dietary behavior change are urgently needed. Given the global nature of this issue, the social disparities that exist in diet quality [7,8,9,10], and the broad role of identity in health behaviors [11,12], it is especially important to identify strategies that can fully account for personal, social, and cultural factors.

Identity is a major construct in the fields of psychology and anthropology that may be relevant to eating behaviors. The most common identity theories when studying eating behavior stem are Identity Theory (IT) and Social Identity Theory (SIT). According to IT, identities are evolving and multifold [13,14,15]. For instance, an individual may be simultaneously a mother, a software developer, a vegetarian, and a recent immigrant, and each of these identities will have a range of internal motivations, values, and goals. IT aims to assess how an individual’s multiple identities interrelate and how they may vary in centrality and in salience by context [13,16,17,18,19,20]. For example, individuals may shift among components of their identity, thinking and acting somewhat differently depending on whether they are at work or with family. Identities may also shift and change in response to more profound changes in context, such as developing a major illness or immigrating to a new country.

Social Identity Theory (SIT), a second major psychological theory of identity, aims to understand identity as a social construct that develops from group membership and social affiliations [21]. SIT examines the role that identification with a group (in-grouping) plays in the reinforcement of behavior, as well as the anticipation, or fear, of judgement for failing to behave congruently with the group. SIT examines the motives and pressures of conformity, such as in-group stigma and out-group discrimination, and the cognitive dissonance of identity threat when individual members violate their own beliefs [22]. In both IT and SIT, behavior reinforces one’s sense of self—values, social roles, and normative expectations—and signals affinity to groups with whom an individual identifies.

In the realm of health behavior theory, there has been increasing interest in including identity [23,24,25], most notably addedto the Theory of Planned Behavior (TPB). Self-identity was first introduced in the TPB to help account for the salience of internal values as a predictor of intention [26,27]. As such, it is thought to help understand and predict differences in the intentions to perform and sustain behavior based on how important the behavior is to the way someone describes themselves [26,27,28,29]. Self-identity is one of the most common extensions of the TPB and has been shown in multiple meta-analyses to explain an additional 4–6% of the variance in intention and up to 9% when controlling for past behavior [28,30,31,32,33,34].

In anthropology, food is considered a central part of identity formation [35,36,37]. Using food, cultures demonstrate in-group affiliation, within-group hierarchies, and within- and between-group rituals that reinforce cultural identity [35,38]. How food is chosen, made, served, and eaten all serve to co-construct identities of multiple orders—self-identity, family identity, ethnic identity, national identity, religious identity, and others [39,40,41,42]. On an individual level, the relationship between food and identity is thought to be even more connected. Food choices signal active allegiances with social groups and reinforce norms, stereotypes, and beliefs; and they do so multiple times a day, throughout an individual’s lifespan [35,37,43]. However, the complex and independent effect of identity on eating behaviors has not been extensively summarized.

Changes in eating behaviors require individuals to make effort to change their habits, maintain the change, and perform continual decision-making relevant to that change. It therefore seems reasonable to suggest that identity processing and salience may help explain variations in both the initiation and maintenance of eating behaviors. However, there is scant literature that actively investigates how to capture various identities and how their salience shifts and evolves with respect to eating behaviors [25,35,44,45,46,47,48,49]. Additionally, the roles of identity, identity formation, contextual identity shifting, and identity change throughout the lifespan are also not a major domain of the health behavior theories that are most frequently used to guide changes in eating behaviors. Indeed, identity is only minimally discussed in the major textbooks widely used to train professionals in health behavior theory [50].

Identity is often explored simplistically, and only recently have there been studies that accounted for the influence of multi-fold identities, and the process of identity formation, that may conflict, shape, or augment identity performance. Studies with less comprehensive measures of identity tended to regard identity as a type of motivation or a characteristic of behavior and may confuse identity with other behavioral constructs, such as cultural norms, self-esteem, and habits. There is a need to increase the dimensionality and standardize the evaluation of identity in order to fully understand its explanatory power for eating behaviors and healthy eating behavior changes. Standardized and complex measures of identity would also help ensure a fuller understanding of how identity processes can be used to assist in the adoption of new eating behaviors and adherence to behavior changes.

Identity theories will likely be useful in improving operationalization. Most studies were informed by IT or SIT; however, only some investigated multiple identities [126,142] or how salience and priorities shift depending on context, life changes, self-determined goals, and other factors [15,17,21,40,48,77,86].

Self-identity is one of the most common extensions of the TPB and has been shown in multiple meta-analyses to independently explain an additional 4–6% of the variance in intention and up to 9% when controlling for past behavior [28,30,31,32,33,34]. Most of the studies reviewed relied on a modified version of the extension of TPB, itself the most commonly used theoretical approach to identity, and therefore most studies assumed that identity was an a priori construct of the self [29,83,126,135,136]. However, one meta-analysis discussed how its findings suggest a need to disentangle identity from the pre-formed aspects of behavior in TPB such as attitude and intention [28]. In a study based on TPB that had more comprehensive identity measurements, identity explained up to 28.5% of the variance in intention and 5.7% of the variance in the behavior itself [83]. This is meaningful, given that, through meta-analytic methods, TPB only explained 21.2% of variance in eating behaviors and 52.4% of intention [34,87]. Theory advancement papers call for the examination of complex, shifting, and evolving identities [25,35,44,45,46,47,48,49], and this review confirms the need for additional research in this area.

 

2.2. Role of Identity in Eating Behaviors

Individual components of identity, such as group affinity, motivation, salience, centrality, and importance, have demonstrated associational and explanatory value when considered independently, and our findings suggest the importance of considering more robust and complete measures, such as formation, multi-fold, and changing identities, in order to better understand the role of identity in eating behaviors. In particular, we found that eating behavior and identity have reciprocal and mutually reinforcing roles, and as such omitting the more complex facets of identity may inaccurately attribute directionality to these specific sub-components, including during experimental manipulations. Further research should investigate identity formation and change in order to better understand this reciprocal relationship. Our findings also suggest that future research might incorporate a wider theoretical perspective to allow for an understanding of more complex relationships between eating behavior and identity.