Keeping and sharing secrets is an integral part of our social lives, yet most research has taken an individual-centered approach, focusing on the cognitive experiences of withholding information without fully accounting for the relationship context in which those secrets are kept. Furthermore, the discreet and often invisible nature of secrets poses many methodological obstacles with regard to understanding the naturalistic implications of secrets in everyday life. Thus, the focus of this dissertation is twofold. First, I establish secrecy as an interpersonal phenomenon by (a) providing a theoretical framework for the sociality of secrets and (b) presenting empirical evidence regarding several social determinants and implications of secrecy. Second, I present novel methodological approaches to studying secrets by (a) establishing and validating a new measure of secrecy burden and (b) employing intensive longitudinal methods in one of the first studies to examine the implicit effects of secrets in everyday interactions.

To summarize, Chapter 1 presents a theoretical overview of the mechanisms by which keeping and sharing secrets may uniquely influence and be influenced by interpersonal relationships in distinct ways from related constructs of self-disclosure and privacy. In Chapter 2, I present empirical data examining the immediate and cumulative effects of repeatedly sharing one’s own (personal) and other people’s (secondhand) secrets over 10 weeks in an undergraduate sample (N = 126). Results show that sharing and receiving more personal secrets was robustly associated with increased closeness and social utility, whereas the relational benefits of receiving secondhand secrets only emerged at the aggregate level. Chapter 3 then shifts the focus to examining the well-being implications of keeping secrets by developing and validating a new measure of secrecy burden. Across three studies (N = 810), I identify a four-factor model of secrecy burden (Daily Personal Impact, Relationship Impact, Pressure to Reveal, and Anticipated Consequences) and show that each burden factor is differentially associated with well-being over two to three weeks. Lastly, Chapter 4 uniquely assesses the implicit effects of secrets without prompting additional salience to those secrets by comparing interactions with secret targets vs. non-targets in a 10-day experience sampling study (N = 114) in which participants were naïve to the study’s focus on secrecy. Results show that keeping a greater number of secrets from an interaction partner was associated with higher burden in daily interactions (i.e., more stress, distractedness, distance, and inauthenticity) as well as lower relationship satisfaction and authenticity at the end of the study. Altogether, this research underscores the importance of studying secrecy through an interpersonal lens and accounting for the specific relationships in which secrets are kept and shared.

Sharing secrets is common in everyday life, yet we know relatively little about how secret-sharing behaviors influence evaluations of those who share secrets. The present studies examined how we evaluate others based on the type of secret (personal vs. secondhand), who shares the secret (close friend vs. acquaintance), and its sharing breadth (exclusive vs. nonexclusive). Participants read brief vignettes about a secret being shared and evaluated the secret-sharer on trustworthiness, closeness with the recipient, and social utility. In Study 1, sharing a personal secret led to higher ratings of trustworthiness and closeness relative to sharing a secondhand secret, regardless of who shared the secret, and sharing a secondhand secret led to higher ratings of social utility. In Study 2, we found a consistent interaction between secret type and sharing breadth. Compared to sharing a secret exclusively with one person, sharing a personal secret nonexclusively decreased closeness ratings, and sharing a secondhand secret nonexclusively decreased trustworthiness ratings. Lastly, social utility was highest for sharing a personal secret nonexclusively. These findings suggest that both the type of secret and how many people it is shared with have important implications for intimacy and trust in close relationships, as well as demonstrating social value.

Although positive emotions have been relatively understudied (in comparison to negative emotions), the existing research suggests that positive emotions can have a beneficial effect on health outcomes and health-relevant biomarkers. It is likely that positive emotions benefit health through buffering individuals from the effects of stress; however, more research is needed to uncover the psychosocial processes through which positive emotions exert this stress-buffering effect. Two studies tested the hypothesis that positive emotions can impact coping, rumination, and perceptions of resources, three processes with important implications for health and well-being. In Study 1, participants were assigned to watch film clips designed to induce the emotions of love or amusement (or to watch a neutral film clip), after which they completed measures of coping and rumination (including an implicit measure of coping). In Study 2, participants were assigned to write about experiences of gratitude (or neutral topics), after which they completed measures of coping, social resources, and personal resources. The present studies failed to find support for the hypothesis that positive emotions impact coping, rumination, and perceived resources. The findings and directions for future research are discussed.

Research on approach and avoidance goal adoption has largely focused on dispositional factors, despite theoretical models accounting for situational influences. Recent work has found that the goal-relevant resources that individuals possess influence approach and avoidance goal adoption, such those with greater resources adopt more approach goals. However, goal-relevant resources have only been examined in the achievement domain. We aim to expand this recent work into the social domain through two studies. Study 1 experimentally manipulated the resources of a hypothetical target as a first step in examining whether or not the social resources of social competence and social support would influence approach and avoidance social and achievement goals. In support of our hypotheses, higher social resources led to ratings of more approach social and achievement goals. Study 2 manipulated participants’ own sense of social competence and then examined social outcome expectancy as a mediator between social competence and social goal adoption. Participants in the high social competence condition had higher social outcome expectancy, which predicted more approach social goals. However, participants did not differ by condition on achievement goals or achievement outcome expectancy. These results lend further support for goal-relevant resources as a situational influence on approach and avoidance goal adoption.

People often modify their emotional responses through the process of emotion regulation (Gross, 1998) for hedonic or instrumental goals (Tamir, 2016). However, these emotion regulation processes can differ based on: (1) whether the emotions are based on one’s group membership (i.e., group-based) or not (i.e., individual-based), and (2) whether another person is involved in this emotion regulation process (i.e., interpersonal) or not (i.e., intrapersonal). Less research has examined how group-based emotions are regulated with others, particularly how group members influence others’ emotions. Over three studies, this dissertation tested a model in which group members attempt to regulate the emotions of ingroup and outgroup members to achieve group goals (Hypothesis 1). In this model, emotions are perceived as instrumental to group goals (or not) based on whether the target is an ingroup or outgroup member (Hypothesis 2). Group members attempt to upregulate instrumental emotions and downregulate non-instrumental emotions for ingroup members, and downregulate instrumental emotions and upregulate non-instrumental emotions for outgroup members (Hypothesis 3). These regulation attempts then result in greater upregulated emotions in ingroup rather than outgroup members (Hypothesis 4). Study 1 experimentally manipulated the group membership of the target to examine whether group members regulate the emotions of ingroup and outgroup members that are instrumental in achieving ingroup goals. The results showed that regardless of the valence of the emotion (i.e., the emotions of happiness and anger), group members attempted to upregulate instrumental emotions and downregulate non-instrumental emotions for ingroup members, and downregulate instrumental emotions and upregulate non-instrumental emotions for outgroup members. Study 2 used archival data to test whether more liberal or more conservative media sources try to influence the emotions of their readers (i.e., ingroup members) about politically group-relevant (i.e., COVID-19 and Ebola) but not politically group-irrelevant events (i.e., celebrity deaths). Results showed that the political identity of media sources influenced how they tried to regulate the instrumental emotions of their readers. More liberal media sources were more likely than more conservative media sources to try to upregulate anxiety (but not other emotions) due to its instrumentality for the politically group-relevant event of COVID-19. Study 3 extended Study 2 to determine whether the regulator’s emotion regulation attempts resulted in the experience of greater upregulated anxiety in ingroup rather than outgroup members (that is, whether ingroup members were more affected by ingroup regulators’ attempts to change their emotions than were outgroup members). However, results indicated that the emotions that the regulators were trying to influence were experienced by both ingroup and outgroup members. Together, these studies suggest that group members attempt to regulate the emotions of ingroup and outgroup members based on the instrumentality of the emotion to group goals, influencing the target's emotions.