238 OVERALL GIRME.LEMAY.AND HAMMOND security and satisfaction.Prior research has shown that partner andurin rela 1999).On the other there is al onship satis eed.self-report data indicate that hat the evels of guilt effor mpanied by other negative affective states such as ster et a 1994).Gre egativity and partner tran rth, d mood 1993 not be ea overcome rdless of how guilty the partner uilt may also for ively les srelationship rewards (support and s commitment and the relati As before.whether guilt has these potential benefits and costs is likely to depend on the of When gu ed that a key way anxious individuals nage relation test their ion.it is likely Thi e,provide highly an individ guilt will be inte any costs of the partner's guilt. enguilt is ind er parnergu might help sustain s of securit en when they do not intentic nally hur the iduals low in attachment a are not upied with thei n their partpe s guilt as evidence of their s ca nd guilt is indu feel isten of high le Is of partne and so we ected that fec would ead to ction for the partners of highly anx cted that d re Summary and Research Overview The entral goal of individuals high in attachment anxiety is to ls high partne closen s and acceptance goal that is served by intens ance of the and as to isolate the en al and beha n fo Benefits and Costs of Feeling Guilt ater hurt feeling Guilt is also likely to be ass ciated with a mix outcomes for the r who feels hould enhan s.Peopl who e more con guilt should provide anxious individual's evidence of their pan 9g v. 1995:R n in their relationshir duals and the ilt and beha ter guilt will be associated with main ing satisfying relatio the se quality. anxious individuals would experience more hurt feelings in thesesecurity and satisfaction. Prior research has shown that partner responsiveness during relationship conflict builds trust, commit￾ment, and satisfaction over time (Wieselquist, Rusbult, Foster, & Agnew, 1999). On the other hand, there is also reason to suspect that the partner’s guilt should undermine satisfaction. Guilt might repair specific transgressions, but high levels of guilt also signify the partner’s responsibility and fault for hurtful transgressions, and may direct attention to the existence of high levels of conflict, hurtful partner behavior, and negative emotions in both partners (Baumeister et al., 1994). Greater negativity and partner transgres￾sions have a deleterious impact on relationship quality (e.g., Kar￾ney & Bradbury, 1995; Kluwer & Johnson, 2007)— damage that may not be easily overcome regardless of how guilty the partner feels or what the partner does to make amends. Thus, greater partner guilt may also forecast more negative perceptions of the partner’s commitment and the relationship. Responses to a partner’s guilt may depend on individuals’ level of attachment anxiety and associated source of that partner’s guilt. We predicted that a key way anxious individuals manage relation￾ship threats is to purposively elicit guilt to coerce the partner’s care and test their partner’s commitment. Successfully inducing guilt in the partner should, therefore, provide highly anxious individuals desired evidence of their partner’s commitment, and these benefits may overshadow any costs of the partner’s guilt. Accordingly, greater partner guilt might help sustain feelings of security and relationship satisfaction for anxious individuals. In contrast, indi￾viduals low in attachment anxiety are not preoccupied with their partner’s acceptance, are unlikely to seek reassurance via guilt￾induction strategies, and, when guilt occurs, are less likely to focus on their partner’s guilt as evidence of their partner’s care. Hence, for low anxious individuals, the partner’s guilt may not have benefits. Instead, when not induced by low anxious individuals, the existence of high levels of partner guilt represents the partner’s culpability for hurtful transgressions, and thus is likely to foster relatively negative evaluations of the partner and relationship. In sum, we expected that greater partner guilt would predict more negative partner and relationship evaluations for individuals low in anxiety, but that the negative impact of guilt would be attenuated for individuals high in anxiety because, for them, the partner’s guilt also has the benefit of providing needed reassurance of their partner’s concern and commitment. Benefits and Costs of Feeling Guilt Guilt is also likely to be associated with a mix of positive and negative outcomes for the partner who feels guilty. On the positive side, prior research has shown that feeling guilt is associated with empathic concern, commitment, and closeness, and triggers reparative actions that should enhance relationships. People who are more com￾mitted and responsive to their partners, and who engage in more constructive relationship maintenance behavior, tend to foster health￾ier and happier relationships (e.g., Karney, & Bradbury, 1995; Rus￾bult, Bissonnette, et al., 1998; Wieselquist et al., 1999). Thus, the pro-relationship orientation underpinning guilt might mean that greater guilt will be associated with maintaining satisfying relation￾ships. In contrast, to the degree that low guilt reflects lack of care and motivation to make amends, it should predict declines in relationship quality. On the other hand, there is also evidence that feeling guilty can have negative consequences. Frequently feeling guilt im￾plies frequent negative experiences, which should undermine relationship satisfaction. Indeed, self-report data indicate that guilt-related cognition and behavior, such as perspective-taking and relationship maintenance efforts, are linked with positive relationship outcomes, but the (aversive) affective experience of guilt is not (Leith & Baumeister, 1998). Guilt is also often accompanied by other negative affective states, such as resentment, disappointment, low self-worth, and depressed mood (Jones & Ku￾gler, 1993), particularly when guilt has been induced by others (Baumeister et al., 1994, 1995). Guilty partners might also feel that they cannot expect love and care from their partner, which combined with a focus on making amends, could result in receiving compara￾tively less relationship rewards (e.g., support and affection; Jones & Kugler, 1993). As before, whether guilt has these potential benefits and costs is likely to depend on the context of guilt experiences. When guilt is created by one’s own care and concerns for the relationship and reflects authentic or internally-generated pro-relationship motiva￾tion, it is likely associated with maintaining relationships. This should be the case for partners of individuals low in anxiety because it is unlikely their guilt will be intentionally induced. In contrast, when guilt is induced or amplified by anxious reactions, the negative consequences of guilt should be more likely. People feel guilty even when they do not intentionally hurt others or when they think the hurt reaction is unreasonable, and this situation tends to highlight discrepancies between couple members’ expectations (Baumeister et al., 1994, 1995). When the hurt individuals’ expecta￾tions are perceived to be unreasonable, the hurt is disproportionate to the offense, and guilt is induced to gain reassurance, feeling guilty will be an overall negative experience. These conditions all reflect the induced guilt we hypothesized would be associated with attachment anxiety, and so we expected that feeling guilty would lead to declines in satisfaction for the partners of highly anxious individuals. Summary and Research Overview The central goal of individuals high in attachment anxiety is to obtain closeness and acceptance—a goal that is served by intense proximity-seeking designed to secure the partner’s care and atten￾tion. Our primary aim was to isolate the emotional and behavioral reactions to relationship threat that capture this primary motivation for connection. Their dependence and intense desire to obtain closeness should lead anxious individuals to experience greater hurt feelings when faced with relationship threats and, in turn, enact exaggerated expressions of hurt to induce guilt and reassurance from their partner. Moreover, although partner guilt should tend to have relatively neg￾ative effects on relationships across time, successfully inducing part￾ner guilt should provide anxious individual’s evidence of their part￾ner’s care and commitment. Thus, we predicted that the partner’s guilt should help anxious individuals maintain feelings of security and satisfaction in their relationship. To test these predictions, we assessed both individuals and their partner’s emotional and behavioral reactions to relationship threats, including when encountering hurtful partner behavior and conflict during daily life (Study 1) and when discussing aspects of the self the partner desired to change (Study 2). We expected that anxious individuals would experience more hurt feelings in these This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. 238 OVERALL, GIRME, LEMAY, AND HAMMOND