Integrating research on attachment anxiety, rejection, self-regulation and health behaviours, I investigated the interactive effect of attachment anxiety and relationship threat on self-regulation. I hypothesized that self-regulation would decrease following a threat to one’s romantic relationship. Moreover, I expected that this association would be moderated by attachment anxiety, such that it would be stronger for individuals high, relative to low, in attachment anxiety. In three laboratory experiments, relationship threat was made salient and participants were given the opportunity to consume snack foods. In a non-experimental diary study, participants’ relationship stress and health behaviours were measured for a period of seven days. The results for Study 1 indicated that women, but not men, high in attachment anxiety experienced self-regulation failure (i.e., ate more jelly beans) when relationship threat was elicited. In contrast, although women low in attachment anxiety demonstrated the same pattern, it was attenuated and non-significant. Study 2 was designed to replicate the findings from Study 1 using a more powerful relationship threat manipulation that I hoped would cause a consistent pattern for both genders. Contrary to hypotheses, both men and women high in attachment anxiety experienced increased self-regulation (i.e., ate fewer brownie pieces) in the experimental, compared to the control, condition. One important difference between these studies was the presence of the partner in the laboratory in Study 2. I hypothesized that this might account for the discrepant findings between these studies and designed Study 3 accordingly, such that participants either participated alone or with their partners. Unfortunately, this study did not reconcile the discrepancies between the first two studies: Only a main effect of relationship threat on number of jelly beans consumed emerged. For Study 4, individuals completed questionnaires for seven days that assessed how fluctuations in relationship stress interacted with attachment anxiety to affect health outcomes. Analyses of the same-day and lagged effects demonstrated several significant interactions consistent with the hypotheses. These four studies provide initial evidence for the interactive effect between attachment anxiety and relationship stress on health outcomes and well-being. Although the findings were inconsistent, two of the studies provided support for my hypotheses. / Thesis (Ph.D, Psychology) -- Queen's University, 2010-08-17 21:23:46.461