Previous research has demonstrated the importance of intimate partner support and criticism to health, but less is known about how these behaviors are regulated and expressed in relationships. The present research examines individual differences and social cognitive processes that may shape support and criticism in romantic relationships. Chapter 1 describes a study designed to test gender differences in intimate partner support. Forty college couples engaged in recorded, laboratory interactions. Using videorecall methods, participants and independent observers rated each partner’s behavior at periodic intervals within interactions. Results indicated that, compared to men on average, women sought more support but received the same amount of support. According to participants’ ratings, women were also more responsive to partners’ varying support needs over the course of an interaction, whereas observers’ data indicated no gender differences in partners’ responsiveness. Findings are discussed in light of previous research on gender differences and methods for behavior measurement. Chapter 2 describes two studies designed to test a dual-process model of criticism and social support in young adults’ romantic relationships. Evidence indicates that intentions play a limited role in guiding social behavior, particularly for behaviors that are well-practiced in stable contexts. The studies hypothesized that individuals’ behavioral intentions would predict their future criticism and support of romantic partners, but that intentions would be a stronger predictor in newer (versus longer lasting) relationships. Study 1 employed daily diary methods to capture actions in everyday life (N = 79 individuals), whereas Study 2 used video recall procedures to measure actions during focused laboratory discussions (N = 50 couples). Results were consistent across studies. Individuals’ intentions predicted their subsequent behavior, but partners in newer (versus older) relationships were more likely to carry out their support intentions. Although the expected pattern of results was found for support, the intention-behavior relationship for criticism was not moderated by relationship length. Implications for intervention and directions for future research are discussed.