UC Berkeley

The demand-withdraw interaction pattern is a common, deleterious pattern in which one spouse blames or pressures while the other spouse avoids or withdraws (Christensen, 1988). Studies consistently show that: 1) there tends to be gender differentiation in the interaction roles, with women demanding and men withdrawing, and 2) demand-withdraw behaviors are associated with marital dissatisfaction. The existing observational research on marital interactions, however, has been overwhelmingly conducted with relatively young couples and does not take into account other known predictors of marital dissatisfaction. The present study examined demand-withdraw behaviors longitudinally in a sample of middle-aged and older couples. Later life stages may be associated with changes in emotion-related behaviors (e.g., Carstensen, 1991) and in gender roles (e.g., Gutmann, 1987). Studying demand-withdraw behaviors over time in middle-aged and older couples enables a determination of whether the manifestations of this set of behaviors, and its negative association with marital satisfaction, change during the later stages of development. This study further evaluated the effect of demand-withdraw on marital satisfaction in relation to other factors known to be negatively associated with this important outcome (e.g., physiological arousal, self-reported negative affect, and negative emotion behaviors during conflict).

A sample of 126 married couples (63 middle-aged, 63 older) were observed at three time points across a 13-year span as they engaged in a 15-minute conversation about an area of relationship conflict. Conversations were videotaped and trained raters used an observational coding system to quantify each partner's demand and withdraw behaviors. During or shortly following the conflict conversations, measures of physiological arousal, self-reported negative affect, and emotion behavior were also collected. The couples also completed self-report measures of marital satisfaction at each of the three time points of observational data collection, as well as at two subsequent time points.

Results showed that demand-withdraw behaviors occur during conflict in both middle-age and older couples at overall comparable rates. Combining this finding with previous work indicates that this pattern is found throughout the life course. Importantly, the specific pattern of behaviors changes with age. There appears to be a marked increase in one type of withdraw behavior, avoidance, over time. Moreover, contrary to theories proposed by Gutmann and others that gender differences diminish in later life, gender differentiation appears to become greater over time with wives in the demand role and husbands in the withdrawing role.

Results further demonstrated that the relationship between demand-withdraw behaviors and marital dissatisfaction remains the same across the lifespan. That is, for both middle-aged and older couples, demand-withdraw behaviors are negatively associated with concurrent levels of marital satisfaction. Furthermore, these behaviors show significant interactions with other factors known to be associated with lower levels of marital satisfaction (e.g., physiological arousal, self-reported negative affect, and negative emotion behaviors). The general pattern was that demand-withdraw behaviors, while deleterious on their own accord, are particularly pernicious when manifest in the context of other factors associated with marital dissatisfaction. When examining the longitudinal effect of demand-withdraw behaviors, however, results indicated that after controlling for initial levels of marital satisfaction, these behaviors were not strong predictors of the trajectory of change in marital satisfaction over time.

The findings are discussed in terms of socioemotional changes couples undergo as they move from middle-age into late life. This study offers evidence not only about changes in demand-withdraw behaviors themselves but also as to how marital processes and gender roles may change over the life course. Future work will be valuable in further elucidating changes in the nature of demand-withdraw behaviors and in the consequence of this interaction pattern for marital satisfaction at different life stages.