Europe’s Journal of Psychology 3/2009, pp. 9-24 www.ejop.org Support Seeking, Caregiving and Conflict Management: Evidence from an Observational Study with Couples

Previous studies of couples have shown theoretically predicted links between 1) support seeking/attachment and marital functioning, and 2) caregiving and marital functioning (Mikulincer & Shaver, 2007). The present study further develops this area of research by investigating the link between conflict management and support seeking in couples, and the relationship between conflict management and caregiving behaviours in couples. Eighty couples completed the Potential Problem Checklist (Patterson, 1976) and a filmed 15-minute interactive problem-solving task. Independent teams of coders rated support seeking/caregiving and conflict management behaviours with the Secure Base Scoring System for Adults (Crowell et al., 1998) and the Global Couple Interaction Coding System (Belanger et al., 1993). It was hypothesized that a partner who asked for and gave support effectively would be less prone to withdraw, dominate, and criticize, and more likely to reinforce and listen to his/her partner and solve problems effectively. It was further hypothesized that men/women with a partner who sought and gave support effectively would be less likely to withdraw, dominate, and criticize, and more likely to reinforce and listen to his/her partner and demonstrate good problem solving skills. The results revealed that men and women who had effective support seeking and caregiving skills demonstrated greater conflict management skills during a discussion about a topic of conflict with their partner. The results further revealed a relationship


Introduction
John Bowlby described love as a dynamic state that incorporates each partner's individual tendencies for attachment (support seeking) and caregiving (Bowlby, 1969(Bowlby, /1982)).Bowlby conceptualized those two dimensions as behavioural systems that evolve in parallel and organize human behaviour to increase the probability of survival and adaptation (Mikulincer, 2006).Every individual's global welfare, including romantic, social, and emotional quality of life, depends to some degree on the proper functioning of these innate behavioural systems (Hazan & Shaver, 1987).
This study is designed to advance understanding of behavioural conflict management in couples by examining it though the lens of attachment theory.
According to Bowlby (1969Bowlby ( /1982), the attachment system (support seeking) is activated any time an individual is confronted with a situation that could threaten his/her survival.Once activated, the system's primary strategy is to seek proximity to the attachment figure (Bowlby, 1969(Bowlby, /1982)).If the attachment figure (e.g.parent, partner) responds consistently to the individual's support seeking needs, he will develop a secure attachment, that is, the belief that he is deserving of love, and confidence in the availability of attachment figures to meet his needs (Bartholomew, 1997).The behavioural caregiving system is fundamental to the development of an effective support seeking strategy.Attachment theory (Bowlby, 1969(Bowlby, /1982) ) suggests that the spousal caregiving system has two major functions: first, to provide a secure base for a partner by providing security when he/she is in distress; and second, to encourage autonomy and exploration when the partner is not in distress (Collins, Guichard, Ford, & J. A. Feeney, 2006;Mikulincer & Shaver, 2007).J. A. Feeney (2002) noted that attachment theory offers a good explanation for individual differences in the positive and negative behaviours manifested in the context of an intimate relationship.Previous studies have essentially shown a predictable link between the attachment/caregiving systems and romantic functioning (see Mikulincer & Shaver, 2007 for a review).Based on attachment theory, the present study therefore seeks to use direct observational measures of behaviour during a dyadic interaction to investigate both support seeking and spousal caregiving in relationship to conflict management.
To our knowledge, only one study (Crowell et al., 2002) has investigated the link between support seeking behaviours and communication behaviours during discussions of topics of conflict.Crowell and colleagues (2002) found that global positive and negative conflict management behaviours were significantly correlated with support seeking behaviours for both men and women, but do not represent the same construct.While these results provide interesting clues about the links between these two types of behaviours, the use of positive and negative global scores do not provide us with information about the specific behaviours at play.The reliability of the information provided by global positive/negative coding systems has been questioned in the past (Markman, 1991;Weiss, 1989).Accordingly, the current study improves on Crowell and colleagues (2002) by exploring the impact of support seeking and caregiving on more specific conflict management dimensions.In addition, it is important to note that Crowell's research sample consisted of engaged couples in their mid-twenties, most of whom were not cohabitating.As it is conceivable that attachment strategies would change as a relationship became more stable and durable, it would be interesting to examine the links between attachment strategy and conflict resolution in couples engaged in long term relationships.The present study responds to this need by studying care seeking and caregiving behaviours in a sample of older, established couples.

Caregiving and Conflict Management in Couples
Effective caregiving plays a major role in determining the quality and stability of conjugal relationships (Collins & B. C. Feeney, 2000).However, little research has examined the caregiving system relative to the attachment/support seeking system in adult relationships (Collins, Guichard, Ford, & B. C. Feeney, 2006).Crowell et al. (2002) has investigated the relationship between spousal caregiving and conflict management.They found that positive and negative conflict management behaviours and spousal caregiving behaviours were significantly correlated for both men and women, but reflect two different constructs.

Partner Effect
Studies have established that marital conflict management skills in one partner may depend on the other partner's attachment security or insecurity (e.g., Collins & Read, 1990;J. A. Feeney et al., 1994;Kobak & Hazan, 1991).The methods used to explore support seeking (Q-sort, self-report) and conflict management (self-report and observational measure) in these studies vary.The authors of the present study are interested in taking one step further and examining the impact of one partner's support seeking and caregiving behaviours on the other partner's conflict management behaviours during conflict resolution.Such a systemic perspective may contribute to a more comprehensive understanding of problems in romantic relationships, including problems with conflict management.

Hypotheses
The hypotheses for this study were: (1) participants' support seeking and caregiving behaviours will negatively predict their withdrawal, dominance, and criticism behaviours, and positively predict their support/validation and problem-solving abilities; (2) participants' support seeking and caregiving behaviours will negatively predict their partner's withdrawal, dominance, and criticism behaviours, and positively their partner's support/validation and problem-solving abilities.Each couple also participated in a videotaped discussion about a topic that was a source of conflict in the relationship.The discussion topic was determined from the couple's responses on the Potential Problem Checklist (Patterson, 1976).Prior to the discussion, the examiner ensured that both partners were willing to discuss the selected topic.

Measures
The Potential Problem Checklist (Patterson, 1976, translated into French by Bourgeois, Sabourin, & Wright, 1990) is a 16-item questionnaire about possible sources of conflict for couples (e.g., family, sexuality, children, finances, etc.).Respondents ranked each conflict theme on a 7-point Likert scale from "strongly disagree" to "strongly agree".

Europe's Journal of Psychology 14
The Secure Base Scoring System (SBSS) for Adults (Crowell, Pan, Gao, Treboux, O'Connor, & Waters, 1998)  The coding is the same as the coding for the four support seeking dimensions.A high score denotes sensitivity to the partner's distress, understanding of the distress, and responsiveness to the distress (Crowell et al., 2002).

Intercoder Agreement
Intercoder reliabilities for the SBSS and the GCICS coding systems were computed for one third of all discussions, using intraclass correlation.Bech and Clemmensen (1983) suggested that an intraclass coefficient between .41 and .60 represents moderate agreement, a coefficient between .61 and .80indicates substantial agreement, and a coefficient over .81shows near-perfect agreement between coders.In the current study, intraclass correlations generally indicated substantial or near-perfect agreement between the coders.Intraclass coefficients for the support seeking summary scale were .82for men and .78for women.Agreement for the caregiving summary scale was .82 for men and .87 for women.Coefficients were calculated for the following GCICS dimensions: withdrawal (.61 for men and .64 for women), dominance (.46 for men and .48 for women), criticism (.84 for men and .73 for women), support/validation (.77 for men and .85 for women), and problem solving (.78 for men and .64 for women).

Support Seeking, Caregiving, and Conflict Management Behaviours
Standard multiple regression analyses were conducted to test the first hypothesis, which stated that participants' support seeking and caregiving behaviours would predict their conflict management behaviours.Analyses were performed for each of the conflict management behaviours.In order to obtain clearly interpretable results, two preliminary steps were taken.First, support seeking and caregiving behaviours that were not related to conflict management behaviours in the correlation analyses were removed from further analysis.The examination of the links between these two sets of variables show that support seeking and caregiving behaviours were generally significantly correlated with conflict management behaviours for both men and women (see Table 1).More specifically, support seeking for men and women was positively correlated with positive conflict management strategies (i.e., support/validation and problem solving), but not significantly related to negative conflict management behaviours (i.e., withdrawal, dominance and criticism), with the exception of women's tendency to withdraw.Caregiving behaviours for both men and women were related to all five conflict management dimensions.
Second, support seeking and caregiving behaviours that were not significant in preliminary regression analyses were removed, and additional analyses were conducted with only the significant predictors.The results confirmed that men's and women's support seeking and caregiving strategies predict many of their conflict management behaviours.Bonferroni correction was applied to avoid Type 1 errors.
The significance level for all regression analyses was set at p = .002.Next, a series of standard multiple regression analyses were conducted to test the second hypothesis, which stated that participants' support seeking and caregiving behaviours would predict their partners' conflict management behaviours.As described above, two preliminary steps were taken prior to running the final analyses.First, support seeking and caregiving behaviours that were not related to conflict management behaviours in the correlation analyses were removed.
Correlations between one partner's support seeking and caregiving behaviours and the other partner's conflict management behaviours are presented in Table 2.
Interestingly, in most cases, participants' support seeking and caregiving behaviours that were related to their own conflict management behaviours also proved to be related to their partner's conflict management behaviours.Men's and women's support seeking and caregiving behaviours were positively related to their partner's constructive conflict management behaviours (i.e., support/validation and problem solving).Men's and women's caregiving behaviours were negatively correlated with their partners' negative conflict management behaviours (withdrawal, dominance and criticism), with the exception of male criticism.Men's support seeking behaviours were correlated with their partner's dominance and criticism.Women's support seeking behaviours were also correlated with their partner's negative conflict management behaviours (e.g.male withdrawal).In a second preliminary step, support seeking and caregiving behaviours that were not significant in preliminary regression analyses were removed and additional analyses were conducted with only the significant predictors.The results of these analyses confirmed that only men's and women's caregiving behaviours predicted their partners' conflict management behaviours.Bonferroni correction was applied to avoid Type 1 errors.The significance level for all regression analyses was set at p = .002.The results revealed that men's caregiving behaviours were significantly negatively correlated with their partners' withdrawal (F(1, 78) = 10.98,p < .001;Beta =

Discussion
The present study examined the relationship between support seeking/caregiving behaviours in couples and conflict management behaviours in couples.In stable couple relationships, both men's and women's support seeking and caregiving behaviours have a significant impact on their own conflict management behaviours.
This result is consistent with a secondary result from a 2002 study (Crowell et al.) that found that support seeking and caregiving behaviours are related to global communication behaviors in both male and female partners in young couples.The current study looked more closely at conflict management behaviours by exploring more subtle and precise interactions between support seeking/caregiving and conflict management behaviours.
The results of the present study reveal that effective support seeking is associated with greater problem solving abilities, possibly as a function of improved problem recognition and greater openness toward the partner.This finding implies that failure of either/both partner(s) to express distress during a discussion of a topic of conflict can decrease both the ability to recognize a problem and the desire to find solutions, making problem solving less likely.Seligman's (1975) theoretical model of learned helplessness may shed some light on these findings.Individuals who are ineffective in seeking support may respond with learned helplessness in uncontrollable situations (Abramson, Seligman, & Teasdale, 1978).They are consequently unable to solve a problem when faced with marital difficulties, even if they can find adequate ways to solve problems in other contexts.The present study also found that male and female caregiving behaviours influence some conflict management behaviours.Both men and women who were sensitive to their partner's distress, understood the distress, and responded appropriately to the distress were less likely to criticize and more likely to listen to their partner.These patterns are independent of gender.
Interestingly, some of the results revealed gender differences in attachment and communication quality.This finding replicates previous findings (e.g., Heene, Buysse, & Van Oost, 2005;Paley, Cox, Burchinal, & Payne, 1999;Wampler, Shi, Nelson, & Kimball, 2003), albeit with different measurement strategies.In the present study, ineffective support seeking in women predicted less withdrawal during communication with her partner.It has been generally observed that, in couples, men tend to withdraw more than women, whereas women are more likely to complain and criticize (Gottman, 1994).Withdrawal in the female partner may be viewed as a defensive decision not to seek care in the relationship based on the woman's belief that she cannot trust her partner.This pattern was not observed in men, for whom withdrawal behaviours may be less attachment-related and more a function of cultural norms.Women who provided good support for their partner showed greater problem solving ability.This relationship was not significant for men, which may imply that men's supportiveness is not essential to problem resolution in the relationship.Some men may be good problem solvers in addition to being sensitive and responsive to their partner's distress.Other men may be effective in recognizing and solving problems, but do not demonstrate sensitivity and responsiveness.Our results revealed that caregiving in men negatively predicted dominance, a result that was not found for women.The tendency to control discussions is less frequent among men who provide effective support to their partner.
Another conclusion suggested by the results of this study concerns the impact of men and women's support seeking/caregiving behaviors on their partner's conflict management behaviours.Although prior studies using other measures have highlighted the importance of examining the relationship between attachment in one partner in a relationship and conflict management in the other partner (Collins & Read, 1990;J. A. Feeney, 1994;Kobak & Hazan, 1991), no research has discussed the importance of the association between caregiving in one partner and conflict management in the other partner.The results of the present study found that one partner's caregiving behaviours influenced some of the other partner's conflict management behaviours.This relationship was gender specific.Women with supportive partners were less likely to avoid discussion and to criticize their partner.
They demonstrated greater skills in listening to their partner, recognizing problems, and seeking appropriate solutions.However, with the exception of more effective problem solving skills, men with supportive partners did not exhibit more constructive conflict management patterns.We can therefore conclude that men's caregiving had an important impact on women's conflict management behaviours, but that women's caregiving had no such effect on men.This difference may reflect the finding that women more than men tend to define themselves by their interpersonal relationships and attach greater meaning to the relationships (Wood, Rhodes, & Whelan, 1989).Our findings suggest a qualitative difference in men's and women's investment in the couple relationship.Unlike men, women need to feel supported by their partner in order to engage in constructive conflict management.
Overall, our findings suggest that men's and women's conflict management behaviours are more related to their own support seeking/caregiving behaviours than to their partner's support seeking/caregiving behaviours.This result is congruent with previous studies that used self-report attachment measures (J. A. Feeney et al., 1994;Simpson, 1990).An individual's ability to give or receive support is partially determined by the quality of his/her past significant attachment relationships (Bowlby, 1973;Bowlby, 1969Bowlby, /1982) and, to a lesser extent, determined by current dyadic interactions.This finding could inform couples therapy aimed at improving communication; the understanding on the part of each partner that his/her communication skills depend more on his/her own behaviours and experience than on his/her partner's behaviours could allow couples to set more realistic expectations for therapy.
There are two methodological limitations to this study.First, cross-sectional results do not allow us to draw conclusions about the direction of the relationship between support seeking/caregiving behaviours and conflict management behaviours (i.e.cause vs. effect).Further longitudinal research is necessary to determine the impact of support seeking/caregiving behaviours on marital quality variables.Second, our results are based on a sample of couples from the general population, and cannot be generalized to clinical populations.

Participants
Eighty heterosexual French-Canadian couples in Quebec participated in the present study.To be eligible to participate, couples had to have cohabitated for a minimum of 5 years.Participant couples were randomly selected from a sample of 312 couples that met the eligibility criteria.All participants were married (n = 106) or cohabiting (n = 54).The couples had been living together for an average of 9.67 years (SD = 3.71, range: 5 to 17 years).Participants had an average of 1.43 children (SD = 1.07).The mean age for men was 37.28 years (SD = 7.10) with 15.88 years (SD = 3.89) of formal education.The mean age for women was 34.40 years (SD = 6.32) with 15.55 years (SD = 2.92) of formal education.Mean annual income for men and women in the sample was $40,635 (SD = 22,825) and $21,033 (SD = 16,154), respectively (in Canadian dollars).Procedure Participants were recruited through newspaper, radio, and television ads.Participating couples attended a one-hour testing session.Men and women completed the self-report research questionnaires and consent forms independently.
was developed to assess support seeking and caregiving behaviours in partners during a dyadic interaction.It was designed to assess both partners' support seeking on four dimensions: (a) directness and clarity of the initial distress signal; (b) maintenance of the signal; (c) approach to the attachment figure; and (d) ability to be comforted.The four support seeking dimensions are coded on a 7point Likert scale and yield a 7-point summary scale of attachment.The summary scale is based on a global evaluation of the four support seeking dimensions and is coded with the SBSS grid.A high score indicates that the individual expresses his/her distress clearly both initially and throughout the discussion, approaches his/her partner with the clear belief that the partner should and will help, and is comforted by the partner's efforts to help.The SBSS also evaluates four dimensions of caregiving for each partner: (a) interest in the partner's distress; (b) recognition of the partner's distress; (c) interpretation of the partner's distress; and (d) responsiveness to the partner's distress.
The Global Couple Interaction Coding System (GCICS;Bélanger, Dulude, Sabourin, & Wright, 1993) measures five dimensions of marital interactions: (a) withdrawal (tendency to avoid discussion); (b) dominance (non-symmetrical control of the discussion); (c) criticism/attack/conflict (tendency to criticize, blame, or disparage the partner, including the use of non-verbal hostility, negative mind-reading, threats, and negative escalation); (d) support/validation (ability to list, validate, or reinforce the partner's statements); and (e) problem solving (ability to recognize a problem and find appropriate solutions).Each category is coded on a 4-point continuum from "absent" to "strong".Bélanger et al. (1993) have demonstrated satisfactory reliability coefficients (Pearson correlations) and validity estimate for this measure.The SBSS coding was completed by a psychologist trained by Judith Crowell and Dominique Treboux at the University of New York at Stony Brook, and two trained graduate students in psychology.The GCICS coding was completed by two graduate students in psychology supervised by Claude Bélanger, the author of the GCICS coding system.The GCICS and the SBSS coding are based on the same 15minute interaction.