This paper reports some of the preliminary findings from a longitudinal study designed to examine an integrated model of pair-bonding developed from an attachment perspective. The longitudinal study comprises data collection at four points during the transition to parenthood from the second trimester of pregnancy until 6 months after the birth. Data from the first two collection points are presented here.

Background

Although high levels of divorce and domestic violence are evident and the negative impact of marital conflict and transitions on children’s development are well documented, theoretical conceptualizations of enduring romantic relationships (or 'pair bonds', as they have been called by evolutionary psychologists) have been somewhat neglected.

The Pair-bond

Drawing on Bowlby’s seminal writings on attachment theory (Bowlby, 1969, 1973, 1980), Shaver, Hazan and Bradshaw (1988) contend that the pair-bond (or enduring romantic relationship) can best be explained by a model integrating three 'behavioural systems': attachment, caregiving and sexual mating. Within this model the attachment system is perceived as pivotal to the establishment of pair-bonds, laying the foundation for the other two systems. Attachment is the first system to develop within the individual and it subsequently influences the development of caregiving during childhood, and sexuality during adolescence.

Attachment, Caregiving and Sexuality in Marital Relationships

Although much of the early research on adult attachment focussed on dating relationships (for example, Hazan & Shaver, 1987; Feeney & Noller, 1990,1991) several recent studies have demonstrated the relevance of attachment styles to marital relationships. For example Feeney, Noller and Callan (1994) demonstrated that security of attachment was associated with constructive communication and marital satisfaction in a longitudinal study of couples in early marriage. Another cross-sectional study of couples sampled across the life-cycle of marriage also produced consistent findings (Feeney, 1994). Again relationships between attachment style and marital satisfaction were demonstrated.

Shaver and Hazan (1988), however, have highlighted two key ways in which bonds in infant-caregiver relationships differ from those in adult romantic relationships. First, they argue that romantic love is characterized by sexual attraction and sexual behaviour, especially in the early phases. Second , romantic love involves reciprocal caregiving while infant-caregiver bonds are asymetrical. These key features of romantic love again highlight the importance of considering the role of all three systems (attachment, caregiving and sexual mating) in the development and maintenance of love relationships.

In fact, the importance of caregiving in intimate bonds has been supported by Kotler (1985) who suggests that marital satisfaction is better predicted by an index of caregiving-carereceiving than by measures of personality, material circumstances and health. Further research is required, however, to clarify the exact nature of the relationship between caregiving and marital satisfaction, and whether there are specific times when caregiving may be particularly salient during the course of a marriage.

Surprisingly, of the three major components of the model of pair-bonding proposed by Shaver et al. (1988) sexuality is the most poorly researched. Although sexuality is acknowledged as a major contributor to relationship satisfaction and stability (Sprecher & McKinney, 1993) relationship researchers have largely ignored sexuality, and little systematic investigation of the role of sexuality in romantic relationships has been conducted. The majority of research on sexuality has been limited to what people do, how often they do it, and how they feel about what they do. Despite this, in their review of sexuality in close relationships, Sprecer and Mckinney (1993) list six distinct functions thought to be served by sex: self-disclosure, intimacy, love or affection, interdependence, maintenance, and exchange. Clearly, further research is required to clarify the contribution of sexuality to marital satisfaction. One time when sexuality may become particularly salient within marriage is during the transition to parenthood.

Transition to Parenthood

The transition to parenthood has been described as one of the most precipitous changes that takes place in most people’s lives (Rossi, 1968), and has been accordingly the focus of much research on adult development and the family life cycle (Terry, 1991a). The role of the parent is acquired abruptly (Miller & Sollie, 1980), is demanding (Terry, 1991b), and is one for which most people have little preparation (Rossi, 1968). Early theorists suggested that this transition is often experienced as a crisis (Hill, 1949), while more recent research has conceptualized it as a specific developmental phase characterised by significant personal, familial and social change (Levy-Shiff, 1994).

This transition in the adult life cycle is particularly salient to the model of pair-bonding examined in this study because of the adjustments required and the implications of these for the three behavioural systems (attachment, caregiving and sexual mating). More specifically, the dyadic marital relationship becomes triadic, and the new member of the triad is highly dependent. Consequently, additional family tasks are created, and typically the female takes major responsibility for these tasks (Levy-Shiff, 1994). Parents have to deal with the stresses of physical demands (e.g. loss of sleep), emotional demands (e.g. uncertainty about competence as a parent), marital strains (e.g. changes to sexual relationship with partner), and lifestyle changes (e.g. financial burden) (Miller & Sollie, 1980). In addition, the experience of providing childcare itself may be stressful (Terry, 1991a).

Many studies have demonstrated changes in marital socio-emotional patterns following the transition to parenthood. Companionate activities decrease postnatally, whereas conflict increases and marital quality may decline in many couples. The transition to parenthood is often a difficult period, with multiple stressors requiring major adjustments in the marital relationship (Levy-Shiff, 1994).

In particular, however, the literature reviewed suggests that considerable change may occur within each of the three behavioural systems during the transition to parenthood. For example, couples will need to make adjustments to their attachment system as the child becomes incorporated within it. They will need to adjust their caregiving practices to meet the extensive demands for caregiving of an infant. New parents may also need to modify their sexual practices considerably.

Hence, the major purpose of this paper is to examine any changes which occur in the three behavioural systems: attachment, caregiving and sexuality, and marital satisfaction during the transition to parenthood.

More specifically, the aims of this paper are:

1. to examine any changes which occur in the three behavioural systems: attachment, caregiving and sexuality, between the 2^nd trimester of pregnancy and the first month after birth.
2. to examine any changes in marital satisfaction which occur between the second trimester of pregnancy and the first month after birth.
3. To examine the contribution of the three behavioural systems: attachment, caregiving and sexuality to marital satisfaction at the second trimester of pregnancy and the first month after birth.

As this is the first study to examine the three behavioural systems during the transition to parenthood specific hypotheses about changes which would occur in each system at specific times could not be generated. In general, however it was expected that 'pregnant' couples would report more change in each of the three behavioural systems than a comparison group of 'childless' couples during the course of the whole study, from the second trimester of pregnancy until 6 months after the birth of the child (approximately 9 months.) It was also expected that greater decline in marital satisfaction may become evident in couples experiencing the transition to parenthood than childless couples, although the extent and exact timing of any decline could not be predicted. Finally, it was expected that each of the three behavioural systems (attachment, caregiving and sexuality) would make an independent contribution to marital satisfaction.

METHOD

Participants

The participants in this study were 107 married couples who were in the second trimester of their first successful pregnancy and 100 married couples who had no children and had decided not to have children in the immediate future. Demographic characteristics of the two samples are shown in Table 1. No significant differences were found between the two samples on age, length of relationship, length of marriage, religiosity, or education level.

This paper reports the findings from data collected at two times: time 1: during the second trimester of pregnancy and time 2: 4-6 weeks after the birth for "pregnant" couples, and at the same time intervals for comparison couples. Some attrition of both samples had occurred by completion of time 1 and again at time 2 for a range of reasons. The most common reason was relocation, but in the "comparison" group 6 couples also became pregnant and withdrew from the study. At the the time of writing this paper the number of "pregnant" couples who had completed time 1 was 98 and the number of "comparison" couples who had completed time 1 was 86. Time 2 data collection was incomplete. Of 96 "pregnant" couples still in the study at time 2, data from 94 couples was available for analysis. Of 85 couples still in the "comparison" group, data from 69 couples was available. In subsequent analyses reported the sample size may vary considerably, however, due to missing data and listwise deletion which occurs with multivariate analyses. Hence, any results reported in this paper for time 2 should be treated cautiously unless they are confirmed by the full data set which will be available shortly.

Attachment, caregiving, sexuality and marital satisfaction were assessed through self-report measures administered at time 1 and time 2 (with a range of other measures not relevant to this report).

The Attachment Style Questionnaire (ASQ) (Feeney, Noller & Hanrahan, 1994) was used to assess attachment dimensions. The ASQ consists of 40 items with a 5 point Likert response scale. Two major attachment dimensions: Anxiety over abandonment and discomfort with closeness, were derived from the ASQ.

The Caregiving Style Questionnaire (Kunce & Shaver, 1994) was used to assess caregiving. This 32 item questionnaire was designed to assess caregiving between romantic partners on four subscales: Proximity versus Distance, Sensitivity versus Insensitity, Cooperation versus Coercion, and Compulsive Caregiving. For the purposes of this study the 5 point response format was altered to a 6 point format (Not at all like me — Completely like me) to be consistent with other measures administered in the same package.

Three measures of sexuality were used:

1. Sexual satisfaction was assessed using the 25 item Index of Sexual Satisfaction developed by Hudson, Harrison and Crosscup (1981). For the purposes of this study the 5 point likert response scale was modified to a 6 point scale (Strongly Disagree to Strongly Agree).

2. Sexual desire was assessed with 14 items selected from the 25 item Index of Sexual Desire developed by Hurlbert (1988). The items selected were those which included reference to the partner and item 23 which had the words "with my partner" added. Only these items were used because the focus of this study was limited to sexual desire within the couple’s relationship. The same 6 point Likert scale response format was used as for sexual satisfaction.

3. Sexual communication satisfaction was measured by the Sexual Communication Satisfaction scale developed by Wheeless, Wheeless and Baus (1984). The response format of this scale was modified to be consistent with the 6 point Likert response scale used for sexual satisfaction and sexual communication.

Snyder’s Marital Satisfaction Inventory (Snyder, 1979) was used to assess marital satisfaction. Due to the length of this scale, 35 items were selected. Eight to ten items were selected from each of the four scales by selecting those items with the highest loadings (r-bis): Global Distress (10 items), Affective Communication (8 items), Problem-solving Communication (9 items) and Time Together (8 items).

Procedure

Participants were recruited by a variety of means. These included advertising in newspapers, during radio interviews, at universities and through maternity wards at two major Brisbane hospitals. Flyers were also distributed at shopping centres and maternity shops in particular.

Volunteers who met the criteria of being married and either planning to remain childless in the near future or being in the second trimester of their first successful pregnancy were scheduled for an interview which both members of the couple could attend. At the interview the purpose of the study was explained, consent forms were signed, and a brief interview designed to obtain background information and demographic data was conducted. At the end of this interview each couple was given two packages of self-report questionnaires to be completed by each participant individually and confidentially, and were asked to return the packages by mail as soon as possible. Approximately 4-6 months later and/or 4-6 weeks after the birth, couples were sent a second pair of packages of self-report questionnaires and were asked to complete these and return them at a second interview scheduled for the following week.

To examine whether change occurred in the three systems between time 1 and time 2, a series of MANOVAs with group (2) as the between subjects variable and gender and time (2) as within subjects variables was conducted. A set of variables within each behavioural system served as the dependent variables for 3 separate MANOVAs. Two dependent variables were entered for the attachment system: discomfort with closesness and anxiety over abandonnment. Four dependent variables were entered for the caregiving system: proximity, sensitivity, cooperative and compulive caregiving. Finally three dependent variables were entered for the sexuality system: sexual satisfaction, sexual desire and sexual communication.

Overall these analyses revealed no main effects for group or time, and no interactions. Main effects for gender were found, however, for each of the behavioural systems.

For attachment the main gender effect was strong (F (2,151) = 12.341, p = .000). Univariate F tests indicated a very significant gender effect on the discomfort dimension of attachment (F(1,152) = 22.590, p = .000). Inspection of the means showed that husbands reported higher discomfort with closeness than wives.

For caregiving the main effect for gender was also strong (F (4,151) = 7.323, p = .000). Univariate F tests indicated gender effects for proximity (F (1, 154) = 9.971, p = .002) and sensitivity (F (1,154) = 14.292, p = .000) in caregiving. Inspection of the means showed that wives were higher on both of these aspects of caregiving.

Finally for sexuality the main effect for gender was also strong (F (3,132) = 7.323, p = .000). Univariate F tests showed a weak effect for sexual satisfaction (F (1,134) = 4.263, p = .041), a very strong effect for sexual desire (F (1,134) = 66.426, p = .000), and a moderate effect for sexual communication (F (1,134) = 6.495, p = .012). Inspection of the means showed that wives report more sexual satisfaction and communicate more about sex than husbands while husbands report higher levels of sexual desire.

To examine whether marital satisfaction declined between time 1 and time 2, a MANOVA with group (2) as the between subjects variable and gender and time (2) as within subjects variables was conducted. Four measures of marital satisfaction: global distress, affective communication, problem solving and time together were entered as dependent variables.

No main effects for time or gender were found on marital satisfaction. A main effect for groups (F (4,149) = 3.535, p = .009), however, did emerge. Univariate F tests showed a group effect on time together (F (1, 152), = 7.882, p = .006). Inspection of the means indicates that the pregnant couples group spend less time together than the comparison group.

In addition an interaction effect between gender and time (F (4,149) = 4.153, p = .003) was found. Univariate F tests showed that gender and time have an interactive effect on global distress (F (1,152) = 6.815, p = .010) and time together (F (1,152) = 7.628, p = .006). To clarify these interactions 4 additional MANOVAs were conducted with time together and global distress as the dependent variables. A MANOVA was conducted on each sample (pregnant husbands and wives, "comparison" husbands and wives) with time as a within subjects variable. These MANOVAs indicate that time effects were significant for "pregnant" wives (F (2,86) = 11.289, p = .000) and "comparison" husbands (F (2,76) = 3.483, p = .036) only. For "pregnant" wives univariate F tests showed a significant time effect for both time together (F (1,87) = 12.765, p = .001) and global distress (F (1,87) = 4.95, p = .029). Inspection of the means indicates that both time together and global distress decreased between time 1 and time 2 for "pregnant" wives. For comparison husbands, univariate F tests showed only global distress was significantly effected by time (F (2,76) = 3.483, p = .036). Inspection of the means indicates that "comparison" husbands reported more global distress at time 2 than time 1.

Furthermore a three way interaction between group and gender and time was also obtained (F (4,149) = 3.463, p = .010). Univariate F tests showed that there was a group x gender x time interactive effect on time together (F (1,152) = 8.862, p = .003).

Two additional MANOVAs were conducted to clarify this interaction with time together as the dependent variable. A MANOVA was performed for each group, with gender and group as within subjects variables. Consistent with the findings described above, the only significant effect revealed by these analyses was an interaction between sex and time in the pregnant group (F (1,90) = 10.40, p = .002). Figure 1 shows the means for time together for "pregnant" husbands and wives at time 1 and time 2. A significant decrease in "pregnant" wives time together is evident in Figure 1, while "pregnant" husbands show no significant change.

Finally two additional weak effects were obtained. While no overall effect was found for a group by time interaction, univariate F tests revealed a weak effect for a group x time interaction on global distress (F (1,152) = 4.196, p = .042). Inspection of the means suggests that the pregnant group may have decreased in global distress between time 1 and time 2 while the comparison group may have increased slightly in global distress. A weak gender effect was also found, although no overall gender effect was found for marital satisfaction. Univariate F tests showed a weak gender effect on affective communication (F (1,152) = 5.56, p = .020). Inspection of the means showed that husbands had a tendency to report higher levels of affective communication than wives.

Marital Satisfaction and the 3 behavioural systems: attachment, caregiving and sexuality

To examine the relationships between attachment, caregiving, sexuality, and marital marital satisfaction at time 1 and time 2 Pearson product-moment correlations matrices were obtained for "pregnant" husbands and wives and "comparison" husbands and wives. For the purposes of these analyses an overall measure of marital satisfaction was obtained by adding scores on the 4 scales of Snyder’s Marital Satisfaction Inventory. Tables 2 and 3 show the correlations obtained. As can be seen from these tables, although minor variations between samples occur, in general attachment, caregiving and sexuality variables are moderately and significantly related to marital satisfaction. Within each behavioural system measures are also generally moderately, and in some cases strongly, and significantly correlated with each other. In addition, consistent relationships are also evident between the three behavioural systems, for example, attachment variables are related to caregiving and sexuality variables.

To examine the contribution of the three behavioural systems to marital satisfaction a series of hierarchical multiple regressions were performed at time 1 and time 2 for "pregnant" husbands and wives and "comparison" husbands and wives. (Note that while analyses including time 1 and time 2 data within the same regression analysis would be valuable, at this preliminary stage of data analysis, within the whole longitudinal study, only separate analyses for time 1 and time 2 are reported here.) The three behavioural systems were entered in the developmental sequence in which they are thought to emerge. Hence attachment variables were entered in step 1, caregiving variables were entered in step 2 and sexualtiy variables were entered in step 3. Summary tables of these regressions appear in Tables 4 and 5. As can be seen from these tables for each sample, at each time a significant amount of the variance in marital satisfaction is explained by the three behavioural systems (R squared = 23-44%, F = 3.03-6.10, p<.01). Some variation is evident, however, across samples and times as to the significance of the contribution to marital satisfaction of each step, whether each behavioural systems makes a unique contribution to marital satisfaction, and the best predictor of marital satisfaction.

Table 6 provides a summary of the results of the hierarchical regressions for the four samples at times 1 and 2. As can be seen in Table 5 for pregnant wives at time 1, each of the three behavioural systems made a significant contribution to marital satisfaction as they were entered into the equation, but only a caregiving and a sexuality variable made a unique contribution to marital satisfaction in the final analysis. The best predictors of marital satisfaction proved to be the caregiving measure of sensitivity and the sexuality measure of sexual satisfaction. By time 2, however, some changes had occurred. Although the three behavioural systems accounted for the same amount of variance in marital satisfaction, and attachment and caregiving made significant contributions to marital satisfaction as these sets of variables were entered into the equation, sexuality no longer made a significant contribution as this set of variables was entered into the equation at step 3. Furthermore, in the final analysis none of the variables entered made a unique contribution to marital satisfaction.

The best predictors were the sexuality variable sexual communication and three of the caregiving variables: caregiving cooperation, proximity and —compulsive caregiving, but none of these made a unique contribution.

For "pregnant" husbands an even greater change between time 1 and time 2 is evident. Here it would appear that the 3 behavioural systems account for a larger proportion of the variance at time 2 (44%) than at time 1 (23%). A different pattern of changes in the relationships between the behavioural systems and marital satisfaction over time also emerges. For "pregnant" husbands at time 1, only the attachment system variables make a significant contribution as they are entered into the equation at step 1, and none of the variables makes a unique contribution to marital satisfaction in the final analysis. The best predictors of marital satisfaction are two of the sexual measures: sexual communication and sexual satisfaction but neither of these variables make a unique contribution to marital satisfaction. At time 2, however, each of the three behavioural systems make a significant contribution as they are entered into the equation, and a caregiving and a sexuality measure make unique contributions in the final analysis. The best predictors of marital satisfaction for "pregnant" husbands at time 2 are caregiving sensitivity and sexual desire.

For "comparison" husbands and wives some change is also evident between time 1 and time 2 in the pattern of relationships between the three behavioural systems and marital satisfaction, but this does not appear to be as strong as change evident in the "pregnant" samples. For example, for "comparison" husbands sexual satisfaction is a unique predictor of marital satisfaction at both times, and for "comparison" wives a similar mix of caregiving and sexuality variables are the best predictors of marital satisfaction, although at time 1 none of these variables make a unique contribution to marital satisfaction while at time 2 cooperative caregiving and sexual satisfaction do make unique contributions.

Table 6 also shows that across samples and across times sexuality variables emerge consistently as the best predictors of marital satisfaction.

1. As data collected at time 2 was incomplete at the time of writing this paper, the results should be interpreted cautiously. On the basis of the data analysed to date, however, it would appear that while the three behavioural systems: attachment, caregiving and sexuality contribute to marital satisfaction, there is little change in these systems during the initial transition into parenthood, marital satisfaction is reasonably robust during this transition, and the three behavioural systems are largely interdependent.

2. With regard to the first expectation that "pregnant" couples would report more change in each of the behavioural systems than a comparison group of "childless" couples during the course of the whole study, there is no support to date from the data analysed at time 1 and time 2. The results of the MANOVAs conducted on the variables within each behavioural system, suggest that attachment, caregiving and sexuality do not change consistently over time and do not vary significantly between the "pregnant" and the "comparison" group throughout the initial transition to parenthood. This is somewhat surprising since the transition to parenthood literature so frequently describes this transition as a highly stressful time and it would be expected that the attachment system in particular would become more salient during times of stress. Whether changes become evident over the final phase of the study remain to be seen. At this point, however, the most consistent finding to emerge from the first set of MANOVAs is the strong gender differences observed across groups and time. In summary, wives were significantly lower than husbands on the attachment dimension of discomfort with closeness, reported higher levels of proximity (or closeness) and sensitivity in their caregiving style, reported less sexual desire, more sexual satisfaction and communicated more about sex.

3. Mixed support was obtained for the second expectation that greater decline in marital satisfaction may become evident in couples experiencing the transition to parenthood than childless couples. On the one hand. the results indicate that on one of the four measures of marital satisfaction: time together, "pregnant" couples were signficantly lower than childless couples, although subsequent analyses suggest that this effect was largely the result of a decline in time together reported by pregnant wives as they moved through the transition to parenthood. "Pregnant" husbands did not show this type of decline. On the other hand, however, there was no evidence of decline in marital satisfaction for the "pregnant" group on two measures of marital satisfaction: affective communication and problem-solving communication. Furthermore, evidence of improvement in marital satisfaction was also obtained on the global distress measure. A gender x time interaction was found for global distress and subsequent analyses indicate that ""pregnant" wives reported less global distress over time, while "comparison" husbands showed more.

4. Partial support was also obtained for the expectation that each of the three behavioural systems (attachment, caregiving and sexuality) make an independent contribution to marital satisfaction. While there is no doubt that attachment, caregiving and sexuality variables were signficantly related to an overall measure of marital satisfaction, it is also apparent that there are significant relationships between the variables from the three different systems and that a considerable proportion of the variance in marital satisfaction explained was often shared by variables from two and sometimes three of the systems. For example, although attachment variables were significant when they were entered in Step 1 in six of the eight hierarchical multiple regressions reported, in the final analyses neither attachment variable made a unique contribution to marital satisfaction. Hence, while attachment variables often do contribute to marital satisfaction it would appear that the variance they explain is shared with caregiving and sexuality variables. Similarly, caregiving variables made a signficant contribution to the regression equation when they were entered in Step 2 in seven of the eight regressions. This suggests that caregiving does make an additional signficant contribution to marital satisfaction than attachment alone. After sexuality variables were also entered, however, in Step 3, caregiving variables continued to make a unique contribution in only four of the eight final analyses. This suggests that sexuality variables often share the variance accounted for by caregiving. In fact, however, when sexuality was entered in Step 3, it only made a signficant increase in the variance explained in three of the eight analyses. On, the other hand, sexuality variables did make unique contributions to marital satisfaction in five of the eight final analyses, with the variable sexual satisfaction making the majority of these unique contributions. Overall, then it does appear that two of the three systems, caregiving and sexualty may make some independent contributions, from those of attachment, towards marital satisfaction. At this stage, however, a consistent pattern has not emerged and larger samples may be required to fully resolve this issue.

5. One consistent finding which did emerge from the regression analyses, however, was the frequency with which sexuality variables appeared as the best predictors, assessed by beta weights, of marital satisfaction. Sexual satisfaction emerged as a particularly good predictor of marital satisfaction, although, interestingly this was not the case for "pregnant" wives after the birth of the child. Sexual satisfaction had been one of the best predictors for "pregnant" wives at time 1, but sexual communication appeared to be more important to marital satisfaction by time 2.

6. Another finding which emerged through the regression analyses is also worthy of discussion. This is the finding of a substantial change in the amount of variance in marital satisfaction accounted for by the three systems in "pregnant" husbands between time 1 and time 2. In light of the evidence for no change occurring within the three behavioural systems during the transition to parenthood reported above, this finding suggests that while little change may occur within the systems, some change may occur in the salience of the three behavioural systems at least for husbands during the transition to parenthood.

References

Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York: Basic Books.

Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiely and ange . New York: Basic Books.

Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss. New York: Basic Books.

Feeney, J. A. & Noller, P. (1990). Attachment style as a predictor of adult romantic relationships. Journal of Personality and Social Psychology 58,281-291.

Feeney, J.A. & Noller, P. (1991). Attachment style and verbal descriptions of romantic partners. Journal of Personality and Social Psychology , 8, 187-215.

Feeney, J.A., Noller, P. & Callan, V.J. (1994). Attachment style, commumication and satisfaction in the early years of marriage. In K. Bartholomew & D. PerIman (Eds.), Advances in personal relationships Vol. 5: Attachment processes in adulthood (pp. 269-308). London: Jessica Kingsley.

Feeney, J.A., Noller, P., & Hanrahan, M. (1994). Assessing adult attachment: Developments in the conceptualization of security and insecurity. In M.B. Sperling & W.H. Berman (Eds.) Attachment in adults: Theory, assessment and treatment (pp. 128-152). New York: Guilford.

Feeney, J.A. (1994). Attachment style, communication patterns and satisfaction across the life cycle of marriage. Personal Relationships, 1, 333-348.

Hazan, C. & Shaver, P.R. (1987). Romantic love conceptualized as an attachment process. Journal of Personality-and Social Psychology, 52, 511-524.

Hill, R. (1949). Families under stress. New York: Harper & Row.

Hudson, W.W., Harrison, D.F., & Crosscup, P.C. (1981). A short form scale to measure sexual discord in dyadic relationships. Journal of Sex Research. 17(2), 157-174.

Kotler, T. (1985). Security and autonomy within marriage. Human Relations, 38, 299-321.

Kunce, L.J., & Shaver, P.R. (1994). An attachment-theoretical approach to caregiving in romantic relationships. In K. Bartholomew & D. PerIman (Eds.) Advances in personal relationships. Vol. 5: Attachment processes in adulthood. (pp. 205-237). London: Jessica Kingsley.

Levy-Shiff, R. (1994). Individual and contextual correlates of marital change across the transition to parenthood. Developmental Psychology, 30, 591-601.

Miller, B.C. & Sollie, D.L. (1980). Normal stresses during the transition to parenthood. Family Relations, 29, 459-465.

Rossi, A.S. (1968). Transition to parenthood. Journal of Marriage and the Family 30,26-39.

Shaver, P.R. & Hazan, C. (1988). A biased overview of the study of love. Journal of Social and Personal Relationships, 5, 473-501.

Shaver, P. R., Hazan, C., & Bradshaw, D. (1988). Love as attachment: The integration of three behavioural systems. In Stemberg, R. J. & Barries, M.L. (Eds), The Psychology of Love (pp. 68-99). New Haven, CT: Yale University Press.

Snyder, D. K. (1979). Multidimensional assessment of marital satisfaction. Journal of Marriage and the Family 41, 813-823.

Sprecher, S. & McKinney, K. (1993). Sexuality . Newbury Park, CA: Sage.

Terry, D. J. (1991 a). Transition to parenthood. In P. C. L. Heaven (Ed.) Lifespan Development (pp. 184-211). Sydney: HBJ.

Terry, D.J. (1991b). Stress, coping and adaptation to new parenthood. Journal of Personal and Social Relationships, 8, 527-547.

Wheeless, L.R., Wheeless, V.E., & Baus, R. (1984). Sexual communication, communication satisfaction, and solidarity in the developmental stages of romantic relationships. The Western Journal of Speech Communication. 48. 217-230.