Psychosomatic Medicine Faster Service from Outside North America
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gallo, L. C.
Right arrow Articles by Matthews, K. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gallo, L. C.
Right arrow Articles by Matthews, K. A.
Related Collections
Right arrow Blood Pressure
Right arrow Social Support
Psychosomatic Medicine 68:253-261 (2006)
© 2006 American Psychosomatic Society

ORIGINAL ARTICLES

Adolescents' Attachment Orientation Influences Ambulatory Blood Pressure Responses to Everyday Social Interactions

Linda C. Gallo, PhD and Karen A. Matthews, PhD

From the SDSU/UCSD Joint Doctoral Program in Clinical Psychology (L.C.G.), San Diego State University, San Diego, California; and the Department of Psychiatry (K.A.M.), University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania.

Address correspondence to Linda C. Gallo, PhD, SDSU/UCSD Joint Doctoral Program in Clinical Psychology, 6363 Alvarado Ct. Suite 103, San Diego, CA 92120. E-mail: lcgallo{at}sciences.sdsu.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 NOTES
 REFERENCES
 
Objectives: Adult attachment theory provides a useful framework for understanding how early developmental experiences affect social processes, associated physiological stress responses, and ultimately, health across the lifespan. The current study examined the effects of attachment orientation on physiological responses to naturalistic social interactions in adolescents.

Methods: Two-hundred five black (49%) and white high school students (14–16 years; 50% boys) completed a measure of anxious and avoidant attachment, and underwent 1.5 days of ambulatory blood pressure and heart rate (AmBP and AmHR) monitoring while they tracked social experiences.

Results: As predicted, individuals with higher avoidant attachment reported fewer interactions with friends (t [299] = –3.18, p < .01) and more anxious adolescents experienced less pleasant interactions both during (t [299] = –3.59, p < .01) and outside of school hours (t [298] = –3.59, p < .01). Individuals who were higher in anxious attachment showed augmented ambulatory diastolic and systolic blood pressure (AmDBP, AmSBP; both p < .05) in conjunction with current or recent interactions with friends. More avoidant adolescents exhibited augmented AmDBP responses to social conflict (p < .05).

Conclusions: The current findings suggest that adolescent attachment orientation may shape later physical health through influences on the nature of, and physiological responses to, social interactions in everyday life.

Key Words: adolescence • attachment style • ambulatory blood pressure • ecological momentary assessment • social environment

Abbreviations: AmBP = ambulatory blood pressure; AmDBP = ambulatory diastolic blood pressure; AmSBP = ambulatory systolic blood pressure; AmHR = ambulatory heart rate; BMI = body mass index; HLM = hierarchical linear modeling.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 NOTES
 REFERENCES
 
Important relationships in early development are believed to have a pervasive and enduring effect on emotion regulation, social functioning, and physiological stress responses (1–4). Attachment theory (5,6) provides a useful theoretical framework for understanding these associations. This conceptual model posits that early interactions with caregivers become internalized into working models, or schemas, of the self, others, and relationships, and that these cognitive structures influence social-emotional processes across the lifespan (7,8). Children who experience a lack of nurturance early in life may be less securely attached as adolescents and adults (9). They may be mistrustful of others and have difficulty forming or utilizing supportive social relationships, thereby fostering vulnerability to stress (10). In contrast, individuals who experience consistent, nurturing caregiving are more likely to develop a secure attachment orientation along with social competencies and emotion regulation skills that help diffuse stress and foster adjustment (10,11).

Given these socioemotional patterns, individuals with a less secure attachment orientation could experience more stress in everyday life. Across time, associated changes in the hypothalamic-pituitary-adrenal and sympathetic-adrenal-medullary systems may engender cumulative damage, creating vulnerability to health problems (4). Indeed, laboratory studies have shown that insecurely attached children and adults, or those who report disrupted or conflictive developmental experiences, display elevated and protracted physiological responses to stressful social situations when compared with securely attached persons (1,12–18). Furthermore, animal models have shown effects of early social deprivation on physiological stress regulation, whereas animal and human studies suggest that supportive social contacts have a downregulating effect on stress responses (19–21). Thus, through social-emotional influences, and concomitant physiological alterations, attachment orientation could ultimately shape later physical health outcomes (2,22,23).

With few exceptions (24–28), the majority of research concerning attachment and social experiences to date has relied on self-report methodology or laboratory manipulations. This is particularly problematic given that a prior study found substantial differences in associations between attachment and social and emotional reports made through aggregate, retrospective assessments, versus those reported in vivo, in everyday life (25). These differences suggested the operation of memory-related biases that were shaped by attachment orientation. Moreover, the few prior studies that have examined attachment and social experiences in everyday life are characterized by several limitations. First, two used an event-contingent reporting procedure (29) that required participants to initiate a diary entry to describe interactions as they occurred (25,26), and one asked participants to recall and record "important" interactions at the end of the day (24). As a result, reports may have been affected by participants' willingness to spontaneously report social experiences, or by their recollection of experiences, which could be related to attachment orientation. These studies also applied categorical methods of assessment, whereas current research and theory considers attachment to be dimensional in nature (30,31). Two additional diary studies have focused exclusively on interactions within dating or married couples (27,28). Most importantly, to our knowledge, no prior study has examined the implications of attachment style for physiological responses to naturally occurring social situations. However, if the supposition that attachment orientation has physical health implications is tenable, social and physiological correlates of attachment should be observable in everyday life across multiple relationships and interactions.

In the current study, adolescents tracked their social experiences in everyday life through an ecological momentary assessment paradigm while their ambulatory blood pressure (AmBP) and ambulatory heart rate (AmHR) were monitored. Participants were asked to describe current or very recent social experiences each time the BP monitor inflated. Hence, all participants were prompted to describe their social experiences at similar intervals and times of day, thus eliminating the potential limitations of an event-contingent or retrospective reporting procedure. In addition, consistent with most recent research (32,33), adolescents completed a scale measuring the two fundamental dimensions that are believed to underlie attachment orientation, specifically, anxious and avoidant attachment. Within close relationships, individuals who are high in anxious attachment tend to be sensitive to rejection cues, demanding, clingy, jealous, vigilant, and emotionally reactive (32,34). In contrast, persons high in avoidant attachment display autonomy, distrust, and discomfort with intimacy; they tend to eschew relationships and prefer to view themselves as not needing others (34,35). Importantly, adolescence represents a key phase during which to examine attachment given key social-developmental changes such as learning to balance closeness and autonomy with parents while initiating mature and intimate relationships with peers (36,37). Moreover, attachment orientation may be central to the ability to negotiate these developmental challenges successfully (38,39). The current study therefore expanded on prior research by studying physiological responses to naturalistic social situations, incorporating a systematic procedure for eliciting social information, assessing attachment orientation from a dimensional perspective, and targeting a sample facing rapid social changes and developmental milestones highly relevant to attachment.

The effects of attachment should be most pronounced within the context of situations that activate the attachment system such as those involving close relationship partners (40,41). Parents are the most obvious attachment figures in childhood, but adolescents and adults frequently characterize friends and romantic partners as falling into this realm as well (40,42). Thus, in the current sample, we examined the implications of attachment for social experiences involving friends (which might include significant others as well as platonic relationships) and parents. In addition, we examined whether attachment orientation affected AmBP and AmHR responses to qualitative aspects of social interaction, specifically, ratings of conflict and pleasant interactions. We also examined the influence of attachment on the number of interactions with parents and friends and on the degree to which interactions were characterized as pleasant or involving conflict. Our predictions followed from attachment theory and the current understanding of its implications for social functioning and health (2,6,35,43) as well as prior empirical studies concerning the daily life social correlates of attachment (24–26). However, in the latter regard, forming direct comparisons was difficult, because the methodology used in the current study—particularly the dimensional rating scale—diverged from that of prior studies.

Consistent with the conceptual definition of avoidant attachment and the results from a prior study of daily social interactions of college students (26), we hypothesized that more avoidant individuals would report fewer interactions with parents and friends than their less avoidant counterparts, whereas anxious attachment would not relate to interaction frequency. Prior diary-based studies have examined attachment in relation to a variety of qualitative responses to social interactions and have produced mixed evidence of differences among attachment groups (24–26). Overall, the available literature suggests that individuals who are higher on either dimension of attachment insecurity might be likely to find social interactions less pleasant than those who do not espouse these characteristics, but for different reasons. Specifically, more anxious persons may experience distress associated with feelings of rejection or interpersonal sensitivity, whereas avoidant persons may find intimacy and social exchange to be aversive. The current study therefore examined the hypothesis that individuals with higher avoidant or anxious attachment would describe their interactions as less pleasant than persons with lower scores on these dimensions. Although behavior within, and emotional responses to, conflict situations are influenced by attachment orientation (44,45), previous research has not shown that attachment orientation predicts the frequency or intensity of conflict in everyday social interactions (25,26). We likewise did not predict associations between attachment and ratings of conflict in the current study.

Given their high levels of vigilance and sensitivity to rejection, we anticipated that more anxiously attached adolescents would show higher AmBP and AmHR when interacting with parents and friends as compared with other readings. We predicted that this effect would be especially salient in relation to friends, considering that peer relationships are often dynamic and intense at this life stage. Because avoidant individuals disengage from emotional closeness and intimacy, and typically view relationships as unimportant (46), we did not predict an effect of avoidant attachment on AmBP responses to interactions with friends or parents. Considering the close association between attachment and conflict tactics and responses (44,45), as well as prior studies suggesting that social conflict experiences foster AmBP increases (47), we predicted effects of attachment on cardiovascular responses to conflict. Avoidant persons may find conflict situations especially aversive given their preference for interpersonal remoteness, whereas conflict may necessitate interchange and responsiveness (23,25). Thus, we predicted a positive association between avoidant attachment and AmBP and AmHR responses to social conflict. We did not expect that anxious attachment would augment AmBP and AmHR responses to conflict because of conflicting results from previous laboratory studies (45,48) and a study of naturalistic interactions of adolescents (25).


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 NOTES
 REFERENCES
 
Participants and Recruitment
The sample consisted of 217 14 to 16 year-olds (50% male, 51% white, 49% black) recruited from two Pittsburgh, Pennsylvania, area high schools with diverse student populations. Study staff traveled to schools, provided an overview of the research, and asked interested students to provide contact information. A letter describing the study was then mailed home to students' parents and was followed by a recruitment phone call. If the parent and child were interested in possible participation, a brief screening was performed. Adolescents were eligible if parents reported that their child was free from cardiovascular diseases and medications with autonomic nervous system effects and did not weigh more than 80% above ideal weight relative to height for the child's age and sex. Study personnel then arranged to meet with parents, obtained informed consent, and arranged for the child to participate in the ambulatory and laboratory portions of the study. Five participants were excluded from the current analyses because they were missing AmBP data and seven were excluded because they did not complete the attachment questionnaire (final n = 205). All procedures for the current study were approved by the University of Pittsburgh's Institutional Review Board. Data were collected between May 1999 and May 2002.

Procedure
Ambulatory monitoring was scheduled for two consecutive school days when participants did not have out of the ordinary events. A research technician met the participant at school on the first day of monitoring, demonstrated the use of the AmBP monitor, calibrated the monitor, explained the diary, and allowed the participant to complete practice entries. AmBP was measured every 30 minutes from 8 am until 10 pm and every 60 minutes between 10 pm and 5 am. The monitor was set to take automatic readings at fixed intervals, to operate in dynamic inflation mode (i.e., to inflate to 30 mm Hg higher than the previous reading), and to perform one retry if an erroneous reading was taken (e.g., as a result of arm movement or insufficient inflation pressure). Participants were permitted to remove the monitor to shower the following morning on wakening (after 5 am). Participants then returned to school where an assistant reconnected the monitor; readings were resumed every 30 minutes on the second school day, from 8 am until 1:30 pm. Participants completed a diary entry following each daytime AmBP reading. Because of differences in sleep and wake schedules, minor variations occurred in the amount of diary entries completed across participants. The current study examined daytime measures only, consistent with the primary focus on social interaction.

Attachment Measure
Participants completed the Measurement of Attachment Quality (MAQ [48]), a 14-item, measure of attachment orientation. Subjects characterized their agreement with each statement on a four-point Likert scale from strongly agree to strongly disagree. Minor modifications were made to accommodate the young age of the sample. Specifically, the phrase "my partner" was changed to "someone I care about" in two items. The item "My desire to merge sometimes scares people away" was changed to "I like to be so close with others that it sometimes scares people away."

Considerable research now indicates that attachment is best described according to two dimensions, which capture the level of anxiety and avoidance in attachment organization (49). Hence, we subjected the MAQ items to principal axis factoring, with varimax rotation, requesting a two-factor solution. The resultant factors had eigenvalues of 3.26 and 2.55 and explained 23% and 18% of the total scale variance, respectively. The first factor consisted of seven items reflecting discomfort with closeness, for example, "I get uncomfortable when someone wants to be very close"; "I prefer not to be close with others." The second factor consisted of seven items reflecting a desire for excessive intimacy and fear of rejection, for example, "I have trouble getting others to be as close as I want them to be"; "I often worry someone I care about will not want to stay with me." Scale scores were created by summing the items that loaded on each factor with higher scores indicating a more avoidant (i.e., first factor) or anxious (i.e., second factor) attachment style. Internal consistencies for these scales were acceptable, as shown in Table 1. The two subscales were weakly but significantly correlated (r [203] = 0.14, p < .05).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Descriptive Statistics for Anxious and Avoidant Attachment and Ambulatory Blood Pressure and Ambulatory Heart Rate

 

Diary
Participants completed a diary entry on a Palm-brand (Palm Inc., Santa Clara, CA) handheld computer after daytime AmBP measurements. The current study examined location and posture at the time of cuff inflation and physical activity and consumption of food, caffeine, and smoking in the 10 minutes before cuff inflation. In addition, we examined whether participants reported talking with friends (or not) and parents (or not) and the degree to which they experienced conflict or disagreement or a pleasant interaction (on a Likert scale with choices as follows: 1 = NO!, 2 = NO, 3 = no, 4 = yes, 5 = YES, 6 = YES!) in the 10 minutes before the reading.

Ambulatory Blood Pressure
AmBP and HR were obtained using the Accutracker DX ambulatory monitor (Suntech Medical Instruments Inc., Raleigh, NC). The monitor uses a noninvasive, auscultatory technique and has been shown to accurately assess cardiovascular changes during simulated ambulatory assessments with mental and physical challenges (50). Data were uploaded to a personal computer using the AccuWin software package (Suntech Medical Instruments, Inc.).

Data Analytic Strategy
The influence of attachment on the quantity and quality of social interaction was tested in hierarchical linear regression analyses performed in SPSS 13.0 (SPSS, Chicago, IL). Outcomes were the proportion of entries during which participants reported talking with friends and parents (separate analyses) and average ratings of interaction quality (i.e., conflict, pleasant).1 Diary entries made during and after school hours were examined separately, because interactions with parents occurred nearly exclusively outside of school (i.e., all but 1%). The covariates ethnicity, sex, and their interaction were entered at the first step of analysis,2 and the anxious and avoidant main effects were entered at the second analytic step. Thus, we examined the degree to which the attachment effects predicted social interaction variables after accounting for demographic covariates.

AmBP and AmHR data were analyzed through multilevel models conducted in HLM 5.04 (51,52). Using this procedure, multiple observations on each individual are viewed as nested within persons rather than the same for all persons as occurs in standard repeated-measures analyses. This allows for unequal numbers and increments of observations across participants, as often occur in diary studies [see (53) for a detailed discussion of random coefficients multilevel modeling analysis of diary data]. In multilevel modeling terminology, two "levels" of analysis were applied with repeated AmBP and AmHR readings and diary entries serving as the unit of analysis at level 1 and individuals serving as the unit of analysis at level 2. At the first, within-person analysis stage, a separate regression equation was calculated for each individual with multiple ambulatory systolic blood pressure (AmSBP), ambulatory diastolic blood pressure (AmDBP), and AmHR measures serving as outcomes in separate analyses and multiple measures of social variables (i.e., interactions with parents; interactions with friends; conflict; pleasant interactions) serving as predictors. Thus, intercepts, representing the average levels of AmSBP, AmDBP, or AmHR, and slopes, representing the influence of each social interaction variable on the outcomes, were estimated for each person. Location (three dummy codes comparing school to home, car, and other), posture (two dummy codes comparing lying down to sitting and standing), consumption of food or substances (i.e., one dummy code, reflecting any versus no consumption), and physical activity (three dummy codes comparing no activity with mild, moderate, and strenuous activity) were included as covariates. Covariate effects were treated as fixed to reduce model complexity, and they were centered about the sample mean so that intercept estimates were adjusted for between-person variations in exposures. To protect against type I error risk, effects of all social interaction variables were modeled simultaneously. Thus, each effect was tested while controlling for all other effects. Social interaction variables were centered about the sample mean before entry to adjust intercepts for between-person differences in the quantity and quality of interaction.

At the second stage of analysis, or level 2, the person-specific intercepts and slope parameters served as outcomes and were predicted by the between-subjects dimensions anxious and avoidant attachment. Anxious and avoidant attachment were standardized (mean = 0, standard deviation = 1) for ease of interpretation. The between-subjects covariates body mass index (BMI; kg/m2), ethnicity, sex, and the ethnicity by sex interaction were included as covariates at level 2 and were centered about the sample mean.

We used a model-building approach so that initial models were first conducted that contained only covariates to determine whether significant between-subjects variability remained in the average levels of AmSBP, AmDBP, and AmHR after controlling for covariates (based on chi-squared tests for variance components). If significant variability was identified, this indicated that the intercept parameters should be treated as randomly varying and that examining level 2 predictors (i.e., attachment effects) would be useful. We next performed random-intercepts models regressing average levels of AmBP and AmHR (i.e., intercepts) calculated at level 1 on the attachment main effects (level 2 predictors). We then added the level 1 social interaction effects to the models to evaluate the degree of between-person variability in the association between the social variables and the physiological outcomes. If significant between-subjects variability was not identified for a given slope parameter, this suggested that participants did not vary substantially in the manner in which social experiences influenced their AmBP or AmHR. In these cases, the slope parameter was treated as fixed, and effects of attachment were not tested. Finally, we performed random-intercepts and slopes models, which tested our primary predictions concerning the moderating effects of attachment orientation on the association between social interaction variables and AmBP and AmHR. These analyses involved examining cross-level interactions, because attachment at level 2 was hypothesized to interact with social experiences at level 1 to influence the physiological outcomes.

Chi-squared tests were used to examine whether the model fit was significantly improved with the addition of the attachment dimensions to the preceding models. R2 values were calculated to provide an estimate of the amount of variance in each parameter (i.e., intercepts, slopes) explained by the attachment effects and were calculated according to the following formula: R2 change = (variance for model without attachment effects – variance for model with attachment effects)/variance for model without attachment effects (51). We also examined individual model parameters to determine whether specific, directional predictions regarding anxious and avoidant attachment were supported.

For all analyses, alpha levels of 0.05 (two-tailed) were used to evaluate associations based on a priori predictions, whereas alpha levels of 0.01 (two-tailed) were used to indicate statistical significance for unpredicted associations. Missing data were excluded on an analysis-specific basis so that degrees of freedom vary somewhat across analyses.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 NOTES
 REFERENCES
 
Attachment and Participant Characteristics
One-way analyses of variance showed no gender, ethnic, or age differences for anxious or avoidant attachment (all p > .10). Descriptive statistics for the attachment variables are shown in Table 1.

Attachment and the Topography of Daily Social Interaction
Table 2 summarizes the hierarchical linear regression analyses that examined the associations between attachment and daily social interactions. On average, participants reported talking to their friends during 79% of readings at school and 35% of readings after school hours. Adolescents reported talking to parents during 32% of readings taken after school. Adolescents generally described their social interactions as pleasant and low in conflict/disagreement.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Results of Hierarchical Linear Regression Analyses Examining the Associations Between Anxious and Avoidant Attachment and the Frequency and Quality of Daily Social Interactions

 

After controlling for effects of sex and ethnicity, the attachment effects accounted for 6% of the variance in frequency of interactions with friends during school hours. As predicted, more avoidant adolescents reported fewer interactions with friends, although this effect did not extend to interactions with friends after school. The attachment main effects predicted the number of interactions with parents (after school hours) with 3% of variance explained. Contrary to predictions, more anxiously attached participants, rather than more avoidant participants, reported fewer interactions with their parents. The attachment variables predicted 6% of variance in pleasant ratings both at home and school. Consistent with predictions, more anxiously attached participants reported less pleasant interactions across the monitoring period. However, avoidant attachment did not relate to this outcome.

Overall Effects of Attachment on Ambulatory Blood Pressure and Ambulatory Heart Rate
Initial models showed significant interindividual variability for AmSBP, AmDBP, and AmHR after accounting for covariates, suggesting the use of examining between-subject predictors. Average levels of AmBP and AmHR and reliabilities of estimates are shown in Table 1. The attachment effects did not explain significant between-person variability in average AmBP or AmHR observed across the monitoring period.

Attachment and Ambulatory Blood Pressure and Ambulatory Heart Rate Responses to Social Interactions
Chi-squared tests revealed significant interindividual variability in the associations between all social interaction variables and AmSBP after accounting for covariates and attachment main effects. Thus, for AmSBP, all of the social interaction slope effects were modeled as random and attachment interaction effects were examined. For AmDBP, only the interactions with friends and conflict effects varied significantly. Interactions with parents and pleasant ratings were therefore modeled as fixed, and attachment predictors were not included for these equations. In the analysis that examined AmHR, the effects of interactions with friends and pleasant interactions showed significant interindividual variability. Conflict and interactions with parents were modeled as fixed effects, and attachment effects were not examined.

Ambulatory Systolic Blood Pressure
Adding the cross-level interactions to the model examining AmSBP did not result in a significant improvement in model fit (Table 3). However, as predicted, parameter tests indicated that higher anxious attachment was associated with significantly greater AmSBP reactivity to interactions with friends (10% of between-persons variability explained).


View this table:
[in this window]
[in a new window]
  		TABLE 3. Results of the Multilevel Modeling Analyses Examining the Moderating Effects of Attachment on the Associations Between Daily Social Experiences and Ambulatory Blood Pressure and Ambulatory Heart Rate

 

Ambulatory Diastolic Blood Pressure
The attachment effects explained significant variability in the effects of social interaction variables on AmDBP. In general, participants displayed increased AmDBP during interactions with friends relative to other readings, and, as predicted, this effect was amplified for participants higher in anxious attachment. The attachment variables accounted for 9% of the variability in AmDBP responses to interactions with friends. Consistent with predictions, a positive association also emerged between avoidant attachment and AmDBP responses to conflict, whereas the overall effect of conflict on AmDBP in the sample was negligible. The attachment variables accounted for 19% of the interindividual variability in AmDBP responses to conflict.

Ambulatory Heart Rate
Addition of the anxious and avoidant main effects to equations predicting AmHR responses to interactions with friends and pleasant ratings did not result in a significant improvement in model fit. However, there was a nonsignificant trend for the effect of interactions with friends on AmHR to be augmented in more anxiously attached participants.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 NOTES
 REFERENCES
 
Recently, researchers have shown growing interest in the physical health implications of early social environments and attachment orientation (1,2). Although most prior studies used cross-sectional frameworks and assessed subjective health outcomes, some evidence suggests that early social experiences could have a more direct association with objective health indicators, including chronic disease burden and mortality (54,55). In part, this association may reflect enduring influences on social-emotional and physiological functioning (1,2,4). To begin to explore this possibility, we investigated the association between attachment orientation and adolescents' social interactions in everyday life and their AmBP and AmHR responses to those social interactions.

Attachment and Social Interactions
Consonant with the predictions of attachment theory (7,8) and with prior research (26), attachment shaped the nature of naturalistic interpersonal interactions. As a group, adolescents reported frequent social interactions with their friends. However, this was less the case for individuals higher in avoidant attachment during school hours. Avoidant attachment did not relate to interactions with parents or friends after school, perhaps because these analyses were based on a smaller number of readings. The association between avoidant attachment and low frequency of peer interactions could reflect the fact that avoidant adolescents perceive fewer individuals as friends. Alternatively, their lack of interaction with peers could hinder the development and continuity of close relationships so that they actually have fewer friends. Adolescence represents a pivotal period of social development, during which social energies typically shift away from parents, and peer and romantic relationships become increasingly intimate and relevant (39,56). A lack of relationships with peers is associated with social and emotional problems in adolescence (57,58), and social adjustment in adolescence predicts well-being in adulthood (59). Given the substantial research suggesting a link between social integration and health (60,61), these findings demonstrate a potential indirect pathway through which avoidant attachment could shape health outcomes.

Individuals with higher anxious attachment experienced less pleasant interactions in everyday life. Social interaction may elicit negative feelings of anxiety and rejection for anxiously attached persons (24) and therefore may entail emotional costs as well as benefits. Unexpectedly, a small but significant effect was also found for more anxious participants to report less frequent interactions with parents, although again, given the limited sampling of interactions outside of school, findings regarding parents should be viewed tentatively.

Contrary to predictions, avoidant attachment was not associated with the degree to which daily interactions were viewed as "pleasant." Findings from prior studies concerning avoidant attachment and qualitative aspects of daily interactions have been quite mixed and furthermore, have varied based on whether avoidant attachment was conceptualized as a single category or as two types (i.e., fearful avoidant, or high avoidant and anxious attachment, and dismissive-avoidant, or high avoidant and low anxious attachment) (25,26). Furthermore, the use of a one-item assessment of positive interaction quality is a limitation in the current study. However, our results tentatively suggest that when levels of anxious attachment are controlled, avoidant attachment may be less strongly related to perceptions that social interactions are, or are not, pleasant.

Consistent with prior research (25,26), attachment did not predict adolescents' social conflict experiences. One might expect that given their jealous and emotional behavior (34), anxiously attached persons would encounter greater social conflict. However, they may actively avoid or seek to diffuse conflict situations because of the potential threat these circumstances entail. For example, in a prior study in which adolescents were observed while they interacted with romantic partners, anxiously attached individuals did not display problematic conflict management behaviors (62). The authors of this study asserted that anxiously attached individuals may "bite their tongue" despite feelings of emotional distress as a result of abandonment fears.

Attachment and Ambulatory Blood Pressure and Ambulatory Heart Rate Responses to Social Interactions
Consistent with predictions, adolescents with higher anxious attachment had augmented AmSBP, AmDBP, and a tendency toward augmented AmHR responses to interactions with friends. The consistency of the effect across physiological outcomes supports the robust nature of this association. Notably, more anxiously attached adolescents did not show elevated cardiovascular responses in interactions with their parents perhaps because they were somewhat less likely than others to interact with their parents. Furthermore, adolescents' view their peer relationships as extremely important; however, these relationships may tend to be unstable and may vary substantially in duration, thus causing unpredictability and augmenting fears of rejection and loss.

We also found that individuals with higher avoidant attachment showed augmented AmDBP responses to experiences of conflict. More avoidant persons may find conflict distressing, in part, because they manage these situations poorly. A prior study showed that dismissive persons (i.e., high avoidance, low attachment anxiety) exhibited negative behaviors such as disdain, hostility, and defensiveness when attempting to resolve conflict with intimate partners (62). However, avoidant persons did not typically display strong, outward displays of anger and aggression. Additional research has shown that avoidant persons often form hostile attributions for negative and even ambiguous social situations, use escapist responses, and lack awareness of physiological indicators of anger (63). Consistently, the fact that AmDBP in particular was augmented in relation to conflict in the current study could suggest a negative and stifled emotive response rather than active attempts to reconcile disagreement. More anxious persons did not display augmented reactivity to conflict experiences, and this may reflect the fact that they view social conflict less negatively given the potential for increased interpersonal intimacy (25).

In combination with prior research showing effects of attachment orientation or experiences on physiological stress responses in laboratory situations (13,15), the current findings suggest a potential route through which early social environments could directly shape health outcomes. Attachment theory suggests that early experiences become internalized and shape cognition, behavior, and emotion in later relationships (9). Anxious attachment may reflect a history of inconsistent and unpredictable caregiving, leading to negative views of self and fears of rejection. Thus, contrary to less anxious individuals who might find the presence of friends or intimate relationship partners to be soothing, more anxious persons may be more likely to exhibit signs of physiological arousal when in the presence of close relationship partners. Individuals who are more avoidant may have been exposed to neglectful or harsh developmental experiences, fostering negative views of others. These negative views may promote maladaptive behaviors and augmented physiological responses during conflict. Although the effects of attachment on AmBP and AmHR were relatively small, the cumulative effects could be important because AmBP measures are highly predictive of cardiovascular health outcomes (64,65). However, it should also be noted that attachment orientation can change in response to traumatic events or later relationships (for discussion and review, see [31]).

Strengths, Limitations, and Future Research Directions
The current findings should be interpreted in the context of several methodological features. First, as a result of the practicalities of assessing high school students, social interaction was examined over a circumscribed time period. Thus, assessment of social experiences outside of school hours (and subsequently, of interactions with parents) was quite limited. Furthermore, information concerning structural aspects of social interaction such as partner sex was not collected. Identities of interaction partners were also assessed broadly ("friends" rather than best friends or romantic partners and "parents" rather than mothers or fathers); however, the effects of attachment may be sensitive to subtle differences among relationship types. Qualitative ratings of interpersonal interactions were likewise evaluated in a limited manner, and additional types of experiences might be relevant to physiological responses. A more refined analysis of social interaction is needed to fully understand the association between attachment and responses to social interaction to help elucidate the specific pathways through which attachment influences physiological responses.

The current findings should also be viewed in light of the number of statistical comparisons made and the fact that not all findings were consistent with predictions. Effects that emerged consistently across outcomes (e.g., the association of anxious attachment with AmBP and AmHR responses to interactions with friends and with ratings of "pleasant" interactions) and which are consistent with attachment theory (e.g., the relation between avoidant attachment and AmDBP responses to conflict) should be given greater emphasis. Overall, future research is recommended to replicate and expand on the current findings.

The use of contemporaneous accounts is a particular strength of the current study in terms of minimizing the strong influence of attachment on information processing and recall (66,67). The timed, stamped records in the dairy were by design conducted within a few minutes of the AmBP readings, thus corroborating that perceptions were recorded in compliance with the study protocol. The addition of physiological end points is an important contribution of the current research not only because the observed associations substantiate the potential health implications of attachment, but also because they show that attachment predicts objective (i.e., physiological) responses to social interactions. Additional strengths of the study include the fact that the data were obtained in a controlled fashion with persons undergoing similar environments for most of 2 days, and the focus on adolescents, who are undergoing rapid social, emotional, and physiological changes that should facilitate the identification of attachment effects.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
NOTES
 REFERENCES
 
Recent research has focused on the possible implications of early familial experiences and attachment orientation for physiological functioning and, ultimately, physical health (1,2,4). The current study provides preliminary support for the tenet that anxious and avoidant attachment could affect health indirectly by shaping the nature of social interactions and directly by moderating physiological stress responses to social interactions. The current research adds to the handful of studies showing that attachment is an important predictor of naturalistic social interactions and is the first to show effects of attachment on ambulatory cardiovascular responses. That these observations were obtained in early adolescence suggests that attachment orientation may have a cumulative effect on relationships and cardiovascular responses to interpersonal interactions, at least in individuals with stable attachment patterns. Future research using diary methods and ambulatory physiological monitoring is recommended to explore some of the questions generated in the current research.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 NOTES
REFERENCES
 
1Preliminary analyses showed that the distributions for frequency of interactions with parents and friends, and conflict experienced at home, were substantially skewed. Thus, the data were transformed using square root transformations. A natural log transformation of reflected data was necessary to normalize the distribution for frequency of interaction with friends at school. Because results of analyses using transformed variables were nearly identical to those using raw values, we maintained the analyses of raw values for parsimony. Back

2Prior research has demonstrated few sex or ethnic differences in associations of attachment and social experiences. Therefore, we controlled statistically for effects of sex and ethnicity rather than exploring two-, three-, and four-way interactions of sex and ethnicity with the attachment dimensions. This approach also reduced the complexity of the analyses and was appropriate given the level of statistical power. Back

This research was supported by NIH training grant HL07560. The first author was supported by NIMH grant MH6610101 during preparation of this article.

DOI:10.1097/01.psy.0000204633.33599.81


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 NOTES
 REFERENCES
 

  1. Luecken LJ, Lemery KS. Early caregiving and physiological stress responses. Clin Psychol Rev 2004;24:171–91.[CrossRef][Medline]
  2. Maunder RG, Hunter JJ. Attachment and psychosomatic medicine: developmental contributions to stress and disease. Psychosom Med 2001;63:556–67.[Medline]
  3. Repetti RL, Taylor SE, Seeman TE. Risky families: family social environments and the mental and physical health of offspring. Psychol Bull 2002;128:330–66.[CrossRef][Medline]
  4. Taylor SE, Lerner JS, Sage RM, Lehman BJ, Seeman TE. Early environment, emotions, responses to stress, and health. J Pers 2004;72:1365–93.[CrossRef][Medline]
  5. Ainsworth MS, Blehar MC, Waters E, Wall S. Patterns of Attachment: A Psychological Study of the Strange Situation. Hillsdale, NJ: Erlbaum; 1978.
  6. Bowlby J. Attachment and Loss, vol 1. Attachment. New York: Basic Books; 1969.
  7. Bowlby J. The making and breaking of affectional bonds. II. Some principles of psychotherapy. The 50th Maudsley Lecture. Br J Psychiatry 1977;130:421–31.[Abstract/Free Full Text]
  8. Bowlby J. The making and breaking of affectional bonds. I. Aetiology and psychopathology in the light of attachment theory. An expanded version of the 50th Maudsley Lecture, delivered before the Royal College of Psychiatrists, 19 November 1976. Br J Psychiatry 1977;130:201–10.[Abstract/Free Full Text]
  9. Bowlby J. A Secure Base: Parent-Child Attachment and Healthy Human Development. New York: Basic Books;1988.
  10. Dozier M, Stovall KC, Albus KE. Attachment and psychopathology in adulthood. In: Cassidy J, Shaver P, eds. Handbook of Attachment: Theory, Research, and Clinical Applications. New York: Guilford Press; 1999:497–519.
  11. Cassidy J. Adult romantic attachments: a developmental perspective on individual differences. Rev Gen Psychol 2000;4:111–31.
  12. Carpenter EM, Kirkpatrick LA. Attachment style and presence of a romantic partner as moderators of psychophysiological responses to a stressful laboratory situation. Pers Relatsh 1996;3:351–67.
  13. Feeney BC, Kirkpatrick LA. Effects of adult attachment and presence of romantic partners on physiological responses to stress. J Pers Soc Psychol 1996;70:255–70.[CrossRef][Medline]
  14. Spangler G, Schieche M. Emotional and adrenocortical responses of infants to the strange situation: the differential function of emotional expression. Int J Behav Dev 1998;22:681–706.
  15. Luecken LJ. Childhood attachment and loss experiences affect adult cardiovascular and cortisol function. Psychosom Med 1998;60:765–72.[Abstract/Free Full Text]
  16. Nachmias M, Gunnar M, Mangelsdorf S, Parritz RH. Behavioral inhibition and stress reactivity: the moderating role of attachment security. Child Dev 1996;67:508–22.[CrossRef][Medline]
  17. Woodall KL, Matthews KA. Familial environment associated with type A behaviors and psychophysiological responses to stress in children. Health Psychol 1989;8:403–26.[CrossRef][Medline]
  18. Luecken LJ, Rodriguez AP, Appelhans BM. Cardiovascular stress responses in young adulthood associated with family-of-origin relationship experiences. Psychosom Med 2005;67:514–21.[Abstract/Free Full Text]
  19. Insel TR, Young LJ. The neurobiology of attachment. Nat Rev Neurosci 2001;2:129–36.[CrossRef][Medline]
  20. Taylor SE, Klein LC, Lewis BP, Gruenewald TL, Gurung RA, Updegraff JA. Biobehavioral responses to stress in females: tend-and-befriend, not fight-or-flight. Psychol Rev 2000;107:411–29.[CrossRef][Medline]
  21. Uchino BN, Cacioppo JT, Kiecolt-Glaser JK. The relationship between social support and physiological processes: a review with emphasis on underlying mechanisms and implications for health. Psychol Bull 1996;119:488–531.[CrossRef][Medline]
  22. Ciechanowski P, Russo J, Katon W, Von Korff M, Ludman E, Lin E, Simon G, Bush T. Influence of patient attachment style on self-care and outcomes in diabetes. Psychosom Med 2004;66:720–8.[Abstract/Free Full Text]
  23. Turner-Cobb JM, Gore-Felton C, Marouf F, Koopman C, Kim P, Israelski D, Spiegel D. Coping, social support, and attachment style as psychosocial correlates of adjustment in men and women with HIV/AIDS. J Behav Med 2002;25:337–53.[CrossRef][Medline]
  24. Kafetsios K, Nezlek JB. Attachment styles in everyday social interaction. Eur J Soc Psychol 2002;32:719–35.[CrossRef]
  25. Pietromonaco PR, Barrett LF. Working models of attachment and daily social interactions. J Pers Soc Psychol 1997;73:1409–23.[CrossRef][Medline]
  26. Tidwell MC, Reis HT, Shaver PR. Attachment, attractiveness, and social interaction: a diary study. J Pers Soc Psychol 1996;71:729–45.[CrossRef][Medline]
  27. Campbell L, Simpson JA, Boldry J, Kashy DA. Perceptions of conflict and support in romantic relationships: the role of attachment anxiety. J Pers Soc Psychol 2005;88:510–31.[CrossRef][Medline]
  28. Bradford SA, Feeney JA, Campbell L. Links between attachment orientations and dispositional and diary-based measures of disclosure in dating couples: a study of actor and partner effects. Pers Relatsh 2002;9:491–506.[CrossRef]
  29. Wheeler L, Reis HT. Self-recording of everyday life events: origins, types, and uses. J Pers 1991;59:339–54.[CrossRef]
  30. Fraley RC, Waller NG, Brennan KA. An item response theory analysis of self-report measures of adult attachment. J Pers Soc Psychol 2000;78:350–65.[CrossRef][Medline]
  31. Gallo LC, Smith TW, Ruiz JM. An interpersonal analysis of adult attachment style: circumplex descriptions, recalled developmental experiences, self-representations and interpersonal functioning in adulthood. J Pers 2003;71:141–81.[Medline]
  32. Brennan KA, Clark CL, Shaver PR. Self-report measurement of adult attachment: an integrative overview. In: Simpson JA, Rholes WS, eds. Attachment Theory and Close Relationships. New York: Guilford; 1998:46–76.
  33. Fraley RC, Shaver PR. Adult romantic attachment: theoretical developments, emerging controversies, and unanswered questions. Rev Gen Psychol 2000;4:132–54.[CrossRef]
  34. Mikulincer M, Shaver PR, Pereg D. Attachment theory and affect regulation: the dynamics, development, and cognitive consequences of attachment-related strategies. Motiv Emotion 2003;27:77–102.[CrossRef]
  35. Bartholomew K, Horowitz LM. Attachment styles among young adults: a test of a four-category model. J Pers Soc Psychol 1991;61:226–44.[CrossRef][Medline]
  36. Allen JP, Land D. Attachment in adolescence. In: Cassidy J, Shaver P, eds. Handbook of Attachment: Theory, Research, and Clinical Applications. New York: Guilford; 1999:319–35.
  37. Collins WA, Sroufe LA. Capacity for intimate relationships: a developmental construction. In: Furman W, Brown B, eds. The Development of Romantic Relationships in Adolescence. Cambridge Studies in Social and Emotional Development. New York: Cambridge University Press; 1999:125–47.
  38. Allen JP, Marsh P, McFarland C, McElhaney KB, Land DJ, Jodl KM, Peck S. Attachment and autonomy as predictors of the development of social skills and delinquency during midadolescence. J Consult Clin Psychol 2002;70:56–66.[CrossRef][Medline]
  39. Scharf M, Mayseless O, Kivenson-Baron I. Adolescents' attachment representations and developmental tasks in emerging adulthood. Dev Psychol 2004;40:430–44.[CrossRef][Medline]
  40. Hazan C, Zeifman D. Sex and the psychological tether. In: Bartholomew K, Perlman D, eds. Attachment Processes in Adulthood. Advances in Personal Relationships, vol 5. London: Jessica Kingsley Publishers; 1994:151–78.
  41. Mikulincer M, Gillath O, Shaver PR. Activation of the attachment system in adulthood: threat-related primes increase the accessibility of mental representations of attachment figures. J Pers Soc Psychol 2002;83:881–95.[CrossRef][Medline]
  42. Fraley RC, Davis KE. Attachment formation and transfer in young adults' close friendships and romantic relationships. Pers Relatsh 1997;4:131–44.[CrossRef]
  43. Hazan C, Shaver P. Romantic love conceptualized as an attachment process. J Pers Soc Psychol 1987;52:511–24.[CrossRef][Medline]
  44. Creasey G, Ladd A. Generalized and specific attachment representations: unique and interactive roles in predicting conflict behaviors in close relationships. Pers Soc Psychol Bull 2005;31:1026–38.[Abstract]
  45. Simpson JA, Rholes WS, Phillips D. Conflict in close relationships: an attachment perspective. J Pers Soc Psychol 1996;71:899–914.[CrossRef][Medline]
  46. Cassidy J, Kobak RR. Avoidance and its relation to other defensive processes. In: Belsky J, Nezworski T, eds. Clinical Implications of Attachment. Child Psychology. Hillsdale, NJ: Lawrence Erlbaum; 1988:300–23.
  47. Kamarck TW, Janicki DL, Shiffman S, Polk DE, Muldoon MF, Liebenauer LL, Schwartz JE. Psychosocial demands and ambulatory blood pressure: a field assessment approach. Physiol Behav 2002;77:699–704.[CrossRef][Medline]
  48. Carver CS. Adult attachment and personality: converging evidence and a new measure. Pers Soc Psychol Bull 1997;23:865–83.[Abstract]
  49. Fraley RC, Waller NG. Adult attachment patterns: a test of the typological model. In: Simpson JA, Rholes WS, eds. Attachment Theory and Close Relationships. New York: Guilford; 1998:77–114.
  50. Goodie JL, Kamarck TW. Use of the Accutracker DX to assess behavioral influences on ambulatory cardiovascular activity: a preliminary investigation. Blood Press Monit 1996;1:135–40.[Medline]
  51. Bryk AS, Raudenbush SW. Hierarchical Linear Models: Applications and Data Analysis Methods. Newbury Park, CA: Sage; 1992.
  52. Raudenbush S, Bryk A, Cheong YF, Congdon R. HLM 5: Hierarchical Linear and Nonlinear Modeling. Lincolnwood, IL: Scientific Software International; 2000.
  53. Nezlek JB. Using multilevel random coefficient modeling to analyze social interaction diary data. J Soc Pers Relat 2003;20:437–69.[Abstract]
  54. Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, Koss MP, Marks JS. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. Am J Prev Med 1998;14:245–58.[CrossRef][Medline]
  55. Russek LG, Schwartz GE. Feelings of parental caring predict health status in midlife: a 35-year follow-up of the Harvard Mastery of Stress Study. J Behav Med 1997;20:1–13.[CrossRef][Medline]
  56. Rice KG, Mulkeen P. Relationships with parents and peers: a longitudinal study of adolescent intimacy. J Adolesc Res 1995;10:338–57.[Abstract]
  57. Demir M, Urberg KA. Friendship and adjustment among adolescents. J Exp Child Psychol 2004;88:68–82.[CrossRef][Medline]
  58. Baumeister RF, Leary MR. The need to belong: desire for interpersonal attachments as a fundamental human motivation. Psychol Bull 1995;117:497–529.[CrossRef][Medline]
  59. Parker JG, Asher SR. Peer relations and later personal adjustment: are low-accepted children at risk? Psychol Bull 1987;102:357–89.[CrossRef][Medline]
  60. Cohen S, Gottlieb BH, Underwood LG. Social relationships and health: challenges for measurement and intervention. Adv Mind Body Med 2001;17:129–41.[Medline]
  61. Berkman LF, Glass TA. Social integration, social networks, social support, and health. In: Berkman L, Kawachi I, eds. Social Epidemiology. New York: Oxford University Press; 2000:137–73.
  62. Creasey G, Ladd A. Negative mood regulation expectancies and conflict behaviors in late adolescent college student romantic relationships: the moderating role of generalized attachment representations. J Res Adolesc 2004;14:235–55.[CrossRef]
  63. Mikulincer M. Adult attachment style and individual differences in functional versus dysfunctional experiences of anger. J Pers Soc Psychol 1998;74:513–24.[CrossRef][Medline]
  64. Staessen JA, Thijs L, Fagard R, O'Brien ET, Clement D, de Leeuw PW, Mancia G, Nachev C, Palatini P, Parati G, Tuomilehto J, Webster J. Predicting cardiovascular risk using conventional vs ambulatory blood pressure in older patients with systolic hypertension. Systolic Hypertension in Europe Trial Investigators. JAMA 1999;282:539–46.[Abstract/Free Full Text]
  65. Verdecchia P. Prognostic value of ambulatory blood pressure: current evidence and clinical implications. Hypertension 2000;35:844–51.[Abstract/Free Full Text]
  66. Pereg D, Mikulincer M. Attachment style and the regulation of negative affect: exploring individual differences in mood congruency effects on memory and judgment. Pers Soc Psychol Bull 2004;30:67–80.[Abstract]
  67. Mikulincer M, Horesh N. Adult attachment style and the perception of others: the role of projective mechanisms. J Pers Soc Psychol 1999;76:1022–34.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Psychosom. Med.Home page
W. M. Troxel, J. M. Cyranowski, M. Hall, E. Frank, and D. J. Buysse
Attachment Anxiety, Relationship Context, and Sleep in Women With Recurrent Major Depression
Psychosom Med, September 1, 2007; 69(7): 692 - 699.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gallo, L. C.
Right arrow Articles by Matthews, K. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gallo, L. C.
Right arrow Articles by Matthews, K. A.
Related Collections
Right arrow Blood Pressure
Right arrow Social Support

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS