School-aged Children’s Psychobiological Divergence as a Prospective Predictor of Health Risk Behaviors in Adolescence


Recent attachment research suggests that children with avoidant attachment often underreport their psychological distress compared to their physiologic indicators of distress (neuroendocrine reactivity, startle response, event-related potentials). This pattern of behavior (referred to as psychobiological divergence) may confer risk for suboptimal coping behaviors, including substance use, sexual risk-taking, and non-suicidal self-injury (NSSI), because individuals who are not aware of or cannot express their emotional needs may engage in maladaptive strategies to regulate their emotions. In the current pilot study (N = 45 youth), we investigate whether psychobiological divergence of neuroendocrine and self-reported reactivity in middle childhood prospectively predicts health risk behaviors (HRBs) in adolescence. The results revealed that divergence was significantly associated with adolescents’ substance use and non-suicidal self-injury (NSSI), but not with their sexual behavior. Among adolescents currently reporting low levels of attachment security, divergence was associated with greater self-reported NSSI. Our results provide initial evidence that psychobiological divergence confers risk for substance use and NSSI in combination with current relational distress. We discuss the implications of our findings for adolescent development and clinical risk.


Attachment theorists predict that the quality of children’s early relationships with caregivers has profound impacts on health throughout the lifespan (Allen and Land 1999; Dozier et al. 2008; Feeney 2000; Greenberg 1999). Secure attachment relationships are thought to promote healthy emotion regulation, decrease stress reactivity, and encourage health-promoting behaviors (Feeney 2000; Pietromonaco et al. 2013), all of which may have downstream health-promoting effects. Though limited empirical research has linked attachment in middle childhood with healthy behaviors in adolescence and adulthood, studies indicate that secure attachment, which is associated with consistent and sensitive caregiving, serves as the foundation for healthy adolescent development (Allen and Land 1999) and promotes a lower level of engagement in health risk behaviors (HRBs) such as alcohol and drug use, non-suicidal self-injury (NSSI), and sexual behavior that can lead to unintended pregnancy and/or sexually transmitted infections (Moretti and Peled 2004). Conversely, insecure attachment in childhood, predicted by exposure to insensitive parenting, forecasts higher levels of physical and mental illness in adulthood (Dozier et al. 2008; McWilliams and Bailey 2010; Puig et al. 2013) and is associated with HRBs in adolescence (Allen and Land 1999; Cooper et al. 1998; Martin et al. 2017; Schindler and Bröning 2015). As a stage during which many behavioral and emotional regulatory processes are in flux, adolescence is a time ripe for the onset of maladaptive coping, including engagement in HRBs (Smith and Cyders 2016; Wills et al. 2006).

According to theory, early experiences with caregivers inform the development of the internal working model (IWM), a framework regarding relationships that impacts emotion regulation (Bowlby 1999; Cassidy 1994). Securely attached children are more emotionally open because their caregivers have demonstrated that they may be relied upon for timely and appropriate emotional attention (Ainsworth et al. 1978). The trust between caregiver and child also allows secure children to feel comfortable enough to explore their environments and return to their caregivers for comfort when distressed (Ainsworth 1979; Bowlby 1999). Conversely, insecurely attached children do not exhibit this openness, ostensibly because their attachment bids have been met with rejection by their caregivers. The most common form of attachment insecurity, attachment avoidance (Ainsworth et al. 1978), is thought to result from consistently rejecting parental behavior. As an adaption to this type of caregiving, theorists contend that avoidant children adopt the strategy of deactivating their attachment systems and suppressing their negative emotions (Ainsworth et al. 1978; Cassidy 1994). Attachment anxiety, another manifestation of insecure attachment, is thought to develop when caregivers are inconsistently responsive, and can result in the hyperactivation of expressed attachment needs and negative emotion in children (Cassidy 1994; Mikulincer and Shaver 2003). Readers should note that the terms used by attachment researchers to describe different types of attachment vary by measurement tool used to assess attachment, as well as by the developmental stage of the participants. For the sake of clarity, in this paper, we use the terms avoidance and anxiety to refer to these dimensions of attachment insecurity – importantly, some studies described employ self-reported measures of attachment whereas others employ coder-rated behavioral observations, the distinctions between which many attachment researchers consider meaningful.

Deactivation and hyperactivation of emotion can be adaptive in the short-term (Cassidy 1994; Main 1981). For example, suppression of negative emotion may protect infants from further rejection, while minimization of the importance of the attachment relationship allows the infant to avert attention from a source of relational conflict (Cassidy 1994). Evidence suggests that adults with greater attachment avoidance demonstrate less performance interference on Stroop tasks containing threatening attachment-related words, perhaps providing potential evidence of the protective benefits of deactivation (Mikulincer et al. 2004). Conversely, hyperactivation is thought to represent the child’s greatest chance of keeping an inconsistently responsive caregiver engaged with and available to the child (Cassidy and Berlin 1994; Cassidy 1994).

However, both of these emotion regulation strategies likely have long-term costs. The performance of avoidant adults deteriorates upon introduction of a cognitive load or when in combination with stressful life events that exceed one’s ability to cope (Fagundes et al. 2012; Mikulincer et al. 2004; Sbarra and Borelli 2013). And adults high in attachment anxiety face the consequences of chronic activation—they have higher rates of emotion dysregulation, distress, and psychopathology (Caldwell and Shaver 2012; Mikulincer and Shaver 2003; Reynolds et al. 2014). Outside of the field of attachment, investigations of different emotional regulation strategies have found similar hidden costs of two strategies similar to those employed by avoidant and anxious individuals. For instance, emotional suppression, a construct similar to deactivation, results in poorer memory (Richards and Gross 2000), increased sympathetic activation (Gross and Levenson 1997), greater experiences of negative emotion, worse interpersonal functioning, and worse well-being (Gross and John 2003). Similarly, rumination, which is similar to hyperactivation, is a well-known risk factor for worsening depression in youth and adults (Compas et al. 2004; Hankin and Abramson 2001; Kovacs et al. 2008). In contrast, flexibility in the use of different regulatory strategies predicts the best emotional adjustment (Bonanno et al. 2004). Thus, although deactivation/hyperactivation may confer short-term relational benefits, over the long-term, and particularly in conjunction with stressors, they may heighten risk for maladaptation.

The majority of research thus far has focused on deactivation, due to the fact that avoidant attachment is more common, and thus easier to investigate, than anxious attachment (Cassidy and Berlin 1994). As a result, in the remainder of our review, we focus most centrally on avoidance and deactivation, while noting that hyperactivation may also confer risk for negative consequences. Deactivation and hyperactivation have been operationalized in the literature by comparing physiologic indicators of stress such as heart rate, galvanic skin response, and salivary cortisol levels with self-reported measures of distress. For instance, studies document that avoidant infants often have greater neuroendocrine reactivity, operationalized as the difference in salivary cortisol levels pre- to post-stressor, in response to separations than do secure infants (Gunnar 1991; Spangler and Grossmann 1993; Spangler 1998), and yet show fewer behavioral signs of distress when the caregiver returns (Ainsworth et al. 1978). Among adults, converging evidence suggests that avoidant individuals display greater stress reactivity than do secure individuals but report less distress (Dozier and Kobak 1992; Spangler and Grossmann 1993). Recently, researchers have identified that children with avoidant attachment have both greater neuroendocrine reactivity than secure children and underreport their distress (Borelli et al. 2014). In other words, these children exhibit what scholars refer to as psychobiological divergence—discordance between physiologic and self-reported indicators of emotionality. Psychobiological divergence can take the form of either high physiologic/low self-reported (deactivation) or low physiologic/high self-reported (hyperactivation) responses. The link between divergence (high physiologic/low self-reported) and attachment has since been replicated with other measures of reactivity, including startle response (Borelli et al. 2013), event-related potentials (White et al. 2012), and anxious non-verbal behavior (Borelli et al. 2017). This work is grounded in the assumption that positive psychobiological divergence, in which one’s physiological arousal outpaces self-reported negative emotion, suggests suppression or lack of awareness of aversive bodily states. Conversely, negative divergence, in which youth’s self-reported negative emotion exceeds their physiological arousal, can result in negative outcomes via the persistent focus on distress.

Given that a number of studies consistently demonstrate a link between avoidant child attachment and deactivation, the next question that emerges is, what are the consequences of psychobiological divergence? What, if anything, happens to the physiological arousal that is not endorsed by avoidant youth? What happens to youth who are inflexibly focused or hyperattuned to their physiological states, such as those with hyperactivating strategies? In the case of deactivation, one possibility is that failure to take steps to directly address aversive arousal leads youth to engage in unhealthy behaviors, including HRBs (e.g., substance use, non-suicidal self-injury, and sexual risk behaviors), to externally regulate themselves. This hypothesis is consistent with theories regarding risky behaviors—the self-medication theory of substance use (Khantzian 1985) posits that individuals may use substances to relieve emotional or physical pain, and researchers similarly posit that non-suicidal self-injury (NSSI) may function as an emotion regulatory outlet when more adaptive means of regulating emotion are unavailable (Gratz 2003).

In the case of hyperactivation, youth who are acutely sensitive to minor fluctuations in physiological arousal may experience a persistent state of subjective malaise, which could cause a sense of frustration, hopelessness, or desperation. Youth in this state could resort to HRBs in an attempt to reduce unpleasant psychological states or solicit the interpersonal support they desire. Alternatively, perhaps hyperactivating youth seek to make physical what exists primarily in the realm of the psychological—in other words, to find a way to embody their subjective experience. This unconscious desire to express physically their subjectivity could lead these children to engage in behaviors that are harmful for their health.

To date no studies have examined the association between psychobiological divergence and HRBs; however, studies have documented associations between insecure attachment, which itself is associated with divergence, and HRBs. Avoidant attachment is associated with substance use in adults (Becoña Iglesias et al. 2014; Schindler et al. 2009; Schindler et al. 2007; Thorberg and Lyvers 2010) and many studies have linked avoidant attachment with substance use in adolescence. Although insecure attachment in general is associated with substance use in adolescence (Schindler and Bröning 2015), to our knowledge no studies report associations between anxious attachment and substance use.

The link between attachment and sexual behavior is less clear. From a theoretical perspective, insecure individuals may engage in risky sexual activity that may lead to sexually transmitted infections and/or unintended pregnancy as a means of coping with relational distress or negative emotion. Extant research suggests that avoidant adults are less likely to engage in sexual activity in order to avoid intimacy (Feeney, Tracy et al. 2003); however, among sexually active adults, those who are avoidant endorse having intercourse outside of committed romantic relationships (Brennan and Shaver 1995; Feeney et al. 1993). Some studies find that both avoidant and anxious adolescents are more likely than secure ones to engage in risky sexual behaviors, such as having sex with a stranger (Cooper et al. 1998) or having casual sex while in a committed relationship (Scharfe and Eldredge 2001). Other studies report that anxious adults are less likely to engage in casual sex (Brennan and Shaver 1995), yet are also less likely to engage in safe sexual behavior, such as using contraception, than secure and avoidant adults (Feeney et al. 1999). Therefore, while both avoidant and anxious adolescents may engage in sexual risk taking, the form that risk-taking may take varies by attachment classification.

Non-suicidal self-injury (NSSI) is intentional, self-inflicted injury to the body by methods such as cutting, hitting, or burning (Heath and Nixon 2009) and is often used as a means of emotion regulation (Adrian et al. 2011; Martin et al. 2016). Adolescents engaging in NSSI endorse reasons for doing so such as releasing tension, alleviating negative emotional states, and stopping dissociative or depersonalizing episodes (Klonsky 2007). Given the high level of emotional dysregulation among insecure adolescents, there may be a strong association between insecure attachment and NSSI. Prior work suggests that anxious, but not avoidant attachment is associated with NSSI (Martin et al. 2017), but other work suggests that avoidant coping strategies in adolescents and young adults are strongly associated with NSSI (Chapman et al. 2006; Gratz 2003).

In the current investigation, we build on previous work by examining the prospective link between psychobiological divergence measured in middle childhood and HRBs (NSSI, substance use, sexual behavior) in adolescence. First, as a replication of prior work, we test whether HRBs in adolescence are concurrently associated with greater self-reported attachment insecurity (Hypothesis 1). Next we examine the prospective physiological and subjective predictors of HRBs, by a) assessing whether neuroendocrine reactivity and self-reported distress in response to a laboratory stressor paradigm measured during middle childhood are correlated with HRBs adolescence (Hypothesis 2), and b) testing whether psychobiological divergence is associated with greater endorsement of HRBs (Hypothesis 3). Finally, grounded in a diathesis stress framework, we evaluate the hypothesis that divergence in middle childhood will be more strongly associated with HRBs in adolescence when adolescent attachment insecurity is high (Hypothesis 4). Because we anticipate that both forms of divergence (positive divergence: high physiologic/low self-report and negative divergence: low physiologic/high self-report) will be associated with greater risk for HRBs, albeit via different mechanisms, we test for the presence of both linear and curvilinear effects.



Children were initially recruited from the local community through flyers and Internet postings ((N = 106; M age = 9.8 years, SD = 1.4 years; Time 1 [T1]); the original sample was lower income and ethnically/racially diverse (see Borelli et al. 2014, for more details). The study was initially designed to be cross-sectional; therefore, no measures were taken to stay in contact with the participants. Approximately 6 years later, we obtained additional funding to follow-up with the original sample, which we did by phone and email. Due to the cross-sectional design of the original study, we had not instituted any procedures for staying in touch with participants, leading to high attrition rates in the follow-up. A total of 40% of the original study participants (n = 45; 58% male; M age = 15.2 years, SDage = 1.4 years) completed the follow-up study. The Time 2 (T2) sample was ethnically (48.5% Caucasian, 27.3%, Hispanic, 18.2% African American, 3% Asian, and 3% Native American) and educationally diverse (39.4% of participants’ primary caregiver did not have a college degree), and did not significantly differ from the T1 sample in divergence, race/ethnicity, income, or gender.


During the T1 data collection, after providing consent(parent)/assent(children), youth completed a self-report measure of attachment to their mothers and a standardized laboratory stressor (described below), before and after which they provided saliva samples and reports of their emotional valence. Approximately 6 years later, after providing consent(parent)/assent(youth), the youth reported on their attachment security to their mothers and their HRBs.

Time 1 Measures

Stressor paradigm

The Vulnerability Vignettes Paradigm (VVP; Borelli et al. 2014), was administered in order to measure neuroendocrine reactivity in response to an attachment related laboratory stressor. In short, thirteen two-sentence vignettes depicting attachment related themes such as sadness, fear, loss, and rejection (e.g. Taylor did not get picked to be on a sports team at school and he felt sad”) and two neutral (control) vignettes were presented to children on a computer screen in one of two random order conditions after a brief acclimation period. As reported previously, children’s self-reported distress was significantly higher in response to the vulnerability vignettes as opposed to the neutral vignettes, suggesting that the experimental manipulation was successful in inducing negative emotion (Borelli et al. 2014).

Self-reported distress

Self-reported distress was assessed after each vulnerability vignette via the single-item Self Assessment Manikin of emotional valence (SAM; Bradley and Lang 1994). The SAM is a pictorial rating system that depicts cartoons figures expressing various levels of emotion. The pictures were also displayed with descriptors such as “happy, pleased, and good” next to the most positive figure and “very unhappy, scared, and sad” next to the most negative figure to assist in the self-assessment. The figures corresponded to a 1–5 rating system, with higher scores indicating more negative emotion.

Neuroendocrine reactivity

We measured salivary cortisol levels from saliva samples collected using standard procedure both before (Time A) and 15 min after VVP administration (Time B). Cortisol levels were measured in duplicate using a fluorescent enzyme-linked immunosorbent assay (ELISA) technique, with a 96-well plate coated in monoclonal cortisol antibodies (Salimetrics, State College, PA). Cortisol levels were expressed as micrograms per deciliter and time of collection was recorded and used as a covariate in subsequent analyses (see MASKED). We were missing data from n = 4 participants due to insufficient saliva.

Time 2 (T2) Measures

Alcohol and substance use

The Adolescent Alcohol and Drug Involvement Scale (AADIS; Moberg 2003), is a 14-item measure that was developed as a research and clinical tool in order to measure the level of alcohol and drug involvement in adolescents at risk for or suspected of substance use/abuse. The scale is an adaptation of the Adolescent Alcohol Involvement Scale (Mayer and Filstead 1979) which demonstrated strong reliability and validity (Mayer and Filstead 1979). Participants rate items on an 8-point scale based on how often they have used 14 different substances (ranging from never used to several times a day), with higher scores indicating greater frequency of substance use behavior. The AADIS had been validated with juveniles in a state correctional facility (Winters et al. 2001).

Non-suicidal self-injury (NSSI)

To measure the frequency of NSSI, we asked adolescents to complete the Functional Assessment of Self-mutilation (Lloyd et al. 1997). On this self-report measure, adolescents report on the frequency with which they engaged in 11 types of self-injurious behavior (e.g., “cutting/burning/scraping skin,” “picking at a wound,” “biting or hitting oneself,” “inserting objects under skin,” “hair pulling”) over the past year. Additional items explore the functional reasons underlying adolescents’ NSSI. Due to our interest in frequeny of NSSI, in the current study we utilized only participants’ endorsement of NSSI. We created a total score of the number of NSSI behaviors reported by each participant.

Sexual behavior

The Sexual Risk Behavior Inventory (SRBI) is a 2-item measure adapted for the purpose of this study based on previous work (Luster and Small 1994) defining high-risk sexual behavior as having multiple partners and not using contraception frequently. Using a 0 to 5 scale, with higher scores suggesting a greater number of partners, adolescents indicate the number of partners with whom they have engaged in sexual intercourse and with whom they have failed to use protection. Here we computed a mean sexual risk behavior score by averaging participants’ scores on the two questions.

Time 1 and Time 2: Children’s Attachment

Children’s attachment security was measured via the Security Scale (Kerns et al. 2001), a self-report measure of attachment with well-established psychometric properties (Diener et al. 2008; Lieberman et al. 1999), at T1 and T2. The Security Scale prompts children to assess the extent to which they believe 15 aspects of parent–child relationships are characteristic of their relationship with attachment figures (in this case, with respect to their mothers/mother figures). Each item is scored on a scale from 1 to 4, with higher scores suggestive of greater security (e.g., Some kids find it easy to trust their mom [really true for me, sort of true for me] but other kids are not sure if they can trust their mom [really true for me, sort of true for me]). Cronbach’s alphas were good, α T1 = 0.86 and α T2 = 0.88.

Data Analyses

Based on the desire to retain all possible data, we used multiple imputation (20 imputation iterations) of T2 variables, enabling us to have a sample of n = 43 in all analyses. AADIS, NSSI, and SRBI scores were positively skewed; therefore, we log transformed them, which resulted in normally distributed variables. We controlled for T2 age in analyses because our bivariate correlation revealed that T2 age was strongly associated with AADIS scores (Table 2). To test for the possibility that both positive and negative divergence scores are associated with HRBs, we tested for linear and curvilinear effects. In these analyses, using a standardized residualized change score, we created independent variables that controlled for relevant covariates (participant age, and in the case of T2 NSSI, T1 NSSI).


As compared to boys, girls reported significantly higher attachment security at T2, t(42) = 2.12, p = .03 (see Table 1). Zero-order correlations revealed that neuroendocrine reactivity was significantly associated with AADIS scores, r = .42, p = 0.01 (see Table 2).

Table 1 Descriptive statistics for key study variables for total sample and by child sex
Table 2 Zero-order correlations for key study variables

Concurrent Associations between Attachment Security and HRBs

Partial correlations controlling for children’s age revealed that attachment security was significantly associated with lower substance use, r = –.29, p = .03, and lower NSSI, r = –.43, p = .003, but not with sexual behavior, r = –.14, p = .41.

Prospective Association between Neuroendocrine Reactivity, Self-Reported Distress, and HRBs

After controlling for children’s age, the step containing our reactivity variables (neuroendocrine reactivity and mean self-reported distress) significantly added to the prediction of youth substance use, ∆R 2 = .19, p = .012. Neuroendocrine reactivity, b = 1.03, p = .004, but not self-reported reactivity, b = –.24, p = .59, was associated with higher AADIS scores (Table 3). These findings remained significant after controlling for T1 attachment.

Table 3 Hierarchical regressions predicting adolescent risky behavior

We conducted regressions examining the contribution of children’s T1 reactivity to the prediction of their NSSI and sexual behavior—none were significant predictors (see Table 3).

Associations between Divergence and HRBs

We probed for both linear and curvilinear effects in these analyses. First, we found that controlling for participant age, divergence exhibited a significant linear effect on T2 youth substance use, R 2 = .14, p = .02, b = .89, with a marginal curvilinear effect, R 2 = .16, p = .05, b 1 = .80, b 2 = .21. The linear effect suggested that higher divergence scores predicted greater substance use, whereas the non-significant curvilinear effect suggested that substance use trended toward being higher among youth with the lowest divergence scores. A follow-up regression in which we controlled for the main effects of neuroendocrine reactivity and self-reported reactivity, as well as age, revealed that psychobiological divergence remained a significant, curvilinear effect, R 2 = .32, p = .004, b 1 = .48, b 2 = 3.83, but the linear effect dropped below significance, p = .16. This latter analysis suggested that substance use was highest among youth with the lowest and highest divergence.

With respect to NSSI, first we explored whether after controlling for age, divergence was associated with T1 NSSI: Results did not reveal a significant linear, R 2 = .08, p = .06, b = –.05, or curvilinear association. The direction of effects suggested that lower divergence scores trended toward an association with higher NSSI. With respect to the prediction of T2 NSSI, in a regression controlling for participant age and T1 NSSI, we found a significant linear association between divergence and T2 NSSI, R 2 = .20, p = .04, b = .30, with effects suggesting that higher divergence scores predict higher NSSI at T2. We did not find a significant curvilinear association, R 2 = .08, p = .18. A follow-up regression in which we controlled for neuroendocrine reactivity and self-reported reactivity, as well as age, revealed that psychobiological divergence remained a significant linear, R 2 = .33, p = .02, b = 2.53, but not a significant curvilinear predictor of T2 NSSI, R 2 = .34, p = .07, b 1 = 1.57, b 2 = 3.02.

In contrast, after controlling for children’s age, psychobiological divergence was not a significant linear, R 2 = .36, p = .05, b = −.23, or curvilinear predictor of sexual risk behaviors, R 2 = .39, p = .14, b 1 = −.22, b 2 = –.03.

Interaction between Middle Childhood Divergence and Adolescent Attachment Insecurity in Predicting HRBs

Finally, we explored whether the association between children’s T1 divergence and their T2 HRBs was moderated by youth’s T2 levels of attachment security. In these analyses, we controlled for children’s age, T1 reactivity variables, and T1 self-reported attachment security on initial steps, using T1 divergence scores as the predictor variables. Note that with respect to all of these analyses, we examined linear effects only, though additional analyses involving curvilinear effects were not significant.

The interaction between T2 attachment and T1 divergence was not a significant predictor of their substance use, ∆R 2 = .01, p = .45, or their sexual behavior, ∆R 2 = .01, p = .62. However, the interaction between T2 attachment and T1 divergence was significantly associated with T2 NSSI, controlling for T1 NSSI, ∆R 2 = .08, p = .03 (see Table 4). We decomposed the simple slopes using PROCESS Model 1 for SPSS, which suggested that the association between attachment security and T2 NSSI was only significant when divergence was medium, b = –.43, p = .025, and high, b = –.82, p = .001, but not low, b = –.04, p = .89 (Fig. 1). Reversal of the independent variable and the moderator revealed that only when attachment security was low was divergence prospectively associated with greater T2 NSSI, b = .29, p = .04.

Table 4 Hierarchical regression predicting adolescent T2 NSSI
Fig. 1

T1 Psychobiological divergence moderates the association between adolescents’ T2 attachment security with their mothers and their T2 NSSI, controlling for demographics and T1 attachment security and NSSI


We examined the link between deactivation/hyperactivation, measured via psychobiological divergence, and HRBs in adolescence. Although it has been theorized that these regulatory strategies confer risk for HRBs in adolescence and adulthood, to our knowledge, no studies have tested this hypothesis. Therefore, the purpose of the current investigation was to test an empirical connection between divergence, as measured the relationship between neuroendocrine reactivity and self-reported distress in response to an attachment-related laboratory stressor, and HRBs in adolescence. Our results generally suggest that positive divergence, in which physiologic reactivity is high and self-reported reactivity is low, is associated with greater risk for both substance use and NSSI in adolescence. Positive divergence may suggest underestimation of physiological response by self-reported response.

Adolescents who previously had shown positive divergence were more likely to have used substances and engaged in NSSI, supporting our hypothesis that positive divergence can create risk for problems in other domains. The causes of positive divergence—whether children are unaware of their emotional states or, although aware, are unable or not willing to express vulnerability—are unclear. We cannot determine the underlying reason driving positive divergence, but either of these explanations could contribute to HRBs. If children are not aware of their emotional states, they may be less likely to receive support when distressed. Although putatively adaptive within the avoidant parent-child relationship (Main 1981), deactivation may limit the support an avoidant child can receive from other people in their lives. Perhaps avoidant children may act out underlying distress even if they may not be explicitly aware of their distress, by engaging in behaviors that may soothe physiological arousal (such as substance use or NSSI), obviating the need for internal or external recognition of mental states. Alternatively, if avoidant children do not feel safe enough within the context of the parent–child relationship to express vulnerability, they may engage in HRBs as a means of regulating affective states (Khantzian 1985). In this way, avoidant children may use substances as a way to relieve emotional pain they are not able to express.

In contrast, no significant association was found between divergence and sexual behavior; however, at a trend level of significance (p = .05), divergence was negatively associated with sexual risk behavior. Perhaps in a larger sample, this effect would reach the level of significance. If this finding were replicated, this would suggest that children whose self-reported negative emotion was high while their cortisol activation was low are more prone to engage in sexual risk behavior. It is also noteworthy that in our initial analysis examining predictors of substance use, we found a trend-level curvilinear effect (p = .05), which in line with the sexual behavior finding, suggests that children with negative divergence also exhibit higher substance use in adolescence.

These two findings suggest the need to examine these research questions in larger samples and the need to think more deeply about the meaning of hyperactivation. When conceptualizing hyperactivation in the context of the divergence metric, we had initially assumed that a hyperactivating strategy would entail a high level of subjective distress in the presence of lower levels of physiological reactivity. In other words, we theorized that hyperactivating children would show heightened sensitivity to physiological reactivity, similar to the heightened sensitivity they show to attachment-related threats (Bowlby 1999). However, we also acknowledge that higher initial subjective distress could lead to higher physiological reactivity over time (for instance, we imagine that when children engage in rumination or worry, they are likely to become physiologically activated as a result of this regulatory process). In the context of slow-moving cortisol reactivity and the broad assessment window tapped in the current study, it seems unlikely that children using hyperactivating emotion regulation would maintain lower levels of cortisol reactivity across the entire task. Thus, perhaps when assessing hyperactivation, it is important to measure physiologic indicators tapping moment-to-moment arousal (such as ERPs or cardiovascular reactivity), rather than cortisol reactivity, which provides a relatively non-specific measure of arousal across a longer window of time. Using physiologic indicators that are more precise in their measurement would allow the assessment of arousal early in a stress exposure, which could ultimately provide a more accurate assessment of hyperactivation. The refinement of the operationalization of negative divergence could lead to a more precise evaluation of the link between hyperactivation and HRBs.

Finally, our last goal in the study was to examine whether current levels of relational adversity were associated with greater HRBs in the presence of divergence. This hypothesis was supported only with respect to NSSI—among adolescents reporting low and mean levels of attachment security at T2, greater divergence was associated with greater NSSI. This effect qualifies our prior finding: it is only among those adolescents reporting higher current levels of insecurity that the association between divergence and NSSI holds. In contrast, however, youth’s divergence is positively associated with substance use regardless of their self-reported attachment security. The differences in the links between divergence and different risk behaviors should be explored in future studies with greater statistical power.

In making sense of these findings, we seek to place them in developmental context. Marked by a tendency towards increased emotional volatility, negative affect, greater participation in risky behavior, and more difficulty inhibiting the self when highly emotional (Cyders and Smith 2008), adolescence is an especially critical period for the impact of emotion regulation on behavior. It may be that difficulties in emotional regulation during a time of greater experiences of emotional turbulence may lead to the adoption of more extreme methods of external regulation. Indeed, adolescents and young adults with lower distress tolerance are more likely to engage in risky behavior, such as substance use or unsafe sexual practices (Daughters et al. 2009; MacPherson et al. 2010; Magar et al. 2008). Similar to neurotic individuals, who engage in risky drinking and sexual behaviors to regulate their high levels of distress (Cooper et al. 2000), adolescents lower in distress tolerance could use risky behavior to displace unbearable emotional and bodily sensations with those that are more pleasurable. Indirect support of this theorizing comes in the form of research demonstrating that alexithymia, difficulty identifying and describing one’s own feelings, is associated with adolescents’ risky behavior, including substance use (Bonnet et al. 2013; Dorard et al. 2008; Zimmermann 2010). For individuals with poor insight in understanding and identifying their specific negative emotional states, it may be easier to understand experiences in terms of physiological rather than psychological states.

Strengths, Limitations, and Future Directions

A central strength of this pilot study is that it explored the connection between deactivation and HRBs; that we explored these hypotheses longitudinally in a diverse sample further strengthens the study’s contributions.

It is also important to note the study’s limitations. The first limiting factor is the relatively small sample, limiting our ability to detect effects. Further, due to the fact that the adolescent follow-up was not planned at the outset of the study, we did not engage in efforts to stay in contact with the families who participated at T1. In this lower income and diverse sample, this lack of contact resulted in a high rate of attrition from T1 to T2, and although the T1 and T2 samples did not differ from one another on many dimensions, it is possible that those families who dropped out were those in which the adolescents were at greater risk.

In addition, the measures we used in this study are limiting. For instance, the AADIS was originally developed to measure the level of substance use behavior in populations at risk for or suspected of substance use disorders. Our sample was non-clinical and thus we found limited variability in AADIS scores. Although we conducted a data transformation to help normalize the distribution of these data, the limited variability may have impacted our ability to find effects. Further the Sexual Risk Behavior Inventory only consists of two items. It will be important to follow-up this investigation with a more thorough assessment of HRBs.

Future studies should extend our preliminary findings to a larger sample. Researchers may wish to measure other problems that increase in adolescence, to extend the connection between deactivation in middle childhood to a broader range of behavior in adolescence. In conclusion, with this pilot study we offer initial evidence of a link between psychobiological divergence and HRBs in adolescence. Though preliminary and limited in scope, our findings provide the foundation for future work on this topic.


  1. Adrian, M., Zeman, J., Erdley, C., Lisa, L., & Sim, L. (2011). Emotional dysregulation and interpersonal difficulties as risk factors for nonsuicidal self-injury in adolescent girls. Journal of Abnormal Child Psychology, 39(3), 389–400.

    Article  PubMed  Google Scholar 

  2. Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Oxford, England: Lawrence Erlbaum.

    Google Scholar 

  3. Ainsworth, M. S. (1979). Infant-mother attachment. American Psychologist, 34(10), 932–937.

    Article  PubMed  Google Scholar 

  4. Allen, J. P., & Land, D. (1999). Attachment in adolescence. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 319–335). New York, NY: Guilford Press.

    Google Scholar 

  5. Becoña Iglesias, E., Fernández del Río, F., Calafat, A., & Fernández-Hermida, J. (2014). Attachment and substance use in adolescence: a review of conceptual and methodological aspects. Adicciones, 26(1). 77-86. Retrieved from

  6. Bonanno, G. A., Papa, A., Lalande, K., Westphal, M., & Coifman, K. (2004). The importance of being flexible: The ability to both enhance and suppress emotional expression predicts long-term adjustment. Psychological Science, 15(7), 482–487.

    Article  PubMed  Google Scholar 

  7. Bonnet, A., Bréjard, V., & Pedinielli, J. L. (2013). Emotional dispositions and substance use: mediating effect of alexithymia. Psychological Reports, 112(1), 289–302.

    Article  PubMed  Google Scholar 

  8. Borelli, J. L., David, D. H., Crowley, M. J., Snavely, J. E., & Mayes, L. C. (2013). Dismissing children’s perceptions of their emotional experience and parental care: Preliminary evidence of positive bias. Child Psychiatry & Human Development, 44(1), 70–88.

    Article  Google Scholar 

  9. Borelli, J. L., Ho, L. C., Sohn, L., Epps, L., Coyiuto, M., & West, J. L. (2017). School-aged children’s attachment dismissal prospectively predicts divergence of their behavioral and self-reported anxiety. Journal of Child and Family Studies, 26(4), 1018–1028.

    Article  Google Scholar 

  10. Borelli, J. L., West, J. L., Weekes, N. Y., & Crowley, M. J. (2014). Dismissing child attachment and discordance for subjective and neuroendocrine responses to vulnerability: dismissing attachment and neuroendocrine response. Developmental Psychobiology, 56(3), 584–591.

    Article  PubMed  Google Scholar 

  11. Bowlby, J. (1999). Attachment and Loss. 2nd ed New York, NY: Basic Books. Vol. 1.

    Google Scholar 

  12. Bradley, M. M., & Lang, P. J. (1994). Measuring emotion: the self-assessment manikin and the semantic differential. Journal of Behavior Therapy and Experimental Psychiatry, 25(1), 49–59.

    Article  PubMed  Google Scholar 

  13. Brennan, K. A., & Shaver, P. R. (1995). Dimensions of adult attachment, affect regulation, and romantic relationship functioning. Personality and Social Psychology Bulletin, 21(3), 267–283.

    Article  Google Scholar 

  14. Caldwell, J. G., & Shaver, P. R. (2012). Exploring the cognitive-emotional pathways between adult attachment and ego-resiliency. Individual Differences Research, 10(3), 141–152. Retrieved from

  15. Cassidy, J. (1994). Emotion regulation: influences of attachment relationships. Monographs of the Society for Research in Child Development, 59(2-3), 228–249.

    Article  PubMed  Google Scholar 

  16. Cassidy, J., & Berlin, L. J. (1994). The insecure/ambivalent pattern of attachment: Theory and research. Child Development, 65(4), 971–991.

    Article  PubMed  Google Scholar 

  17. Chapman, A. L., Gratz, K. L., & Brown, M. Z. (2006). Solving the puzzle of deliberate self-harm: The experiential avoidance model. Behaviour Research and Therapy, 44(3), 371–394.

    Article  PubMed  Google Scholar 

  18. Compas, B. E., Connor-Smith, J., & Jaser, S. S. (2004). Temperament, stress reactivity, and coping: Implications for depression in childhood and adolescence. Journal of Clinical Child & Adolescent Psychology, 33(1), 21–31.

    Article  Google Scholar 

  19. Cooper, M. L., Agocha, V. B., & Sheldon, M. S. (2000). A motivational perspective on risky behaviors: The role of personality and affect regulatory processes. Journal of Personality, 68(6), 1059–1088.

    Article  PubMed  Google Scholar 

  20. Cooper, M. L., Shaver, P. R., & Collins, N. L. (1998). Attachment styles, emotion regulation, and adjustment in adolescence. Journal of Personality and Social Psychology, 74(5), 1380

    Article  PubMed  Google Scholar 

  21. Cyders, M. A., & Smith, G. T. (2008). Emotion-based dispositions to rash action: Positive and negative urgency. Psychological Bulletin, 134(6), 807

    Article  PubMed  PubMed Central  Google Scholar 

  22. Daughters, S. B., Reynolds, E. K., MacPherson, L., Kahler, C. W., Danielson, C. K., Zvolensky, M., & Lejuez, C. W. (2009). Distress tolerance and early adolescent externalizing and internalizing symptoms: The moderating role of gender and ethnicity. Behaviour Research and Therapy, 47(3), 198–205.

    Article  PubMed  Google Scholar 

  23. Diener, M. L., Isabella, R. A., Behunin, M. G., & Wong, M. S. (2008). Attachment to mothers and fathers during middle childhood: Associations with child gender, grade, and competence. Social Development, 17(1), 84–101.

    Google Scholar 

  24. Dorard, G., Berthoz, S., Phan, O., Corcos, M., & Bungener, C. (2008). Affect dysregulation in cannabis abusers. European Child & Adolescent Psychiatry, 17(5), 274–282.

    Article  Google Scholar 

  25. Dozier, M., & Kobak, R. R. (1992). Psychophysiology in attachment interviews: Converging evidence for deactivating strategies. Child Development, 63(6), 1473–1480.

    Article  PubMed  Google Scholar 

  26. Dozier, M., Stovall-McClough, K. C., & Albus, K. E. (2008). Attachment and psychopathology in adulthood. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (Vols. 1–2, pp. 718–744). New York, NY: Guilford Press.

    Google Scholar 

  27. Fagundes, C. P., Diamond, L. M., & Allen, K. P. (2012). Adolescent attachment insecurity and parasympathetic functioning predict future loss adjustment. Personality and Social Psychology Bulletin, 38(6), 821–832.

    Article  PubMed  Google Scholar 

  28. Feeney, J. A. (2000). Implications of attachment style for patterns of health and illness. Child: Care, Health and Development, 26(4), 277–288.

    Google Scholar 

  29. Feeney, J. A., Kelly, L., Gallois, C., Peterson, C., & Terry, D. J. (1999). Attachment style, assertive communication, and safer-sex behavior. Journal of Applied Social Psychology, 29(9), 1964–1983.

    Article  Google Scholar 

  30. Feeney, J. A., Noller, P., & Patty, J. (1993). Adolescents’ interactions with the opposite sex: Influence of attachment style and gender. Journal of Adolescence, 16(2), 169–186.

    Article  PubMed  Google Scholar 

  31. Gratz, K. L. (2003). Risk factors for and functions of deliberate self-harm: An empirical and conceptual review. Clinical Psychology: Science and Practice, 10(2), 192–205.

    Google Scholar 

  32. Greenberg, M. T. (1999). Attachment and psychopathology in childhood. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 469–496). New York, NY: Guilford Press.

    Google Scholar 

  33. Gross, J. J., & John, O. P. (2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85(2), 348–362.

    Article  PubMed  Google Scholar 

  34. Gross, J. J., & Levenson, R. W. (1997). Hiding feelings: The acute effects of inhibiting negative and positive emotion. Journal of Abnormal Psychology, 106(1), 95–103.

    Article  PubMed  Google Scholar 

  35. Gunnar, M. R. (1991). The psychobiology of stress in early development: Reactivity and regulation. Minneapolis, MN: International Society for the Study of Behavioral Development.

    Google Scholar 

  36. Hankin, B. L., & Abramson, L. Y. (2001). Development of gender differences in depression: An elaborated cognitive vulnerability–transactional stress theory. Psychological Bulletin, 127(6), 773

    Article  PubMed  Google Scholar 

  37. Heath, N. L., & Nixon, M. K. (2009). Assessment of nonsuicidal self-injury in youth. In N. L. Heath & M. K. Nixon (Eds.), Self-injury in youth: The essential guide to assessment and intervention (pp. 143–170). New York, NY: Routledge.

    Google Scholar 

  38. Kerns, K. A., Aspelmeier, J. E., Gentzler, A. L., & Grabill, C. M. (2001). Parent–child attachment and monitoring in middle childhood. Journal of Family Psychology, 15(1), 69–81.

    Article  PubMed  Google Scholar 

  39. Khantzian, E. J. (1985). The self-medication hypothesis of addictive disorders: focus on heroin and cocaine dependence. American Journal of Psychiatry, 142, 1259–1264.

    Article  PubMed  Google Scholar 

  40. Klonsky, E. D. (2007). The functions of deliberate self-injury: A review of the evidence. Clinical psychology review, 27(2), 226–239.

    Article  PubMed  Google Scholar 

  41. Kovacs, M., Joormann, J., & Gotlib, I. H. (2008). Emotion (Dys)regulation and links to depressive disorders. Child Development Perspectives, 2(3), 149–155.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Lemelin, C., Lussier, Y., Sabourin, S., Brassard, A., & Naud, C. (2014). Risky sexual behaviours: The role of substance use, psychopathic traits, and attachment insecurity among adolescents and young adults in Quebec. The Canadian Journal of Human Sexuality, 23(3), 189–199.

    Article  Google Scholar 

  43. Lieberman, M., Doyle, A. B., & Markiewicz, D. (1999). Developmental patterns in security of attachment to mother and father in late childhood and early adolescence: Associations with peer relations. Child Development, 70, 202–213.

    Article  PubMed  Google Scholar 

  44. Lloyd, E. E., Kelley, M. L., & Hope, T. (1997, April). Self-mutilation in a community sample of adolescents: Descriptive characteristics and provisional prevalence rates. In annual meeting of the Society for Behavioral Medicine, New Orleans, LA.

  45. Luster, T., & Small, S. A. (1994). Factors associated with sexual risk-taking behaviors among adolescents. Journal of Marriage and the Family, 56(3), 622–632. doi:

  46. MacPherson, L., Magidson, J. F., Reynolds, E. K., Kahler, C. W., & Lejuez, C. W. (2010). Changes in sensation seeking and risk‐taking propensity predict increases in alcohol use among early adolescents. Alcoholism: Clinical and Experimental Research, 34(8), 1400–1408.

    Google Scholar 

  47. Magar, E. C., Phillips, L. H., & Hosie, J. A. (2008). Self-regulation and risk-taking. Personality and Individual Differences, 45(2), 153–159.

    Article  Google Scholar 

  48. Main, M. (1981). Avoidance in the service of attachment: A working paper. Behavioral Development: The Bielefeld Interdisciplinary Project, 651–693.

  49. Martin, J., Bureau, J. F., Lafontaine, M. F., Cloutier, P., Hsiao, C., Pallanca, D., & Meinz, P. (2017). Preoccupied but not dismissing attachment states of mind are associated with nonsuicidal self-injury. Development and Psychopathology, 29(2), 379–388.

    Article  PubMed  Google Scholar 

  50. Martin, J., Bureau, J. F., Yurkowski, K., Lafontaine, M. F., & Cloutier, P. (2016). Heterogeneity of relational backgrounds is associated with variation in non-suicidal self-injurious behavior. Journal of Abnormal Child Psychology, 44(3), 511–522.

    Article  PubMed  Google Scholar 

  51. Martin, J., Raby, K. L., Labella, M. H., & Roisman, G. I. (2017). Childhood abuse and neglect, attachment states of mind, and non-suicidal self-injury. Attachment & Human Development, 1–22.

  52. Mayer, J., & Filstead, W. J. (1979). The Adolescent Alcohol Involvement Scale. An instrument for measuring adolescents’ use and misuse of alcohol. Journal of Studies on Alcohol, 40(3), 291–300.

    Article  PubMed  Google Scholar 

  53. McWilliams, L. A., & Bailey, S. J. (2010). Associations between adult attachment ratings and health conditions: Evidence from the national comorbidity survey replication. Health Psychology, 29(4), 446–453.

    Article  PubMed  Google Scholar 

  54. Mikulincer, M., & Shaver, P. R. (2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. Advances in Experimental Social Psychology, 35, 53–152.

    Article  Google Scholar 

  55. Mikulincer, M., Dolev, T., & Shaver, P. R. (2004). Attachment-related strategies during thought suppression: Ironic rebounds and vulnerable self-representations. Journal of Personality and Social Psychology, 87(6), 940–956.

    Article  PubMed  Google Scholar 

  56. Moberg, D. P. (2003). Screening for alcohol and other drug problems using the Adolescent Alcohol and Drug Involvement Scale (AADIS). Madison, WI: Center for Health Policy and Program Evaluation, University of Wisconsin-Madison. Retrieved from.

    Google Scholar 

  57. Moretti, M. M., & Peled, M. (2004). Adolescent-parent attachment: Bonds that support healthy development. Paediatrics & Child Health, 9(8), 551–555.

    Article  Google Scholar 

  58. Pietromonaco, P. R., Uchino, B., & Dunkel Schetter, C. (2013). Close relationship processes and health: Implications of attachment theory for health and disease. Health Psychology, 32(5), 499–513.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Puig, J., Englund, M. M., Simpson, J. A., & Collins, W. A. (2013). Predicting adult physical illness from infant attachment: A prospective longitudinal study. Health Psychology, 32(4), 409–417.

    Article  PubMed  Google Scholar 

  60. Reynolds, S., Searight, H. R., & Ratwik, S. (2014). Adult attachment styles and rumination in the context of intimate relationships. North American Journal of Psychology, 16(3), 495–506.

    Google Scholar 

  61. Richards, J. M., & Gross, J. J. (2000). Emotion regulation and memory: the cognitive costs of keeping one’s cool. Journal of Personality and Social Psychology, 79(3), 410–424.

    Article  PubMed  Google Scholar 

  62. Sbarra, D. A., & Borelli, J. L. (2013). Heart rate variability moderates the association between attachment avoidance and self-concept reorganization following marital separation. International Journal of Psychophysiology, 88(3), 253–260.

    Article  PubMed  Google Scholar 

  63. Scharfe, E., & Eldredge, D. (2001). Associations between attachment representations and health behaviors in late adolescence. Journal of Health Psychology, 6(3), 295–307.

    Article  PubMed  Google Scholar 

  64. Schindler, A., & Bröning, S. (2015). A Review on Attachment and Adolescent Substance Abuse: Empirical Evidence and Implications for Prevention and Treatment. Substance Abuse, 36(3), 304–313.

    Article  PubMed  Google Scholar 

  65. Schindler, A., Thomasius, R., Petersen, K., & Sack, P.-M. (2009). Heroin as an attachment substitute? Differences in attachment representations between opioid, ecstasy and cannabis abusers. Attachment & Human Development, 11(3), 307–330.

    Article  Google Scholar 

  66. Schindler, A., Thomasius, R., Sack, P.-M., Gemeinhardt, B., & Küster, U. (2007). Insecure family bases and adolescent drug abuse: A new approach to family patterns of attachment. Attachment & Human Development, 9(2), 111–126.

    Article  Google Scholar 

  67. Smith, G. T., & Cyders, M. A. (2016). Integrating affect and impulsivity: The role of positive and negative urgency in substance use risk. Drug and Alcohol Dependence, 163, S3–S12.

    Article  PubMed  PubMed Central  Google Scholar 

  68. Spangler, G. (1998). Emotional and adrenocortical responses of infants to the strange situation: The differential function of emotional expression. International Journal of Behavioral Development, 22(4), 681–706.

    Article  Google Scholar 

  69. Spangler, G., & Grossmann, K. E. (1993). Biobehavioral organization in securely and insecurely attached infants. Child Development, 64(5), 1439–1450.

    Article  PubMed  Google Scholar 

  70. Thorberg, F. A., & Lyvers, M. (2010). Attachment in relation to affect regulation and interpersonal functioning among substance use disorder in patients. Addiction Research & Theory, 18(4), 464–478.

    Article  Google Scholar 

  71. Tracy, J. L., Shaver, P. R., Albino, A. W., & Cooper, M. L. (2003). Attachment styles and adolescent sexuality. In P. Florsheim (Ed.), Adolescent romance and sexual behavior: Theory, research, and practical implications (pp. 137–159). Mahwah, NJ: Lawrence Erlbaum Associates.

  72. White, L. O., Wu, J., Borelli, J. L., Rutherford, H. J. V., David, D. H., Kim–Cohen, J., & Crowley, M. J. (2012). Attachment dismissal predicts frontal slow-wave ERPs during rejection by unfamiliar peers. Emotion, 12(4), 690–700.

    Article  PubMed  Google Scholar 

  73. Wills, T. A., Walker, C., Mendoza, D., & Ainette, M. G. (2006). Behavioral and emotional self-control: relations to substance use in samples of middle and high school students. Psychology of Addictive Behaviors, 20(3), 265–278.

    Article  PubMed  Google Scholar 

  74. Winters, K. C., Botzet, A., Anderson, N., Bellehumeur, T., & Egan, B. (2001). Screening and assessment study, Wisconsin Division of juvenile corrections, alcohol and other drug abuse. Madison, WI: University of Minnesota. Minneapolis: Center for Adolescent Substance Abuse Research. Retrieved from

  75. Zimmermann, G. (2010). Risk perception, emotion regulation and impulsivity as predictors of risk behaviours among adolescents in Switzerland. Journal of Youth Studies, 13(1), 83–99.

    Article  Google Scholar 

Download references

Author Contributions

J.B.: conceptualized the design, executed the study, conducted the data analyses, and wrote the manuscript. L.H.: collaborated with the writing of the Introduction and Discussion. L.E.: designed the adolescent follow-up assessment, conceptualized the hypotheses, and contributed to analyses and writing.

Author information



Corresponding author

Correspondence to Jessica L. Borelli.

Ethics declarations

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee of Pomona College and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all individual participants included in the study. Parents completed consent forms, and children completed assent forms.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Borelli, J.L., Ho, L. & Epps, L. School-aged Children’s Psychobiological Divergence as a Prospective Predictor of Health Risk Behaviors in Adolescence. J Child Fam Stud 27, 47–58 (2018).

Download citation


  • Psychobiological divergence
  • Deactivation
  • Neuroendocrine reactivity
  • Health risk behaviors
  • Attachment