Self-regulation is assumed to play an important role in adolescents’ social functioning. High self-regulation has been related to lower levels of externalizing behavior and to higher levels of empathic concern (e.g., Fabes et al.
1994; Oldehinkel et al.
2007). Resting respiratory sinus arrhythmia (RSA) is generally seen as a physiological marker of dispositional self-regulation (Porges
1991; Thayer and Lane
2000). Individuals low in resting RSA, indicating reduced parasympathetic activation of the heart, are thought to have more difficulties in physiological and behavioral self-regulation and are therefore less capable of displaying situation-appropriate reactions than individuals high in resting RSA (Porges
1991). Low resting RSA indeed has been related to higher externalizing behavior in clinical or at risk samples (see Kibler et al.
2004) and to lower empathic concern (e.g., Fabes et al.
1993). Besides potential direct effects, there are conceptual reasons to expect resting RSA to moderate the association between environmental influences and adolescents’ adjustment, although theories regarding the nature of this interaction are diverging (see Eisenberg et al.
2012). Therefore, we examined in a community sample of adolescent boys and girls (1) whether low resting RSA predicted higher externalizing behavior, (2) whether low resting RSA predicted lower empathic concern, and (3) whether and how resting RSA moderated the link between parent-adolescent relationship quality and adolescents’ social functioning.
Discussion
Our first aim was to examine whether resting RSA predicted adolescents’ empathic concern and externalizing behavior. Our second aim was to investigate whether and how adolescents’ resting RSA moderated the associations between parent-adolescent relationship quality and, respectively, adolescents’ empathic concern and externalizing behavior. We examined whether the interaction patterns were consistent with either the notion that high resting RSA functions as a protective factor (Ellis et al.
2011), or the notion that high resting RSA functions as a susceptibility factor (Beauchaine
2001; Thayer and Lane
2000). Whereas previous studies have almost exclusively focused on the effects of environmental adversity on children’s dysfunction, we explored the interaction effects of resting RSA with both negative and positive environmental factors on both negative and positive outcomes. Moreover, we investigated not only whether adolescents high in resting RSA were more susceptible to environmental influences than adolescents low in resting RSA, but we also tested whether they were stronger affected by both environmental adversity and benefit.
Despite conceptual reasons to expect resting RSA to be negatively linked to externalizing behavior and positively linked to empathic concern, our findings did not support this. With regard to externalizing behavior, we found no main effects of boys’ and girls’ resting RSA. Our finding is in contrast with results in clinical samples (e.g., Beauchaine et al.
2007; Mezzacappa et al.
1997), but is in line with several studies conducted in community samples that also did not find a significant association between resting RSA and externalizing behavior (e.g., Calkins et al.
2007; El-Sheikh and Whitson
2006). This suggests that low basal RSA is a marker of dysregulation for youth showing externalizing behavior in the clinical range rather than for relatively well-functioning adolescents. In a community sample of adolescents, certain levels of externalizing behavior are part of the normative development instead of an expression of pathological dysregulation (Moffitt
1993). Also with regard to empathic concern, our results did not support the expectation that high resting RSA would be a positive predictor (e.g., Fabes et al.
1993). Only for boys, we concurrently found a tendency towards a positive correlation, but the longitudinal analyses revealed the inverse association (which was qualified by a significant interaction, interpreted below). Thus, our findings as well as the inconsistent results of previous studies in community samples, suggest that above a certain threshold inter-individual differences in resting RSA may have less impact on social functioning than at lower levels. Future research may test this by comparing adolescents with scores on problem behavior in the clinical range with adolescents who score within the normal range. Further, a relationship between biological factors and problem behavior may emerge rather in interaction with environmental risk factors than as a direct association (for reviews see Raine
2005; Moffitt
2005).
Our findings did reveal support for resting RSA as a moderator in the association between parent-adolescent relationship quality and adolescents’ adjustment. For boys, resting RSA interacted with negative interaction in the prediction of empathic concern. For girls, resting RSA interacted with negative parent-adolescent interaction in the prediction of externalizing behavior, and with parental support in the prediction of empathic concern.
Looking across the interaction patterns, no support was found for high resting RSA as a buffer for the impact of low environmental quality; the effects of high negative interaction with parents or low parental support were not stronger for adolescents with low RSA than for adolescents with high RSA. In fact, the most consistent finding was that adolescents high in resting RSA appeared to be hampered relatively more by a negative environment than were adolescents low in resting RSA. That is, boys high in RSA reported less empathic concern if they experienced higher negative interaction with their parents, and girls high in RSA reported less empathic concern if they experienced lower parental support. This finding is consistent with previous studies that also found, only for children high in RSA, the association between environmental quality and social adjustment to be negative (Eisenberg et al.
2012; Blandon et al.
2008; Shannon et al.
2007). The fact that our findings were not in line with the notion that individuals with high RSA levels are better able to cope with environmental stressors, may be explained by the restricted range of scores on the relationship quality measures in this relatively low risk sample. Low scores may not reflect a truly adverse environment. Perhaps because of better attentional skills (Thayer and Lane
2000), adolescents with high resting RSA are more aware of relatively small negative changes in the relationship with their parents than are adolescents with low resting RSA.
Yet, the interaction patterns were also not clearly in line with the notion of differential susceptibility. According to the first necessary condition for establishing differential susceptibility, girls high in resting RSA indeed appeared to be more susceptible to parental support than girls low in resting RSA. That is, parental support was positively related to empathic concern for girls high in resting RSA, whereas the association was non-significant for girls low in resting RSA. Yet, the interaction pattern did not meet the second necessary condition, that is, that high susceptible individuals should be stronger affected both by positive and negative factors. Girls who were higher in resting RSA reported lower empathic concern if they experienced low parental support, but they did not report higher empathic concern if they experienced high parental support. Thus, girls high in resting RSA are, compared to girls low in resting RSA, not more susceptible to parental support for both ‘better and worse’, but only for the adverse effects of low parental support. In general, our results suggest that the negative rather than the positive end of the spectrum makes the difference (i.e., low support, high negative interaction; see Fig.
1a2 to c2). In the current sample, the average level of parenting may already be “good enough” for adolescents to be well-adjusted, and therefore differences above this level may have little effect.
The interactions between boys’ and girls’ resting RSA and negative interaction with parents showed a remarkable pattern. That is, whereas the association between boys’ negative interaction with parents and empathic was non-significant for boys high in RSA, boys with low levels of RSA reported higher levels of empathic concern if they experienced more negative interaction with their parents. Similarly, whereas girls high in RSA reported more externalizing behavior if they experienced more negative interaction with parents, girls low in resting RSA reported less externalizing behavior if they experienced more negative interaction with their parents. Thus, instead of being less responsive than adolescents high in resting RSA, as expected based on differential susceptibility theory, adolescents low in resting RSA tended to be reversely affected by negative interaction with their parents. An explanation for this finding may be that children with low resting RSA are less easy to socialize than children with high resting RSA because of their difficulties in self-regulation. As suggested by Kochanska (
1991), children with certain characteristics need more forceful parental strategies in order to develop conscience. In this way, parents’ efforts to discipline and monitor these adolescents may go together with negative interaction.
Significant gender effects were found in the prediction of empathic concern. For girls (but not boys) high in resting RSA, parental support predicted empathic concern. A possible explanation is that girls generally experience more support in their relationships and exhibit a stronger relational orientation during adolescence than do boys (e.g., De Goede et al.
2009). Girls may therefore be more attuned to and affected by changes in parental support than are boys. The finding that only for boys, the interaction between resting RSA and negative interaction predicted empathic concern, is puzzling and needs further research.
Our results should be interpreted in light of some limitations. First, self-reports were used to assess parent-adolescent relationship quality, externalizing behavior and empathic concern. Especially our finding of no significant association between resting RSA and externalizing behavior may be due to adolescents’ underreports of externalizing behavior. However, previous studies have found adolescents to be more reliable reporters of their own problem behavior than parents (Edelbrock et al.
1985; Verhulst and Van der Ende
1992). Second, due to our broad measure of externalizing behavior, it is unclear whether resting RSA, though not being associated with externalizing behavior in general, may be negatively related only to specific forms of externalizing behavior (e.g., with reactive, but not proactive aggression; Scarpa et al.
2010; Xu et al.
2014). Third, resting RSA was assessed during a laboratory session several weeks apart from the home visit during which the Time 1 questionnaires were assessed. Differences in the period in between the laboratory session and the home visit may have confounded the concurrent associations. However, resting RSA has been found to be quite stable across time (e.g., El‐Sheikh
2004). Fourth, resting RSA was assessed at the beginning of the laboratory session whereas later during this session participants had to perform a public speech task which they were encouraged to prepare before at home. In an earlier study performed in a community sample of adolescents, moderate anticipatory physiological stress responses were demonstrated preceding this task (Westenberg et al.
2009). This may have influenced the assessment of resting RSA in our study although participants watched an aquatic video which has been proven to be effective in lowering cardiovascular activity levels (Piferi et al.
2000).
By investigating in a community sample the interaction of resting RSA with both positive and negative facets of the parent-adolescent relationship, the current study extended previous research that mainly has been done in clinical or at risk samples and has primarily focused on negative contextual influences. In contrast to results of clinical studies, resting RSA was not predictive of externalizing behavior or empathic concern. However, we found resting RSA to interact with parent-adolescent relationship quality. Our findings revealed no clear support for resting RSA either as a protective factor or a susceptibility factor, but provide initial evidence for the notion that, also beyond childhood, parental efforts may differentially affect individuals with varying levels of resting RSA. The marked differences in the interaction patterns across gender, and across the full scope from negative to positive environmental factors and outcomes, highlight the importance of applying such a thorough approach. The current study shows that adolescents may respond differently to parental socialization depending on their resting RSA level, but calls for future research to disentangle the patterns in parent-adolescent interaction that may explain this differential reactivity.