Longitudinal Relations between Early Childhood Observed Behavioral Inhibition and Reports of Behavioral Inhibition in Adolescence

• In Event: Poster Session 01
  In Poster Session: PS 01 Topic - Social, Emotional, Personality

Abstract

Behavioral inhibition (BI) is a temperamental trait and one of the best early predictors of anxiety (Fox et al., 2005). However, there is important heterogeneity in the research literature in how BI is measured. Behavioral observation in early childhood is thought to be the “gold standard” as it avoids reporter and recall bias. On the other hand, parent reports provide an efficient and scalable way to assess temperament that also provides a broader perspective that includes a variety of naturally occurring situations. As such, there is a need to examine how these different approaches relate to each other, especially since BI obtained via questionnaire is usually collected in older children and adolescents. Researchers have developed methods to assess BI across development, including later in childhood and adolescence, but their relations to behavioral measures have only been examined concurrently and in early childhood (Dyson, 2011 & Bishop, 2003). Using a unique longitudinal cohort, we examine the longitudinal relation between observed BI in early childhood to parent- or child-reported BI in adolescence. In addition, we examine which subcomponent of BI reports is most strongly linked to early childhood observed BI.

Participants were originally selected to be part of a longitudinal sample (N=291). At 24 and 36 months of age, participants were brought into the laboratory and assessed for BI using a standardized paradigm of novel stimuli and situations (i.e., unfamiliar room, an adult stranger, and a novel toy). At 13 years of age, participants (n=172) and their parents (n=167) completed the Behavioral Inhibition Questionnaire (BIQ). The BIQ includes subscales about novelty in the context of peers, physical challenges, separation, performance situations, adults, and unfamiliar situations.

Results showed that early childhood BI was related to parent-reported BIQ, r(155)=.20, p=.01, but was not significantly related to child-reported BIQ, r(170)=.11, p=.15, at 13 years. These effects remained unchanged when controlling for gender, maternal education, and ethnicity, and did not vary by gender. We used a regression that included all the subscales for parent and child reports separately to examine subcomponents and found that the only significant subscale was parent-reported social novelty in the context of unfamiliar adults (p = .02.). When including all the child-report subscales, none of them were significant.

These results suggest that parent reports of BI in adolescence capture related processes to the BI observed in the laboratory a decade earlier. However, the relation was modest and it was not significant when using child reports. Moreover, an exploration of the different subcomponents of BI suggests that early-observed BI most strongly related to children's reactions to social novelty in the context of unfamiliar adults. These findings provide important information for linking investigations using parent-reports of temperament in middle late childhood and adolescence with studies using observational measures in early childhood. Moreover, they open new avenues of research as social novelty in adolescence is often examined in the context of unfamiliar peers rather than unfamiliar adults.

Authors

• Jamie Listokin, University of Maryland - College Park
  Presenting Author

• Santiago Morales, University of Southern California
  Non-Presenting Author

• Kathryn A Degnan, Catholic University of America
  Non-Presenting Author

• Heather Henderson, University of Waterloo
  Non-Presenting Author

• Nathan A Fox, University of Maryland - College Park
  Non-Presenting Author