Subjective fear, interference by threat, and fear associations independently predict fear-related behavior in children

Background and objectives: Several information-processing models highlight the independent roles of controlled and automatic processes in explaining fearful behavior. Therefore, we investigated whether direct measures of controlled processes and indirect measures of automatic processes predict unique variance components of children's spider fear-related behavior.
Method: Seventy-seven children between 8 and 13 years performed an Affective Priming Task (APT) measuring associative bias, a pictorial version of the Emotional Stroop Task (EST) measuring attentional bias, filled out the Spider Anxiety and Disgust Screening for Children (SADS-C) in order to assess self-perceived fear, and took part in a Behavioral Assessment Test (BAT) to measure avoidance of spiders.
Results: The SADS-C, EST, and APT did not correlate with each other. Spider fear-related behavior was best explained by SADS-C, APT, and EST together; they explained 51% of the variance in BAT behavior.
Limitations: No children with clinical levels of spider phobia were tested. The direct and the different indirect measures did no correlate with each other.
Conclusions: These results indicate that both direct and indirect measures are useful for predicting unique variance components of fear-related behavior in children. The lack of relations between direct and indirect measures may explain why some earlier studies did not find stronger color-naming interference or stronger fear associations in children with high levels of self-reported fear. It also suggests that children with high levels of spider-fearful behavior have different fear-related associations and display higher interference by spider stimuli than children with non-fearful behavior.