Cognitive bias for pictorial and linguistic threat cues in children

The present study was designed to test the conditions under which threat-related cognitive bias can be observed in anxious children. Measures of cognitive bias for threatening words and pictures were obtained from spider fearful children (N = 55) and non-fearful children (N = 58) aged 8-11 in the first experiment, and from spider fearful children (N = 44), and two control groups (N = 41; N = 36) aged 8 in the second experiment. Cognitive bias was assessed by the emotional Stroop task. In line with our previous findings, all children aged 8 showed a bias for spider words, but not for spider pictures. However, a relation between spider fear and bias was observed when age was taken into account: bias for spider words decreased with age in the non-fearful children whereas this bias maintained in the fearful group. This differential age effect too replicated earlier findings (Kindt, Bierman, & Brosschot, 1997). It is suggested that a bias for threat words is a normal characteristic in children aged 8. During development, normal children learn to inhibit this processing bias, whilst fearful children fail to learn this ability.