Objective: Chess and Thomas suggested that temperament might make a contrib
ution to social phobia and other forms of extreme social anxiety. This stud
y provides the first investigation of the outcomes in adolescents who had b
een inhibited (subdued to and avoidant of novelty) or uninhibited (approach
ing novelty) in the second year of life, utilizing both direct interview an
d direct observation. Method: Seventy-nine subjects, aged 13 years, who had
been classified as inhibited or uninhibited in the second year were assess
ed with both standardized interview and direct observation. Results: There
was a significant association between earlier classification of a child as
inhibited and generalized social anxiety at adolescence, but no association
with specific fears, separation anxiety, or performance anxiety. The adole
scents who were classified as socially anxious made fewer spontaneous comme
nts than those without social anxiety; no relation was seen between any oth
er type of fear and the number of spontaneous comments. Adolescent girls wh
o had been inhibited as toddlers were more likely to be impaired by general
ized social anxiety than boys. Conclusions: The interview and observational
data indicate that important aspects of an inhibited temperament are prese
rved from the second year of life to early adolescence, which predispose an
adolescent to social anxiety.