This study aimed to evaluate the relationship between risk for, and manifestation of, anxiety disorders, and acquisition, extinction and spontaneous recovery of aversive associations. Three pilot groups of youths were compared: anxiety disordered (generalized anxiety, social anxiety, panic or separation anxiety, n519), high risk (anxious parents but non-anxious youths, n515) and controls (non-anxious parents and non-anxious youths, n512). After a series of habituation trials, youths underwent 16 trials of differential conditioning, including 8 trials in which a CS1(geometric figure) is paired with a loud tone. Then, following a 5minute interval, youths underwent 8 extinction trials.One week later, they were assessed for spontaneous recovery with another 8 extinction trials. Measures included skin conductance, startle blink magnitude and subjective ratings of valence and arousal. Results indicate that anxious youths show less differential conditioning and rate the CS1as more arousing and more negative relative to high risk youths, but not relative to controls. In addition, anxious youths show delayed extinction relative to high risk youths, but not relative to controls. Finally, anxious youths showed more spontaneous recovery one week later than either high risk youths or controls. The findings are moderated by gender effects. It is conceivable that aversive learning is not a marker of risk for anxiety but is rather a manifestation of anxiety; it is also conceivable that high risk youths actually represent ''''super normals'''.