Introduction The attribution of intent during social interaction is a cornerstone of the social adjustment process ( Crick and Dodge, 1994 ). In particular, the tendency to attribute hostile intent has consistently been associated with and argued to play a role in the development and maintenance of aggressive behavior in childhood and adolescence ( Dodge, 2006 ). However, little is known about the neural correlates of hostile intent attributions ( Erdley et al., 2010 ). We used fMRI and behavioral measures to investigate intent attributions and attention biases for hostile social stimuli using social laughter processing in children and adolescents under consideration of gender and age effects. Methods 35 girls and 28 boys in three age groups (9–12, 13–15 and 16–18y; mean 15.3, SD 2.39) without current psychiatric diagnoses rated 1.5s audiovisual laugh sequences in a 3T MRI scanner. 20 friendly, 20 tickling and 20 taunting sequences ( Szameitat et al., 2009 ) were rated as friendly or taunting on a four-point scale. Rating, reaction time (RT) and attention bias (RT friendly – RT taunting laughter) differences were calculated with three 2 (gender) × 3 (age group) × 3 (laughter type) mixed model ANOVAs. Participants with excessive movement or spurious responding were removed from analysis. Neural responses were calculated in SPM8 in a whole brain analysis of the BOLD response to each laughter type and the correlations to hostile intent rating, physical aggression ( Buss and Perry, 1992 ) and callous unemotional traits ( Frick, 2003 ). Results A significant main effect of laughter type ( F = 100.62, p F = 2.829, p = .068) and the main effect of age on attention bias ( F = 2.634, p = .083) approached significance, with younger participants showing attention bias to taunting and older participants to friendly laughter. Callousness and physical aggression correlated with hostile ratings of friendly ( r = .288, p r = .315, p r = −.404, p A broad network showed increased BOLD response during laughter processing, including bilateral medial and superior temporal gyri (MTG, STG), inferior and medial frontal gyri (IFG, MFG), caudate nuclei and left superior parietal lobe (whole brain FWE p Conclusions The current study found an extended neural network involved in laughter processing in children and adolescents, similar to that found in adults ( Szameitat et al., 2010 ). In addition, attribution of hostile intent in laughter was found to associate with changes in BOLD signal strength across these processing areas. This association strength was modulated by youth’s aggressive behavior and callous personality traits.