Abstract: Deficits in social communication are a hallmark of autism spectrum disorders (ASD). Through a series of papers, this dissertation explored the neural basis of interpreting communicative intent in high-functioning children and adolescents with ASD. Paper 1 examined differences in the neural networks underlying facial emotion processing in children with ASD and typically developing (TD) children. When matching facial expressions, the ASD group showed reliably less activity than the TD group in the fusiform face area. However, no differences were observed while assigning a verbal label to facial expressions, suggesting that individuals with ASD may be relatively unimpaired in the cognitive assessment of emotions, yet still show deficits in the automatic perception of facial expressions. Paper 2 explored age-related changes in the neural circuitry supporting the interpretation of communicative intent in TD children and adults, using irony comprehension as a test case. A developmental shift was observed from greater reliance on prefrontal regions in children to posterior occipitotemporal regions in adults, which may reflect the automatization of basic reasoning about mental states. Paper 3 used the same paradigm to investigate the neural basis of impairments in interpreting irony in children with ASD. Reduced activity in regions involved in reasoning about others' mental states was observed in ASD relative to TD children during the perception of potentially ironic scenarios. Importantly, explicit instructions to attend to facial expression and tone of voice yielded increased activity in the medial prefrontal cortex only in the ASD group. Finally, paper 4 examined the role of context and intonation in interpreting irony at both the behavioral and neural levels. ASD children were reliably less accurate than TD children at taking advantage of contextual cues detailing the outcome of an event. Overall, the ASD group showed reliably greater activity in frontal-temporal regions that were also recruited by TD children, perhaps reflecting more effortful processing when interpreting a speaker's intended meaning. Taken together, these papers suggest that there is no fundamental deficit of neural functioning in regions that support the normal processing of important social information in individuals with ASD. Instead, high-functioning individuals with ASD may be impaired in the automatic engagement of this circuitry when implicitly processing social cues, but do show activation in crucial networks when task demands require or encourage explicit attention to the relevant information.