Neuroimaging Studies of Hunger and Impulsivity

187 Cognitive Neuroscience Society expressions. This study investigated how differentially valenced musical contexts would bias amygdala responses to surpised facial expressions. During fMRI scans, subjects listened to 16 minutes of valenced (i.e., either positive or negative) music followed by alternating blocks of surprised faces and fi xation. Because amygdala signal changes tend to habituate with repeated presentations, blocks of surprised faces were divided into early and late trials within each run. A signifi cant music by time interaction was observed in the right amygdala. Subjects primed with negative music showed a heightened right amygdala signal to surprised faces during early trials that quickly habituated. Conversely, subjects primed with positive music showed moderate right amygdala signal that was sustained through both the early and late trials. This amygdala response pattern to surprised faces preceded by negative music strongly resembles amygdala responses to fearful faces devoid of any musical context (see Whalen et al., 2009). This suggests that negative music biased perceivers to interpret surprised faces as more negative. However, amygdala responses to surprised faces preceded by positive music were typical of those seen to surprise faces devoid of a musical context, suggesting that the positive music did not necessarily bias the interpretation of surprise.