This study investigated whether social context could affectthe perception of smile and sneer in individuals with high levels of social anxiety. Social anxiety and depressino symptoms of undergraduate and graduate participants were measured using self-report questionnaires, Subject then participated in an experiment using a computer. Smiling and sneering faces were morphed to 7 levels from 2:8 to 8:2 and they were used as facial stimulus in the experiment. The context in which two people were palced back-to-back or face-to-face was used as social context stimulus. Under conditions where no context appeared, only one facial stimulus was presented briefly. Under conditions in which context was presented the social context stimulus was presented first followed by facial expression stimulus simultaneously for a short period of time. Subsequently, participants evaluated whether the presented facial stimulus was smile or sneer. Only under conditions in which the context appeared, they evaluated whether the gender of the people in the context stimulus was the same or different to confirm that they were paying attention to the context. The point of subjective equality (PSE) was estimated by curve-fitting data collected from participants into a cumulative normal distribution function. As a result of correlation analysis, there was a positive correlation between social anxiety and PSEs in context conditions where the context did not appear and the two were placed back-to-back. However, PSEs in context condition where the two were placed face-to-face were not correlated with social anxiety. This suggests that the higher the social anxiety, the more likely it is to judge sneer. However, in a positive social situation where the two are facing each other, the bias in social anxiety might disappear.