The importance of eyes for virtual characters stems from the intrinsic social cues in a person's eyes. While previous work on computer generated eyes has considered realism and naturalness, there has been little investigation into how details in the eye animation impact the perception of an avatar's internal emotional state. We present three large scale experiments (N≈500) that investigate the extent to which viewers can identify if an avatar is scared. We find that participants can identify a scared avatar with 75% accuracy using cues in the eyes including pupil size variation, gaze, and blinks. Because eye trackers return pupil diameter in addition to gaze, our experiments inform practitioners that animating the pupil correctly will add expressiveness to a virtual avatar with negligible additional cost. These findings also have implications for creating expressive eyes in intelligent conversational agents and social robots.