This research set out to find whether the use of gestures by virtual embodied agents could affect perceptions of integrity by humans. 48 participants played a quiz with the assistance of a gesturing and an idle agent. Likert scales were used to evaluate their perceptions of integrity, which showed no significant difference between either agent. Interestingly, the actions of the participants showed something different: participants chose the gesturing agent to provide assistance a significantly higher number of times than the idle agent. One possible explanation for this result is that even though participants did not consciously notice a difference between the two agents, they subconsciously did while performing the quiz task. Another explanation is that choosing the gesturing agent more did not have to do with perceived integrity, but rather perceived ability, benevolence, or something else entirely.