Purpose. Non-verbal communication researchers have identified specific categories of hand gestures but deception researchers typically ignore these. This experiment refined and developed some of these categories and examined whether there is a difference in the frequency of speech prompting and rhythmic pulsing gestures between liars and truth tellers. Methods. Twenty truth tellers and 20 liars (all undergraduate students) described a person who entered a room where they were playing a game with a confederate. Truth tellers gave a truthful description of an event they had participated in. Liars had previously taken money from a wallet in the room but had not played a game with the confederate, or seen anybody enter the room, they just pretended they did during their interview. Results. Truth tellers made more rhythmic pulsing gestures than liars indicating this type of gesture may be connected with the prosodic flow of speech. Liars made significantly more speech prompting gestures than truth tellers, supporting the notion that greater cognitive load may be experienced during deceptive accounts. Conclusions. This study demonstrates the benefit of examining subcategories of gestures when investigating deceptive behaviour.