Handwritten signature is commonly used for identity purpose in our daily life. Much research has suggested that personality, self-esteem and social status have significant effects on the size of signature. Nevertheless, whether signature position and prior awareness of signature modulate individual’s behavior is still an open question. Here, two experiments were conducted to investigate the influence of signature on ethical behavior. Experiment 1 was conducted to examine the effects of signature position and prior awareness of signature on honesty. 123 subjects participated in the experiment. It consisted of three phases: pre-task phase, task phase and post-task phase. In the pre-task phase, all participants were asked to learn 30 moral words. Then, they performed a color judgment task. In the task phase, subjects were instructed to play a dice game, in which they threw a six-sided dice by pressing a key. Each subject had five chances and the total scores (preset to 12 for all subjects) were calculated by adding the five scores. Before the dice game, half of subjects were informed that they had to sign their names to ensure honesty after the dice game, while another half were not informed. After this game, participants were told that the higher points they threw the higher reward they would get. Participants were asked to report their total scores and sign their names on an answer sheet. Half of them signed on the top (top-signature group), while another half signed their on the bottom (bottom-signature group). Subjects were considered as cheating if their self-reported scores were higher than 12. The percent of cheating was calculated for each condition. In the post-task phase, subjects performed a recognition task on moral words. Results showed that signature position had significant effects on honesty for subjects without prior notification of signature, but not for subjects with prior notification of signature. Specifically, the top-signature group behaved more honestly (i.e., reporting higher scores) than the bottom-signature group in the dice game if they did not know the requirement of signature, while there were no differences between the two groups if they knew the requirement of signature. Furthermore, subjects who lied in the dice game remembered less positive words than those who behaved honestly. Experiment 2 was conducted to examine whether the ways of commitment (i.e., handwritten signature and verbal promise) modulate the effect of commitment on ethical behavior. Similar to Experiment 1, subjects performed the color judgment task and the dice game in Experiment 2. Subjects were randomly assigned into three conditions: handwritten signature condition, verbal promise condition, and no-commitment condition. The three conditions only differed in the way of commitment they made. In addition to the total scores, subjects were asked to report the time spent in the dice game. They were told that they would get more reward if they spent more time in the dice game. Cheating was assessed using the following two indices: percentage of cheating based on self-reported scores and self-reported time spent in the dice game. Results showed that subjects in the handwritten signature condition and those in the verbal promise condition behaved more honestly (i.e., smaller percentage of cheating and shorter self-reported time) than those in the no-commitment condition. There were no significant differences between subjects in the handwritten signature condition and those in the verbal promise condition. Taken together, results from the two experiments suggest that both signature and verbal promise make people behave more honestly by priming the individual’s self-identity.