| With the rapid development of the modern technology, we have enter an era of the Internet. With popularized and deepened of the Internet-based interpersonal interactions, Internet brings people efficiencies and conveniences, as well as online interpersonal trust problems. Online interpersonal trust is the expectation of one party towards another party’s actions and speeches while they are interacting with each other online. Unlike the real life "face-to-face" interactions, online interpersonal trust is built on the virtual interactions, with an unknowable reality and unpredictable risk. Researchers research mainly the factors of online interpersonal trust, e.g. the difference of individual’s trust propensity, net age, gender, etc. Most of the past researches were based on survey, a method to study factors of online interpersonal trust in a static state; moreover, in the most times, individuals only face other’s avatars on the internet and the avatars are symbols of our web image. Some researchers point out that the anthropomorphic degree of the avatars will affect the determinants of trust. Based of the above two points, this study integrated with a popular avatar design program called the’Face Q", adopted real human face avatars and cartoon avatars as study targets, and used trust game as a dynamic study method, via two experiments to investigate the difference of initial trust level between real human face avatars and cartoon avatars.Experiment number one used a hybrid experiment design of 2 avatar types (real human face and cartoon),2 avatar genders (male, female) and 2 experimentees (male, female) to study characteristics of college students’trust level on real human face avatars and cartoon avatars. In this experiment, avatar types and avatar genders are within-subject variables, experimentee’s genders are between-subject variables. The result shows the avatar types have a more significant effect on the trust level; real human face avatars has a higher trust level than cartoon avatars. And more deeply, real human female face avatars have a higher trust level than real human male face avatars, but the trust level both genders cartoon avatar was about the same. On the other hand, both genders’experimentees showed about the same level of trust on real human face avatars; but male experimentees tends to trust cartoon avatars more than female experimentees.Experiment number two used the trust level scores obtained on experiment number one to categorize both cartoon and real human face avatars with both high and low trust levels, and select out 4 avatars in each categories (two male avatars and two female avatars) as study targets, used multi-rounds trust game as a study method to investigate experimentee’s trust level on different types of avatars. This experiment is a between-subject design experiment with two avatar types (cartoon and real human face), two trust level (high and low). With the experimentee’s trust level being the dependent variable, to complete, (1) the average score after 10 rounds and, (2) the score of the first round, fifth round and tenth round being the quantitative indicators. The result shows, the avatar type did not cause a big difference in expeimentee’s score. Moreover, initial trust level did not cause the difference in experimentee’s score in the first round, but affected the overall! average score.In summary, avatar type showed different results in two experiments, in the "first impression", real human face avatar has a higher trust level than cartoon avatar, but after the multi-rounds trust game, avatar type did not cause the difference in scores. In experiment two, with specified analysis on each steps’score, and the non-unified effects of the initial trust level, this indicates that with the development of online interpersonal trust, its effects on the factors are different too. Therefore, online interpersonal trust is just the same as the traditional offline trust, which is a process of dynamic development... |
PDF Full Text Request