Abstract: Previous studies have mostly focused on the process of understanding single individual’s neutral or innocent intentions. The ecological validity of stimuli which were used before, such as point-lights, geometries and virtual cartoons, was not as high as that of photos or movie clips. Recent years, the intention researches have begun to study the emotional intentions performed by two or multiple individuals’ whole body movement. The photos and movie clips used in the experiments were captured from the real figures by digital camera. However, the role of stimulus types and genders playing in the intention understanding still remains unclear. In order to address this issue, we carried out the experiment to explore the behavioral cognition characteristics underlying the process of understanding interactive intentions. One hundred college students participated in the experiment. Half of them (25 males and 25females) were randomly assigned to complete a standard intention inference task to discriminate kind, hostile and non-interactive intentions from each other which were displayed by two virtual cartoon figures, while the other half (25 males and 25 females) were assigned to understand those intentions displayed by photos. It should be noted that the photos which were captured from two male and two female college students were taken by a digital camera. The cartoons were professionally drawn by an excellent undergraduate majoring?in?fine arts according to the photos. The physical characteristics of cartoons and photos in the formal experiment matched very well. The mean accuracy and reaction time were analyzed using 2(stimulus ecologicalness: cartoons and photos) × 2(subjects’ gender: male and female) × 2(actors’ gender: male and female) × 3(intention types: kind intention, hostile intention and non-interactive intention) repeated ANOVA test. The stimulus types and subjects’ genders were between-group factors, while the actors’ gender and intention type worked as within-group factor. The accuracy and reaction time of the participants were recorded when they performed the intention inference task. There was a main effect of stimulus ecologicalness on accuracy, which demonstrated a better performance on photos compared with that of cartoons. The main effects of actors’ genders on accuracy and reaction time were also very significant, which demonstrated a remarkable male advantage. The results also indicated the highest accuracy and fastest reaction time for hostile intention among the intention conditions displayed by both cartoons and photos. It is consisted with the negative bias theory. The analysis also revealed an interaction effect between subject’s gender and stimulus type on accuracy. And a marginal interaction effect between subject’s gender and stimulus type on reaction time also existed. It should be noted that an interaction effect between actors’ gender and intention types was found on reaction time. The hostile intention displayed by male actors showed a faster reaction time than that displayed by female actors, while there was no significant difference on kind intention displayed by male and female actors. In conclusion, the current study demonstrates that whatever the virtual cartoon figures or real persons are, the behavioral processing may vary with different types of social motor intentions, and is also modulated by the actors’ gender.

Key words: kind intention, hostile intention, stimulus ecologicalness, gender, social cognition

摘要： 为考察材料生态性、被试性别和角色性别对理解友好与敌对意图的影响，分析了100名被试观察由2名男性或2名女性表演的三类互动意图的照片或卡通时的正确率和反应时。结果：理解真实人物的正确率高于虚拟角色；女性被试理解虚拟角色的反应时长于真实人物，而男性被试却未出现材料生态性差异；理解男性角色敌对意图的反应时短于女性角色的，而理解友好意图时却未出现角色性别差异。研究为负性偏向等理论提供实证证据，揭示材料生态性、角色性别对理解友好和敌对意图的影响机制。

关键词: 友好意图 敌对意图 材料生态性 性别 社会认知