Processing efficiency theory (PET, Eysenck & Calvo, 1992) and attentional control theory (ACT, Eysenck, Derakshan, Santos, & Calvo, 2007) adopt new perspectives of executive functions to explain the relationship between anxiety and performance. There are two main assumptions for PET: 1) Anxiety impairs processing efficiency but not performance effectiveness; 2) Anxiety impairs the central executive system of working memory. ACT was proposed on the base of PET. There are also two main assumptions for ACT: 1) Anxiety impairs goal-directed attention system; 2) Anxiety impairs efficiency of the inhibition and switching functions. The issue of anxiety-performance relationship is one of the most important issues in the competitive sport field. There are already several studies exploring the assumptions of ACT in competitive sport field, most of which focused on the first hypothesis of ACT: anxiety impairs goal-directed attention system. The present research aimed to examine the second assumptions of ACT, anxiety impairing efficiency of the inhibition function, to provide evidence for the application of ACT in the competitive sport field. The interference effect of competitive trait anxiety on inhibition measured by the antisaccade task was examined in two eye-tracking studies. Experiment 1 hypothesized that competitive trait anxiety would impair efficiency but not performance of inhibition function. Twenty five athletes were divided into competitive trait anxiety high-score group and low-score group. Three dependent variables, antisaccade latency and error rate of antisaccade indicating efficiency, and accuracy rate indicating performance effectiveness were recorded. No significant difference was found between the two groups on antisaccade latency, however, the error rate of antisaccade in the high-score group was significantly higher than that in the low-score group. No significant difference on accuracy rate was shown. In sum, the hypothesis of competitive trait anxiety impairing efficiency but not performance of inhibition function was partially supported. Experiment 2 was designed to examine the hypothesis that competitive trait anxiety would impair efficiency but not performance of inhibition function in stress condition. Twenty three athletes were divided into competitive trait anxiety high-score group and low-score group, and both groups performed the antisaccade task in the stress and non-stress situation. The results demonstrated that the high-score group exhibited significantly longer antisaccade latency and higher error rate of antisaccade than the low-score group in both the stress and non-stress situation. No significant difference on the accuracy rate was found between the two groups. All in all, it was the efficiency but not the performance of inhibition function that was interfered by competitive strait anxiety in experiment 2. The results not only supported the hypothesis proposed by Experiment 2 that competitive trait anxiety impaired efficiency but not performance of inhibition function in stress condition, but also supported the hypothesis set up in Experiment 1 for non-stress condition. To sum up, the present research demonstrated that: competitive trait anxiety impaired processing efficiency more than performance effectiveness, and competitive trait anxiety impaired processing efficiency of inhibition function. It is concluded that processing efficiency theory and attentional control theory are applicable for explaining the influence of anxiety on performance in the competitive sport field.