A group of 20 internals and a group of 20 externals, counterbalanced for sex, were served as subjects in an experiment of yoked pairs design, which was purported to determine if there was any difference in the trends of emotional changes between these two groups. Based on Rotter's (1966) theory, it was hypothesized that the internals would achieve better control of emotion than the externals might do when they were instructed to be calm and relaxed as soon as possible, and that consequently there must be an Internal-External × Time Block interaction effect as the experiment continued. Both EKG and BSR were simultaneously recorded during four time blocks for each member of each yoked pair. Data thus obtained were analyzed by multivariate analysis. Results showed that there was no significant interaction effect as predicted. Since both internals and externals showed the same trends in decelerating their HR and skin conductance, these results did not seem to support the notion that HR and GSR could be intentionally controlled.