The intervention effects of the computer programs of rapid eye movement (REM) and relaxation on acute stress in adults were investigated. In addition, the mechanism of working memory of REM was explored. A randomized controlled trial was conducted for 90 healthy adult males who were randomly assigned to the control, REM, and relaxation groups with or without the intervention or control computer program after acute stress induction. The Positive and The negative effect Schedule, Reaction Time data, and Visual Analogue Scales were collected in the experiment. The data analysis results revealed that reductions in the negative effect were more significant in both the REM (P=0.030) and relaxation group (P=0.036) than in the control group with intervention. Users' reaction time was longer both in the REM ($P\lt0.001)$ and relaxation group ($P\lt0.001)$ than in the control group. Visual scale analysis results showed that users' clarity (P=0.028), emotionality (P=0.013), pain (P=0.002), and stress (P=0.007) of the recall all decreased more significantly in the REM group after the intervention compared to the control group. These results demonstrated that both computer programs of REM and relaxation attenuated the negative effected and reduced acute stress in adults effectively. There were differences during the process as the REM program reduced the clarity and stress scores of the recall. Such results of the relaxation group were not significant, suggesting the REM program alleviated the acute stress response by increasing users' working memory load.