The car-lights span task and the car-memory game, with story background, were used as behavioral and physiological measurement tools for children's visual-spatial working memory, respectively. 104 children (54 males, average age 71.11 months) living in Hsinchu City from 5 to 6 years old participated in the behavioral experiment of this study. Among them, 22 children also participated in the electrophysiological experiment. The results of the correlation analysis showed that the performance of 5- to 6-year-old children in the car-lights span task was significantly positively correlated with their performance in short-term memory, simple working memory, and complex working memory tasks, and was also significantly positively correlated with their performance of Toni non-verbal intelligence, WPPSI–IV working memory, and WPPSI–IV spatial. It showed that the car-light span task has construct validity. The performance of 5- to 6-year-old children in the car-lights span task was also significantly positively correlated with their K-scores and accuracy in the car-memory game, showing the behavior performance of children on complex span task and change detection task was consistent. In the car memory games, the CDA peak amplitude of children aged 5 to 6 showed the effect of memory load. It is observed that whether it is in the P7/P8 or P3/P4 brain regions when the number interval of the set size was greater than 2, the CDA peak amplitude increased significantly with the increase of the set size. These findings showed that the two tasks designed in this study were suitable as behavioral and electrophysiological measurement tools for the visual-spatial working memory of children aged 5 to 6 years old in Taiwan.