Previous?researches investigated children’s?cognitive flexibility?of varying age groups by using different?paradigms. Wisconsin?Card?Sorting?Test (WCST) was one of the most widely used tests of executive function in clinical and research contexts. These researchers drew?different?or even?contrary?conclusions. Predecessors' researches?pay a lot of attention to cognitive flexibility which refers to a scene when previous rule is shown to be ineffective then flexibly switch to a new rule according to feedback. To explore a new?unified method in studying cognitive flexibility in rule acquisition for wider age range (5-11 years) children, in the present work,?a?novel test named Target Choose Task (TCT）was designed based on WCST. TCT concerns individual cognitive flexibility, which specifically refers to whether one could resist or switch initial hypotheses flexibly according to different feedback in rule acquisition. Another aim is to carry out a preliminary discussion of development and fast period of children’s cognitive flexibility. Additionally, to study the reliability, validity, difficulty and discrimination of TCT. TCT consisted of 63 three-dimensional stimuli boxes. Each of them comprised three perceptual dimensions, shape (square, trapezoid, triangle or circle), color (red, green, blue or yellow), and pattern (star, gridding, streak). TCT was composed of 9 trials, in each trial of which participants were provided with 1 target stimulus box and 6 reference stimuli boxes, and were asked to match one of the 6 reference stimuli boxes to the target stimulus box based on a classification rule. Participants received feedback after each match. The cognitive flexibility (set shifting) was investigated with TCT in 181 participants (from kindergarten to fifth grades; mean age 8.4 years ; age rage: 5-11 years old; 86 males) selected randomly from an ordinary kindergarten and an ordinary primary school. Twenty-four participants (grade: from second to third grade; mean age 8.9 years; age rage: 7-10 years; 10 males) also participated in the computer version of Wisconsin Card Sorting Test (WCST). The participants completed the test individually. The results were following, (1) All TCT scores show a positively skewed distribution; (2) There is a positive correlation between TCT accuracy and WCST accuracy; (3) The TCT score respectively has significant correlations with Categories Completed (CC) and Conceptual Level Responses（CLR）of WCST; (4) While, no statistically significant difference is found between “B type error” of TCT and Percent Perseverative Errors（PPE）of WCST; (5) The split-half reliability coefficient was calculated. One half text items significantly correlate with the other half; (6) High-scoring group have significantly higher level comparing to low-scoring group in TCT score; (7) TCT scores of all subjects rage from 0 to 8 (mean score ± SD: 3.78 ± 1.9). (8) The pass rate of each TCT test for all participants is from 0.32 to 0.66, implying that the difficulty index of TCT is moderate. In summary, psychometric properties analysis showed that TCT achieved to an excellent level of internal consistency, criterion-related validity, distinction index and moderate level of difficulty. It is a reasonable new method of exploring cognitive flexibility.