ObjectiveTo investigate the psychometric features of MATRICS Consensus Cognitive Battery （MCCB） in adolescents with bipolar disorder， so as to evaluate its appropriateness for the measurement of cognitive deficits in adolescents with bipolar disorder. MethodsAdolescents with bipolar disorder （n=38）， adolescents with major depressive episode （n=40） and healthy controls （n=41） matched on age， sex and educational background were enrolled. Adolescents with bipolar disorder were assessed using Montreal Cognitive Assessment Scale （MoCA） and MCCB at baseline and 2 weeks later， while the rest were only assessed using MCCB at baseline. Thereafter， the psychometric features of MCCB such as internal consistency， test-retest reliability and criterion-related validity， discriminant validity and structural validity were evaluated using Cronbach's α coefficient， Pearson correlation analysis， analysis of covariance and confirmatory factor analysis. Results①The Cronbach's α coefficient of MCCB in adolescents with bipolar disorder was 0.784 at baseline and 0.773 at two weeks later， respectively. ②Among adolescents with bipolar disorder， the test-retest reliability over a two-week interval of each dimension in MCCB ranged from 0.630 to 0.812 （P<0.01）. ③ The criterion-related validity denoted that the score of short-term memory domain in MoCA was positively correlated with the speed of processing， verbal learning and working memory in MCCB （r=0.487， 0.522， P<0.05 or 0.01）. ④ Discriminant validity analysis implied that the scores of the processing speed， attention/vigilance， working memory， verbal learning and memory， visual learning and memory， reasoning and problem solving in MCCB yielded statistical differences among adolescents with bipolar disorder， adolescents with major depressive episode and healthy controls （F=3.790~7.243， P<0.01）. ⑤ Exploratory factor analysis showed that cumulative total variance contribution rate of MCCB amounted to 71.65% of four factors， and the confirmatory factor analysis indicated that the ideal 7-factor model had poor structural validity. ConclusionMCCB has good internal consistency， retest reliability and acceptable validity in adolescents with bipolar disorder.