Objective To investigate the cognitive functions in first-degree relatives of patients with schizophrenia by meta-analysis. Methods Collect case-control studies of cognitive function in first-degree relatives of patients with schizophrenia,then extract data and assess the risk of bias of included studies,and use RevMan 5.3 to perform meta-analysis. Results A total of 27 studies involving 1708 people were includ-ed. Trail Marking A test(MD=3.59,P<0.01) and B test (MD=12.71,P=0.02),Stroop Color test((MD=-7.22,P<0.01),Stroop Color-Word test(MD=-7.25,P<0.01),Digit Symbol Coding test(MD=-7.29,P<0.01),Digit Span test(MD=-1.17,P<0.01),Hopkins Verbal Leaning test-Revised(MD=-4.58,P<0.01), Verbal Fluency test(MD=-2.92,P<0.01),Hanoi Tower(MD=-5.38,P<0.01),the number of corrected of WCST(MD=-3.10,P<0.01),number of completed categories of WCST (MD=-0.74,P<0.01),Persevera-tive Errors of WCST (MD=4.82,P<0.01) and Nonperseverative Errors of WCST (MD=8.00,P<0.01) scores between the first-degree relatives groups(FDR)and health control groups (HC)showed significant differences. Conclusion Cognitive impairment appears in the first-degree relatives populations of patients with schizophrenia,including the attention,verbal learning,working memory,and executive function.