A better understanding of the causal relationship between spousal cognitive functioning and depression levels among middle-aged and older adults is vital for effective health policymaking under the globally severe aging challenge. However, the related evidence is often limited by potential omitted-variable bias and reverse causation. This study uses an instrumental variables approach, namely the two-stage least squares (2SLS) method, to examine the impact of spousal cognitive functioning on depression levels among middle-aged and older adults in China. The data were sourced from the China Health and Retirement Longitudinal Study (CHARLS) of 2020, including a total of 3,710 couples aged 45 years and above. Depression levels were measured using the Center for Epidemiologic Studies Depression Scale (CES-D-10), while cognitive functioning was assessed using the Mini-Mental State Examination (MMSE). Spousal social participation was employed as the instrumental variable to address omitted-variable bias and reverse causation. Additionally, an interaction effect test between gender and spousal cognitive functioning was conducted. The results show that for each one-point increase in the spouse's MMSE score, the CES-D-10 score of middle-aged and older adults decreased by 17.1% to 68.2%. The OLS results indicated that women, rural residents, and middle-aged individuals were more sensitive to these changes. The interaction effect test results confirmed that women were more affected by changes in spousal cognitive functioning. However, after a more reliable 2SLS analysis, the results for age groups shifted, showing that middle-aged individuals were more sensitive to these changes, with a decrease in depression levels reaching 70.0%, compared to 60.2% for the elderly group. Nonetheless, given the prevalence of depression among the elderly, the impact of spousal cognitive decline on depression in this group should not be overlooked. Our findings highlight the importance of spousal cognitive health in managing depression among both middle-aged and older adults, with particular attention to women and rural populations.