BACKGROUND: Air pollution is a risk factor for poor cognitive function, while a plant-based dietary pattern is associated with better cognitive function. We aimed to explore their interaction with cognitive function among older adults. METHODS: We used a prospective cohort of old individuals, including 6525 participants of the Chinese Longitudinal Healthy Longevity Survey (CLHLS), aged 65-110 years and with normal cognition at baseline. Air pollution measurement was derived using satellite-derived annual average fine particulate matter (PM2.5) concentrations based on residential locations. Plant-based diet index (PDI) was calculated using survey responses to assess the dietary pattern. Repeated measures of the Mini-Mental State Examination (MMSE) were utilized to assess cognitive function. We applied the Cox proportional hazard regression to explore the associations and further stratified the analysis by PDI. FINDINGS: During a median of 5·6-year follow-up, 1537 (23·6%) out of 6525 participants with normal cognition at baseline developed poor cognitive function (MMSE <18). Living in areas with the highest quintile of cumulative PM2.5 was associated with a 46% increase in the risk of developing poor cognitive function (hazard ratio (HR): 1·46, 95% confidence interval (CI): 1·20, 1·77), compared to those living in areas with the lowest quintile. We observed a significant interaction between cumulative PM2.5 and PDI (p-interaction: 0·04), with the corresponding associations of cumulative PM2.5 being more pronounced among participants with lower PDI (HR: 1·68, 95% CI: 1·26, 2·24) than those with higher PDI (HR: 1·28, 95% CI: 0·98, 1·68). INTERPRETATION: Plant-based dietary pattern may attenuate detrimental impacts of PM2.5 on cognitive function among older adults. Adherence to the plant-based dietary pattern could be used to prevent adverse neurological effects caused by air pollution, especially in developing regions.