Prior studies using global cognitive measures have shown that social connectedness is linked with cognitive performance. We investigate the role of different social network dimensions on performance across distinct cognitive domains among dementia-free middle-aged and older rural Indians. We utilized baseline assessment data of 2,525 participants (≥45 years) of an ongoing, prospective, aging cohort in rural India (Srinivaspura Aging, Neuro Senescence, and COGnition) for this study. We evaluated social networks using Cohen's Social Network Index on three dimensions: network diversity, network size, and network embeddedness. We measured cognitive performance on memory (narrative recall), visuospatial ability (geometric figure spatial recognition, visuospatial span), language (verbal fluency, semantic association, word comprehension, reading comprehension), and attention domains of a culturally adapted, computerized, neurocognitive test battery. Linear regression models, adjusted for age and sex, were used for statistical analysis. In the linear regression model adjusted for age and sex, we found that greater network diversity was significantly associated with better performance in narrative recall, geometric figure spatial recognition, reading comprehension, semantic association, and attention tests. Participants with a larger network size had significantly better performance in verbal fluency and semantic association tests. Further, participants with greater network embeddedness had better visuospatial span and verbal fluency. Individuals having a greater network diversity, larger network size, and greater network embeddedness had better cognitive performance in multiple distinct cognitive domains. Following up these participants with serial cognitive monitoring can help understand if social networks play a role in delaying cognitive decline and protecting against dementia. (PsycInfo Database Record (c) 2024 APA, all rights reserved).