Background: This study assesses the relationships between lymphocyte and monocyte subsets and intelligence quotient (IQ) scores in antiretroviral therapy (ART)-naive, HIV-infected Thai children without advanced HIV disease.Findings: Sixty-seven ART-naive Thai children with CD4 between 15-24% underwent cognitive testing by Weschler intelligence scale and had 13 cell subsets performed by flow cytometry including naive, memory and activated subsets of CD4+ and CD8+ T cells, activated and perivascular monocytes and B cells. Regression modelling with log10 cell count and cell percentage transformation was performed.Median age (IQR) was 9 (7-10) years, 33% were male, CDC stages N:A:B were 1:67:31%, median CD4% and count (IQR) were 21 (18-24)%, 597 (424-801) cells/mm3 and HIV RNA (IQR) was 4.6 (4.1-4.9) log10 copies/ml. Most (82%) lived at home, 45% had a biological parent as their primary caregiver, and 26 (49%) had low family income. The mean (SD) scores were 75 (13) for full scale IQ (FIQ), 73 (12) for verbal IQ (VIQ) and 80 (14) for performance IQ (PIQ). Adjusted multivariate regression analysis showed significant negative associations between B cell counts and FIQ, VIQ and PIQ (p < 0.01 for all); similar associations were found for B cell percentages (p < 0.05 for all).Conclusions: High B cell counts and percentages were strongly associated with poorer FIQ, VIQ and PIQ scores. Prospective, long-term assessment of cell subsets and determination of relevant B cell subpopulations could help further elucidate associations between lymphocyte subsets and neurocognitive development. © 2014 Ananworanich et al.; licensee BioMed Central Ltd.