Quantitative structure activity relationship (QSAR) studies of twenty four 5-cyano, N1, 6- disubstituted, 2-thiouracil derivatives were performed for their central nervous system (CNS) depressant (locomotor) activity using VlifeMDS3.5 software. Partial least square (PLS) linear regression analysis coupled with stepwise variable selection method was applied to derive QSAR models which were further validated for statistical significance and predictive ability by internal and external validation. The best QSAR model was selected, having correlation coefficient r2 = 0.9014 and cross validated squared correlation coefficient q2 = 0.8120 with external predictive ability of pred_r2 = 0.6692. The QSAR model indicated that the vdWSurfaceArea (van der Waals surface area of the molecule), dipole moment, YcompDipole (y component of the dipole moment) and T_2_F_1 (count of number of double bounded atoms (i.e. any double bonded atom, T_2) separated from fluorine atom by 1 bond in a molecule) were the important determinants for CNS depressant (locomotor) activity.