This study investigates possible connections between the observed dayside equatorward wind in the upper thermosphere by the HIWIND balloon experiment and magnetospheric lobe-cell convection in the polar cap using numerical simulations from the first-principles, physics-based coupled magnetosphere-ionosphere-thermosphere (CMIT) model. Observations have shown that these lobe cells occur frequently in the summer hemisphere during IMF By dominant periods. Theoretical studies have shown that the lobe-cell phenomenon is a magnetospheric response to merging site asymmetries imposted by dipole tilt, IMF Bx and By. When the high-latitude forcing is from a global magnetosphere model driven by realistic upstream conditions, the thermosphere-ionosphere electrodynamic general circulation model (TIEGCM) is capable of reproducing the observed equatorward neutral wind in the upper thermosphere. When the high-latitude forcing is empirical models, the TIEGCM predictions of meridional neutral winds deviate significantly from observations. Data-model comparisons suggest that magnetospheric lobe-cell convection may have significant influence on thermospheric winds in the polar cap during geomagnetic quiet time through ion-neutral coupling.