[1] In this paper we investigate the role of different solar wind magnetosphere coupling functions on the Dst index calculated by the low-order nonlinear dynamical WINDMI model. In our previous work we have shown that the geotail current dynamics has a significant role in the two-phase decay of the Dst index. During that investigation we used the rectified solar wind electric field v x B z as a baseline for the simulations and analysis. Here we include an evaluation of four other coupling functions in addition to the rectified vB s . These coupling functions emphasize different physical mechanisms to explain the energy transfer into the magnetosphere due to solar wind velocity, dynamic pressure, magnetic field, and Mach number. One coupling function is due to Siscoe, another by Borovsky, and two by Newell. Our results indicate that for a majority of cases, at most only v x , B y , and B z are needed to sufficiently account for the supply of energy to the ring current and geotail current components that contribute to the Dst index. The more complex coupling functions sometimes perform extremely well on storm data sets but at other times do not reproduce the Dst index faithfully. The AL index was used as an additional constraint on the allowable geotail current dynamics and to further differentiate between coupling functions when the Dst performance was similar. The solar wind dynamic pressure contribution appears to be correctly accounted for through the calculation of the Dmp formula of Burton et al. (1975). The degree to which the B y component affects the Dst index is not entirely clear from our results, but in most cases its inclusion slightly overemphasizes the ring current contribution and slightly underemphasizes the geotail current contribution.