The microscopic short time dynamics of glass-forming materials seen by quasielastic neutron scattering, dynamic light scattering, high frequency dielectric relaxation and molecular dynamics simulation in the picosecond region show the existence of the primitive relaxation, exp(-t/tau(0)), for t < t(c), where t(c) approximate to 2 ps is the cross-over time of the coupling model. However, such an interpretation is not unique because the same data have also been considered to be evidence of the fast beta-relaxation of the model coupling theory. An apparently unique test of the coupling model is via the macroscopic long time relaxation data of ortho-terphenyl confined in nanometre glass pores with a size that is smaller than the length-scale of cooperative motions of the molecules. Under this condition, motions of the majority of the molecules cease to be cooperative and the experimentally observed relaxation time is the primitive relaxation time, so, of the coupling model. On the other hand, tau(0) can also be obtained from relaxation data of bulk Qrtho-terphenyl fitted to a stretched exponential correlation function, exp[-(t/tau)(1-n)], by the relation tau = [t(c)(-n)tau(0)](1/(1-n)) of the coupling model. Good agreement found between these two independently determined values of so provides strong support of the coupling model. A similar test of the coupling model is to compare the Johari-Goldstein beta-relaxation time, t(beta), with the calculated tau(0) at temperatures above the glass transition temperature. Both the beta-relaxation and the primitive relaxation time being noncooperative in nature implies that tau(beta) is comparable in order of magnitude to that of the calculated to, which is shown to be true for the fragile glass-former ortho-terphenyl as well as a non-fragile glass-former, cyclo-octanol.