There are many unknowns in the formation of subdwarf B stars. Different formation channels are considered to be possible and to lead to a variety of helium-burning subdwarfs. All seismic models to date, however, assume that a subdwarf B star is a post-helium-flash-core surrounded by a thin inert layer of hydrogen. We examine an alternative formation channel, in which the subdwarf B star originates from a massive (>~2 Msun) red giant with a non-degenerate helium-core. Although these subdwarfs may evolve through the same region of the log g-Teff diagram as the canonical post-flash subdwarfs, their interior structure is rather different. We examine how this difference affects their pulsation modes and whether it can be observed. Using detailed stellar evolution calculations we construct subdwarf B models from both formation channels. The iron accumulation in the driving region due to diffusion, which causes the excitation of the modes, is approximated by a Gaussian function. The pulsation modes and frequencies are calculated with a non-adiabatic pulsation code. A detailed comparison of two subdwarf B models from different channels, but with the same log g and Teff, shows that their mode excitation is different. The excited frequencies are lower for the post-flash than for the post-non-degenerate subdwarf B star. This is mainly due to the differing chemical composition of the stellar envelope. A more general comparison between two grids of models shows that the excited frequencies of most post-non-degenerate subdwarfs cannot be well-matched with the frequencies of post-flash subdwarfs. In the rare event that an acceptable seismic match is found, additional information, such as mode identification and log g and Teff determinations, allows us to distinguish between the two formation channels.