Observed daughter-parent amplitude ratios $A_d^{\mathrm{\prime }}/A_p^{\mathrm{\prime }}$ of candidate resonances of SPB stars detected in models of light curves measured by Kepler, compared to their theoretically predicted equivalent observables, the monochromatic stationary daughter-parent surface luminosity perturbation ratios |𝔏_d| / |𝔏_p|, which are computed with the theoretical oscillation model described in Sect. 2. These quantities are shown as a function of the minimal inertial-frame frequency ν_min of the three signals in the considered candidate resonance, or, the minimal mode frequency observable ${\mathrm{\Omega }}_{\min ,\mathfrak{i}}^{\mathrm{NL},s}/2\pi$ among the three mode frequency observables of the considered triad, respectively. Left panel: $A_d^{\mathrm{\prime }}/A_p^{\mathrm{\prime }}$ as a function of ν_min. The light curve models were generated using strategy 3 of V21, which uses the signal-to-noise ratio of the detected signals to determine their significance. Colors indicate the pseudo-class defined in V21: high-f_sv (‘hfsv’), high-mode-density (‘hmd’), or outbursting (‘outb’). Right panel: |𝔏_d| / |𝔏_p| as a function of ${\mathrm{\Omega }}_{\min ,\mathfrak{i}}^{\mathrm{NL},s}/2\pi$. Colors indicate whether all stability criteria are fulfilled (‘Stab.’), or if additionally the AE validity criterion (‘Stab. + AE’) or all validity criteria (‘Stab. + Val.’; i.e., the isolated triads listed in Table 6) are fulfilled. We show alias signals in the frequency range of 0 to 24.4512 d⁻¹ as fainter orange-red (‘Stab.’) symbols, because we do not account for the splitting of the alias signals (which will affect the amplitude ratios).