This study investigates in detail the effects of predose on TL and OSL signals for BeO dosimeters. The TL and OSL signals were deconvoluted to each peak and component. As a result of deconvolution, the variations in the kinetic parameters (E, Tmax, b) for the TL signals and the lifetime values of the OSL signals were investigated. In addition, sensitivity changes according to predose were monitored for each peak and component. Finally, dose response curves were studied using the dose linearity index (f(D)) for each peak and component. Accordingly, the peak structures and kinetic parameters did not change according to the predose for the TL signal, whereas variations in the lifetime values for the OSL signal were observed, especially at the initial dose values (0.1 and 0.2 Gy). There was no change in sensitivity according to the predose for the total area condition although each peak and component exhibited independent behavior. Therefore, TL and OSL signals should be evaluated based on the total area in predose applications. The TL and OSL dose response curves exhibited different behaviors according to predose. TL dose response curves were not affected by the predose except for 1000 Gy, while the OSL dose response curves were affected by the predose considering the total area condition. The possible reason for the differences between the TL and OSL dose response curves is the significant transfer effect in the OSL signal at low doses, which results in greater changes at low doses compared to the TL signal. Also, thermal quenching effects may have resulted in lower intensity in the case of the TL signal. In future studies, preheating tests and thermal quenching corrections on TL peaks at high predoses may increase our understanding of deep trap interactions in BeO dosimeters.