Establishing chronologies of archaeological sites by using a single dating method may not always reliably constrain the age of a deposit or a fossil, as potential biases may naturally arise, particularly in complex sedimentary settings such as caves. In this study, we performed a multi-technique dating approach that targets different materials in two caves from the Cueva Mayor-Cueva del Silo karst system, Sierra de Atapuerca, Spain. Two new fossil teeth from lithostratigraphic unit LU4 (GE-I pit) at Galería de las Estatuas have been dated by combined U-series/ESR and Amino Acid Racemisation (AAR) methods. The former provides ages of 117 ± 13 and 87 ± 9 ka that agree with previous dating studies of this Neandertal site, confirming the assignment of level LU4 to Marine Isotope Stage (MIS) 5. In contrast, AAR analyses produce very scattered ages, pointing to significant post-burial chemical modifications in dentine. A first set of ages is presented for the Cata Litario pit at Sala de los Cíclopes, a palaeontological site formed exclusively of cave bear fossils belonging to the Middle Pleistocene species Ursus deningeri. We provide U-series/ESR and AAR ages for two bear teeth from lithostratigraphic unit LU5, and extended-range luminescence ages for the sediment deposits that host, and immediately underlie and overlie, the bear accumulation. Depositional ages of the sediment were obtained using single-grain thermally-transferred optically stimulated luminescence (TT-OSL) dating of individual quartz grains and post-infrared-infrared stimulated luminescence (pIR-IR) dating of K-feldspars. Several speleothems, both in situ and reworked fragments embedded in the different lithostratigraphic units, have additionally been dated by U-series to constrain the timing of the sequence formation. The results indicate that the sediment originally entered the cave between MIS-12 and MIS-10 (444 ± 28 ka to 367 ± 32 ka). Nevertheless, the AAR and U-series/ESR fossil ages, as well as some of the speleothem fragment ages from LUs 5 and 6 are significantly younger. Leaving aside methodologically questionable results for one of the bear tooth samples, the most reliable AAR and U-Series/ESR ages range from 287 ± 23 ka to 256 ± 24, while the speleothem fragments indicate several formation episodes between 444 ± 19 and 284 ± 8 ka. The apparent difference between the U-Series/ESR ages on the fossils and the luminescence ages on the host sediment are not likely to be wholly explained by uncertainties on uranium uptake modelling or gamma dose rate evaluation, and thus is most likely related to taphonomic processes or a complex site formation history.The presence of fossils and eroded speleothem clasts of younger age within these levels is consistent with either an event of prior erosion, transport and resedimentation of LUs 4, 5 and 6, and/or a complex post-depositional mixing history for the stratigraphic sequence at Cata Litario. This study highlights the importance of using different dating methods to reconstruct reliable chronological frameworks, and to understand the geological factors that can affect the dated materials. This includes the accumulation and taphonomic histories of fossils, post-depositional mixing or reworking processes affecting sediments during the residence time within the karst system.