River channel profiles and river terraces are two typical landscape features to investigate river incision history. However, incision history inferred from the two markers can be inconsistent. For example, along the Jinshan Gorge in the middle Yellow River, pseudo-terraces (a virtual metric derived from channel profile modelling) suggest a recent acceleration in incision rates only in the southern gorge, and the northern keeps steady, while river terraces in the trunk channel indicate recently increased incision rates all along the gorge. To mitigate these discrepancies, we investigate six tributaries along the Jinshan Gorge by mapping and dating several strath terraces at various distances from the confluence to the trunk channel. By comparing tributary terraces of similar age, we find generally increasing incision rates towards the confluence with the trunk river in the southern Jinshan Gorge. North to south increasing incision rates along the gorge are also observed from coeval tributary terraces. In addition, two different landscapes between the southern and northern tributaries near the confluence are recognized, one with a deeply incised gorge and another one with floodplains, suggesting incision rate in the northern gorge is steady through time. Integrating the new infrared stimulated luminescence (IRSL) and Accelerator Mass Spectrometry (AMS) 14C data from tributaries with the abundant literature data of terrace samples mostly in the Jinshan Gorge dated with various methods, we suggest published ages from optically stimulated luminescence (OSL) might be too young. The methodological limitation associated to the dating method can explain the observed mismatch of incision history in the northern gorge inferred from pseudo-terraces and from river terraces in the trunk channel, and reconcile some contradicting patterns in published trunk terrace data.