We investigate the impact of erosion on the geometry and kinematics of the Aconcagua fold-and-thrust belt in central Argentina using analog and numerical models. In these models, the surfaces are eroded according to a rule in which mass removal is limited by the rate of fluvial bedrock incision. This approach unifies principles of frictional failure used in critical Coulomb wedge theory with a quasi-mechanistic erosion rule, which allows us to explicitly relate temporal changes in erosional efficiency in this fold-and-thrust belt to its kinematics. We show that theoretical predictions of the fold-and-thrust belt geometry, as well as the kinematics predicted by both physical and numerical experiments, are both internally consistent and correctly predict the interpreted and measured field geometries. Specifically, the geometric evolution of the Aconcagua fold-and-thrust belt requires relatively high erosion efficiency values (K) during the initial stage of deformation, and relatively low K values during the latter stages, consistent with the progressive exposure of different rock types during the different stages of deformation. Model results indicate that the activity of the faults in the hinterland is high when erosion is most efficient during the initial stage of deformation; this activity is facilitated by increased out-of-sequence thrusting. In contrast, the models predict that forward-propagating thrusts dominate the latter stages of deformation when erosion is far less efficient. Comparing the geometry of models subjected to erosion with those in which the surface remains uneroded suggests that erosional mass removal is required to explain the geometry of this fold-and-thrust belt. By implication, the results indicate that the kinematics of this, and perhaps other fold-and-thrust belts, may be intimately tied to the temporal history of erosion of the surface topography of these features.