Shallow saturated subsurface flow, frequently observed on hillslopes of headwater catchments in humid temperate climates, often dominates hydrologic responses of the catchments to major rainfall events. Typically, these responses are significantly affected by the presence of preferential flow. Reliable prediction of runoff from hillslope soils under such conditions remains a challenge. In this study, two approaches to modeling hillslope responses to rainstorms, which differ in dimensionality and thus also in the complexity of geometric, material, and boundary conditions, were tested and used on the hillslope discharge data observed in an experimental trench. In the one-dimensional (1D) approach, 1D variably saturated vertical soil water flow is combined with 1D lateral saturated flow above the soil–bedrock interface. In this approach, vertical flow is modeled by means of a dual-continuum concept involving two coupled Richards’ equations (representing flow in the soil matrix and in the preferential pathways), while lateral flow is described by the diffusion wave equation. In the two-dimensional (2D) approach, the movement of water in a variably saturated hillslope segment is modeled as vertical planar flow (i.e., the vertical and lateral flow components are fully integrated into one flow system). Similar to the 1D approach, the preferential flow effects are implemented in the 2D model by means of the dual-continuum concept. The two model approaches (1D and 2D) resulted in similar hillslope discharge hydrographs, characterized by short-term runoff peaks followed by zero-discharge periods, but the 2D model showed closer agreement between observed and simulated soil water pressure heads near the trench. The sensitivity analysis of soil and bedrock properties confirmed a significant influence of the bedrock saturated hydraulic conductivity on simulated hillslope discharge. The simpler 1D approach, based on the combination of 1D vertical flow and 1D lateral flow, was found to provide a useful approximation of the more complex and flexible 2D system and to be far more efficient in terms of computing time.