Soil aggregate breakdown is the initial stage of splash erosion. Preventing aggregate disintegration can effectively control the material source for sediment transport. The impact factors of aggregate disintegration have been partially studied, but the effect of soil moisture content on slaking and physico-chemical dispersion remains unclear. In this study, simulation tests of splash erosion were carried out on three sandy clay loam soils with different years of vegetation restoration (S1, S2 and S3 were 55, 80 and 100 years of reforestation and land restoration, respectively). Three rainfall heights (1.5 m, 2 m and 2.5 m) were used to simulate aggregate disintegration caused by different rainfall kinetic energies. Soil aggregates with five antecedent moisture contents (air drying, 5 %, 10 %, 15 %, and 20 %) were subjected to three different pretreatments: fast wetting (FW), slow wetting (SW) and shaking after prewetting (WS). It was found that the mean weight diameter in the fast wetting (MWDFW) treatment increased and then decreased with increasing moisture content. Aggregate sensitivity to slaking decreased with increasing moisture content. After rainfall, the macroaggregates content (>0.25 mm) increased and then decreased with increasing moisture content. In the S1 sample, the contribution of physico-chemical dispersion to splash erosion was dominant for increasing splash erosion rate at the moisture content of 15–20 %. The slaking contributed more to decreasing splash erosion rate with increasing moisture content in the S2 and S3 samples. This study clarifies the response of slaking and physico-chemical dispersion to moisture content, which is helpful for thoroughly understanding the disintegration mechanisms of soil aggregates.