We are developing a mechanism-based fracture model for brittle solids that can help understand the effects of interactions on damage-evolution and softening under high-rate compressive loading. The microstructure of such solids is generally comprised of a polycrystalline matrix material interspersed with micro-scale heterogeneities such as pre-existing cracks, voids, second-phases, precipitates, and pores. The heterogeneities, which can be considered to be randomly distributed in space for a given material domain and across realizations of the material, are often initiators of micro-cracking. This is in turn related to the emergent non-linearity in the stress-strain response of such materials. The constitutive response under high-rates is particularly sensitive to the microstructural stochasticity as a range of heterogeneities can initiate cracks (in contrast to quasi-static loading where typically the ‘weakest’ heterogeneities initiate cracks). We also perform numerical experiments using the theoretical framework to understand the effective behavior of the material under the influence of these heterogeneous microstructures within dynamic loading problems.