Accurate modelling of the phenomena occurring in loaded soil materials is the primary source of development of the current normative regulations as well as smarter and more efficient engineering practice. The aim of this research is to develop a more accurate description of the complex nature of the soil material and thus a more accurate evaluation of the subsoil-construction interaction in the whole range of stress, not only in chosen limit states. The proper description of the behaviour of soil in the range of small strain is an extremely important element to forecast displacements of construction interacting with subsoil. Due to this fact, it has a huge influence on the quality of mapping the actual internal forces in the whole structure - including the foundation. Stiffness moduli for small strains are actually recognized as one of the most important properties of the soil. RC/TS is the reference apparatus in which they are determined. A technique based on the analytical solutions was used to calculate the dynamic shear modulus, dynamic modulus of elasticity and the damping coefficient of soils and rocks. In theory, the movement of the tested soil particles resulting from the propagation of elastic waves is non-destructive. Despite the apparent differences in the behaviour of soil material under dynamic loads with a significant frequency (approx. 60- 160Hz - RC test) and quasi-static loads, characterized by very low frequency amplitude (below 10Hz, in practice 0.1Hz below - TS test), small strain conditions are modelled numerically using a combination of linear and non-linear laws of elasticity. The main drawback of standard RC/TS research methodology is the assumption that soil material is subject only to viscoelastic strain. This assumption is manifested, among others in the interpretation of the damping coefficient during TS test, which is defined as the proportion of the energy dissipated by the material during the cyclic torsional shearing to the potential energy accumulated in the material during elastic deformations. However, observations confirm the hypothesis that the behaviour of the soil, even in the range of small strains, should be considered as elasto-plastic - including the generation of a plastic, irreversible deformation. The research method involves executing multiple (repeated) torsional shear tests on natural and artificial (repeatable) structure samples of soil. Samples were subjected to varying conditions of the amplitude and frequency of cyclic load. Using modified procedure of registration test results allows the observation of changes in viscoelastic strain of material and measurement of permanent deformation after the completion of the torque impact. Multiple repetitions of tests on samples of the natural and artificial structure allows to identify the cause of any differences that may arise in the individual reactions of the material (phase elastic, viscous, plastic).