The manifestation of different psychophysical visual effects is intensively studied in modern graphic communication. One of the effects which occur most often in graphic design is the Munker-White effect. It is caused by the painted geometric structure called the Munker-White grid. This effect is manifested as a shift of manifestation of colour perception ( ) of rectangular grid elements in relation to the physical colour value. The shift of brightness manifestation is thereby manifested the most prominently ( ). This thesis defines new variations of the achromatic Munker-White effect. The new variations of grids with different percentage of coverage and with different brightness were used as test samples. A psychophysical visual experiment with 38 participants was then undertaken and the colorimetric differences of perceived colours in relation to their physical quantities were calculated in CIE system. The regression analysis showed the regularities of the Munker-White effect. The results are obtained by comparing different regression models until the models the best adjusted to experiment results were obtained. In this way the squared function dependence of the variable in dependence with the parameter of coverage percentage was determined. Furthermore, the squared function connection of the variable and the brightness of rectangular elements , as well as the paraboloid function, which gives the dependence of the variable on variables and , were also determined. This function can be considered the visual law of perception of rectangular elements of the Munker-White grid. A logistical regression analysis was also implemented. This was done in the case of differences which the observers notice, see, see well and obviously differentiate. The obtained logistical regression models enable the calculation of chances that the observer will perceive (notice, see, see well and obviously differentiate) the Munker-White effect in graphic reproductions in the mentioned cases. All obtained models are statistically tested and are proved to be of the highest statistical quality. The determined mathematical models enable the efficient study of the intensity as well as of the chances that the Munker-White effect would occur in different cases and as such represent a significant contribution to modern graphic communication.