Metamphetamine (METH) is a psychostimulant that is widely abused around the world. There is an increasing prevalence of addiction that can potentially be associated with offspring whose parents were addicted, making the offspring a risk group. METH in the body affects the morphological and functional changes of neural networks – neural plasticity. Our model for studying neural plasticity is METH self-administration, which we use to implement artificial selection to investigate the neurobiological basis of the of the difference in high (HP) and low (LP) preference. The aim of this experimental work was to determine whether preferential METH self-administration will show changes in the redox parameters of F1 filial generation of the 22nd and 28th generations of HP and LP strains, and whether these redox parameters will be stable in the F2 and F3 filial generation of the 22nd and 28th generations of HP and LP strains after the end of selection. The obtained results show that by measuring the concentration of hydrogen peroxide, advanced glycation product and catalase specific enzyme activity, the levels were lower in the whole body homogenates of the F1 filial generation of the selected strains of the 22nd and the 28th generations of the HP and LP strains compared to the control group. By measuring the superoxide dismutase (SOD) specific enzyme activity, no differences were observed between the F1 filial generation of the experimental groups, nor was there a difference compared to the control group. In the F2 and F3 filial generations of the 22nd generation HP and LP strains, there is no difference in the measured redox parameters, except for SOD specific enzyme activity, while in the F2 and F3 filial generations of the 28th generation HP and LP strains, all parameters change except SOD specific enzyme activity. In this experimental work, it was shown that METH consumption of the parental generation affects the change of measured redox parameters in the filial generations. Instability in the 28th generation HP and LP strains indicates potential epigenetic changes. Since they were not observed in the 22nd generation, the assumption is that the preference of the 22nd generation is due to the enrichment of gene variants that resulted in the HP and LP phenotype. In order to investigate this assumption, it is necessary to conduct future research.