Purpose — to apply a set of complementary statistical and information-theory methods for an objective, quantitative, and multiparametric assessment of learning processes in behavioral experiments involving animals. In particular, this work is aimed at assessing skill acquisition in mice during training in a spatial task in virtual reality and at identifying behavioral changes that cannot be explained solely by an increase in the number of received rewards.

Methods. Learning processes in mice performing a behavioral task were analyzed using conventional behavioral metrics and statistical tests. The following parameters were considered: the number of received rewards, total distance traveled, running time, time spent in the reward zone, and the number of entries into the reward zone. The dynamics of these parameters were described using the Wilcoxon, Friedman, and Page tests, as well as piecewise linear regression. The spatial organization of behavior was assessed using locomotor heatmaps, the Rayleigh test, Shannon information entropy, normalized entropy, Kullback–Leibler divergence, and a trajectory similarity coefficient. Spearman correlation analysis, simple linear regression, and multiple linear regression were used to identify relationships between learning-related metrics.

Results. The combination of conventional behavioral metrics, statistical tests, piecewise linear regression, and entropy-based trajectory characteristics allows formalizing the dynamics of behavioral task acquisition in virtual reality. It was shown that a plateau in the number of received rewards does not indicate complete stabilization of behavior. The proposed approach reveals differences between groups of animals under altered behavioral testing conditions that cannot be described solely by general locomotor activity or by changes in the number of received rewards.

Conclusion. Multiparametric analysis supports the interpretation of learning as a restructuring of behavioral organization, including reward acquisition performance, spatial organization, movement directionality, and coherence among behavioral metrics.