Lévy statistics define anxiety and depression in mice subjected to chronic stress

Current rodent models for anxiety and depression assessment face methodological challenges compromising both scientific rigor and animal welfare.

This study introduced a novel approach using the Lévy flight (LF) statistical method to analyze spontaneous movement in open spaces. We employed three models: Chronic unpredictable mild stress (CUMS), Electric shock stress (ES), and Chronic Restraint Stress (CRS)-utilizing a total of 540 mice for LF fitting. A support vector machine algorithm was applied to distinguish each model group based on the two-dimensional distribution of the variables γ and μ in the LF. Statistical analysis was performed using a two-dimensional Kolmogorov-Smirnov test before and after drug administration.

We found that the ES model primarily exhibited anxiety-like behaviors, the CRS model predominantly exhibited depression-like behaviors, and the CUMS model displayed both depression-like and anxiety-like behaviors. All three stress models were suitable for LF fitting, with the distribution of CUMS in the γ-μ plane lying between the ES and CRS groups. To assess the therapeutic effect of fluoxetine (FXT) on CUMS, we excluded the CUMS-resistant mice and performed LF analysis. Following FXT treatment, the mice gradually shifted toward the normal area in the γ-μ plane, with a more pronounced shift toward the depression area observed as the modeling time increased.

This study identifies a previously unrecognized statistical locomotor pattern in mice with anxiety and depression. By integrating scientific rigor with ethical considerations, this approach also presents a humane paradigm shift in preclinical assessment, accelerating translational breakthroughs in neuroscience research.