[Anxiety models on mice: experimental evaluation and criticism of the method]

The analysis of publications in the field of experimental psychopharmacology (in particular, anxiety models) shows a lack of data concerning the general methodological problems such as reproducibility and reliability of data. There are only a few works where important characteristics concerning solution pH, sham injections, body temperature, locomotion indices (basal and injection-modulated), etc, are reported. Figures on the illumination level in experimental chambers are generally missing. Three simplest, fastest, and most commonly used anxiety models (dark-light compartments, social isolation test, elevated plus-maze) are considered with special attention to the mistakes and artifacts most frequently encountered in registration of the experimental animal behavior and in the data interpretation.