The use of antipsychotic drugs represents an important approach for the treatment of schizophrenia. However, their efficacy is limited to certain symptoms of this disorder, and they induce serious side effects. As a result, there is a strong demand for the development of new drugs, which depends on reliable animal models for pharmacological characterization. The present review discusses the face, construct, and predictive validity of classical animal models for studying the efficacy and side effects of compounds for the treatment of schizophrenia. These models are based on the properties of antipsychotics to impair the conditioned avoidance response and reverse certain behavioral changes induced by psychotomimetic drugs, such as stereotypies, hyperlocomotion, and deficit in prepulse inhibition of the startle response. Other tests, which are not specific to schizophrenia, may predict drug effects on negative and cognitive symptoms, such as deficits in social interaction and memory impairment. Regarding motor side effects, the catalepsy test predicts the liability of a drug to induce Parkinson-like syndrome, whereas vacuous chewing movements predict the liability to induce dyskinesia after chronic treatment. Despite certain limitations, these models may contribute to the development of more safe and efficacious antipsychotic drugs.

Anxiety and stress-related disorders are severe psychiatric conditions that affect performance in daily tasks and represent a high cost to public health. The initial observation of Charles Darwin that animals and human beings share similar characteristics in the expression of emotion raise the possibility of studying the mechanisms of psychiatric disorders in other mammals (mainly rodents). The development of animal models of anxiety and stress has helped to identify the pharmacological mechanisms and potential clinical effects of several drugs. Animal models of anxiety are based on conflict situations that can generate opposite motivational states induced by approach-avoidance situations. The present review revisited the main rodent models of anxiety and stress responses used worldwide. Here we defined as “ethological” the tests that assess unlearned/unpunished responses (such as the elevated plus maze, light-dark box, and open field), whereas models that involve learned/punished responses are referred to as “conditioned operant conflict tests” (such as the Vogel conflict test). We also discussed models that involve mainly classical conditioning tests (fear conditioning). Finally, we addressed the main protocols used to induce stress responses in rodents, including psychosocial (social defeat and neonatal isolation stress), physical (restraint stress), and chronic unpredictable stress.