Several psychiatric disorders such as depression and schizophrenia and many others can not be well studied experimentally due to hard to have a perfect animal models that reflects changes similar to human.
Modeling human disorders characterized by changes in emotional and cognitive processing will always be difficult to do in animals. However, scientists can use animal models to gain insights into the neural mechanisms contributing to certain aspects of the disease and see if they parallel what is known in humans. For example, if you are interested in studying depression, you can study prominent aspects of the disease such as reduced sociability, sleep disturbances, anhedonia and behavioral despair in rodent by measuring social behavior, examining changes in circadian rhythm, sensitivity to rewards (sucrose consumption), and immobility duration in the Forced Swim Test.
Also, there are multiple hypotheses regarding the origin of many psychiatric disorders, which makes it difficult to find animal models that encompass all of them.
Thank you Milllie, I agree with you that it is impossible to have a model that reflects real situation for human as a disease but not healthy animal forced to be e.g, depressed, since there are many other contributing factors as style-life, stress, social interaction, age related diseases and many others. What's important is that are we really measure in behavior what we want in animal models ...... is real and reflecting the studied disorder as in depression, sleep disorder, food changes low motor activity and other symptoms of MDD of depression models.
I believe that canine anxiety-related behaviour problems represent a real-world animal model of human mental illness. They live in a similar environment to humans, they suffer from a wide variety of phobias, anxieties and fears. They have both a heritable and an environmental aetiology. These anxiety-disorders are also successfully managed in many cases with a combination of cognitive behavioural therapy and anxiolytic medication (such as benzodiazepines and SSRIs). One of the secondary aims of my research is to help build a bridge between the clinical behaviour problems we see in these dogs and the scientific rigor of behavioural paradigm work used in rodents.
In the pain field, there is increasing interest in using measures of emotional disturbance (anxiety, depression etc, to use the clinical terms) as a proxy for the spontaneous pain - these usually take the form of complex behaviours which take into account ethological considerations.
e.g.
Open Field - measurement of thigmotaxis in rats can be used to detect alterations in the animals innate assessment of a novel (and potentially threatening) area - pain decreases exploration of exposed areas (increasing thigmotaxis, or movement along walls), and treatment with analgesic agents can restore this.
http://www.ncbi.nlm.nih.gov/pubmed/18926876
http://www.ncbi.nlm.nih.gov/pubmed/24651240
Burrowing - this relatively new paradigm has been shown very sensitive in detecting behavioural alterations associated with decreased "Quality of life"