There are several variations of the CAR test. The most common form of the test is the two-way active avoidance test (also known as the two-way discriminated shuttle box procedure). The animal does this by performing a specific behavioral response, like moving to the other compartment of the box, and this response is referred to as "avoidance" or "conditioned avoidance". If the animal is late in performing the avoidance, the aversive stimulus is presented until the animal responds by moving to the compartment. This is referred to as "escape". If the animal does not escape within a certain amount of time, it is designated "escape failure". As such, there are three variables that can be measured in the CAR test: avoidance, escape, and escape failure. Drugs that are considered to show antipsychotic-like effects
selectively suppress the
avoidance response without affecting
escape behavior. Conversely, drugs that are not considered to have antipsychotic-like effects either have no effect in the CAR test or suppress
both avoidance behavior and escape behavior at the same doses. Examples of drugs that inhibit both avoidance and escape responses include
sedatives like
benzodiazepines,
barbiturates, and
meprobamate and
antidepressants like many
tricyclic antidepressants (TCAs). The CAR test is considered to have high
predictive validity in the identification of potential antipsychotics and is frequently used in
drug development. However, its
face validity and
construct validity have been described as low or absent. Moreover, a described major limitation of the model is that drugs active in the test work by impairing a normal self-preservation function; that is, avoiding an unpleasant or
painful stimulus. Another limitation of the CAR test is that selective suppression of avoidance responses by drugs is procedure-specific. In procedures besides the one-way discriminated pole jump procedure and the two-way active avoidance test, such as the Sidman procedure, antipsychotics block avoidance behavior and escapes at almost the same doses. Conversely, benzodiazepines selectively suppress avoidance behavior without affecting escape behavior in the Sidman procedure. This is opposite to what is generally described as reflecting antipsychotic-like activity. Hence, selective suppression of avoidance responses is not a specific predictor of antipsychotic efficacy, or at best, selective suppression of avoidance responses as a predictor of antipsychotic activity is dependent on the specific CAR procedure employed. ==Drugs affecting the test==