Behavioral research using animal models may need improvement

Posted Oct.1, 2002

When scientists use animal models to study mental disorders such as schizophrenia and autism, they do their best to standardize the environment of the animals to eliminate confounding factors that could bias data. A blank slate, untainted by the experience of a complex environment, has been thought to produce the most valid results. New evidence, however, suggests that this just might not be the case.

According to Joseph Garner, PhD, a postdoctoral researcher at the University of California-Davis, researchers using rodent models to study mental disease may not be getting accurate results, if they are housing their animals in barren cages. Dr. Garner presented evidence for this theory, which is stirring up controversy, at the Fourth World Congress on Alternatives and Animal Use in the Life Sciences, held in August in New Orleans. Noncomplex environments can have profound effects on the brain and behavior and may lead to erroneous results, Dr. Garner said.

The scientist gathered data for this theory from studying stereotypies in animals including mice. Stereotypies, actions repeated over and over again for no reason, are common in captive animals that are stressed by their environment. For example, bar-mouthing, an animal rubbing the gap between its front and back teeth, is common among rodents. Stereotypies are prevalent in certain strains of mice and sometimes vary by sex.

Through recent experiments, Dr. Garner has discovered that these abnormal behaviors are correlated with dysfunction in certain areas of the brain. In rodents, for example, stereotypy correlates with impairments of basal ganglia function. Barbering or hair plucking, another abnormal behavior in captive mice, correlates with impairments in prefrontal cortex function.

"What we know at this point is that the worse these behaviors are in [the mice], the worse the dysfunction in that area of the brain appears to be," Dr. Garner said. "And if you enrich the environment, the stereotypy goes away in younger animals." Now, he says, efforts are being made to connect the dots and prove definitively that the environment causes the abnormal behaviors by inducing brain dysfunction.

Dr. Garner recommends enriching environments for rodents in two ways: providing enough material so that the animal can build a nest and reducing ultrasonic noise. Ultrasonic noise, such as that from computers, causes stress, and a lack of covering causes fear. "Without cover, they spend their entire lives exposed to perceived, potential predation, and you can imagine how that might lead to a chronically stressed animal," he said. "[This] might affect things like its physiological response to pretty much any drug you give it, its immune response, its breeding response, etc."

In addition to possibly altering the effects of pharmacology studies, noncomplex environments may also interfere with genetic studies. "A lot of the genetic work that is being done is being done to find aspects of brain development, the effects of different neurotransmitter- or neuroreceptor-regulating genes on the brain and behavior," Dr. Garner said. "If you are doing that in a strain of mice that is particularly susceptible to the effects of the environment, then it is possible that you could discover that one gene has one effect in one environment and a completely different effect in another."

Far from standardizing laboratory animals, barren environments may induce severe brain abnormalities, Dr. Garner said, and these abnormalities call the validity of a wide range of experiments into question.

"What concerns me, simply in terms of animal welfare, is that you want people to be doing the best possible experiment they can with the fewest animals," Dr. Garner said. "I am not saying that these experiments shouldn't be done. I am just saying that we should think about the environment as a possible problem in the current way experiments are performed." According to him, good animal welfare is good science.